phase ii trial of erlotinib and docetaxel in advanced and refractory hepatocellular and biliary cancers: hoosier oncology group gi ‐ - chiorean - - the oncologist - wiley online library 	skip to article content 	skip to article information [pdf] phase i/ii study of a bnc p/everolimus regimen for progressive metastatic renal cell carcinoma (mrcc) following prior tyrosine kinase inhibitors (hoosier oncology group). | semantic scholar skip to search formskip to main content> semantic scholar's logo search sign increate free account you are currently offline. some features of the site may not work correctly. 	doi: . /jco. . . _suppl. corpus id: phase i/ii study of a bnc p/everolimus regimen for progressive metastatic renal cell carcinoma (mrcc) following prior tyrosine kinase inhibitors (hoosier oncology group). @article{hutson phaseis, title={phase i/ii study of a bnc p/everolimus regimen for progressive metastatic renal cell carcinoma (mrcc) following prior tyrosine kinase inhibitors (hoosier oncology group).}, author={thomas e. hutson and l. dang and r. lauer and a. starodub and r. hauke and t. logan and k. bylow and m. galsky and david c. bibby and g. kremmidiotis and e. doolin and tina c. lavranos and annabell f. leske and n. hahn and g. sonpavde and c. sweeney and j. sarantopoulos}, journal={journal of clinical oncology : official journal of the american society of clinical oncology}, year={ }, volume={ _suppl}, pages={ } } 	thomas e. hutson, l. dang, + authors j. sarantopoulos 	published 	medicine 	journal of clinical oncology : official journal of the american society of clinical oncology background: bnc p is a vascular disruption agent (vda) that destabilizes tubulin polymers leading to selective damage of tumor vasculature, hypoxia and associated tumor necrosis. bnc p also has a direct anti-proliferative action on cancer cells. up regulation of the mtor pathway has been identified as a survival response by the tumor to hypoxic insult. preclinical investigations demonstrated that bnc p is effective at selectively damaging the vasculature in primary and metastatic… expand view on pubmed bionomics.com.au save to library create alert cite launch research feed share this paper citations view all topics from this paper 	bnc p 	everolimus 	metastatic renal cell cancer 	sunitinib 	neoplasms 	renal cell carcinoma 	tyrosine kinase inhibitors [moa] 	hypoxia 	epithelial cell of renal tubule 	thrombocytopenia 	adverse reaction to drug 	kidney neoplasm 	tumor necrosis 	anemia 	inflammatory disease of mucous membrane 	cancer cell 	dlst wt allele 	protein tyrosine kinase 	vascular diseases 	milligram per square meter citations citation type citation type 	all types 	cites results 	cites methods 	cites background has pdf publication type author more filters more filters filters sort by relevance sort by most influenced papers sort by citation count sort by recency combination therapy in metastatic renal cell cancer. 	a. ravaud, m. gross-goupil, j. bellmunt 	medicine 	seminars in oncology 	 	 save alert research feed overexpression of c-abl predicts unfavorable outcome in epithelial ovarian cancer. 	s. zhou, liangdan tang, + authors r. berkowitz 	medicine 	gynecologic oncology 	 	 save alert research feed references molecular cancer therapeutics: from bench to bedside 	d. v. hoff 	biology 	 	 save alert research feed related papers 	abstract 	topics 	 citations 	 references 	related papers stay connected with semantic scholar sign up about semantic scholar semantic scholar is a free, ai-powered research tool for scientific literature, based at the allen institute for ai. learn more → resources datasetssupp.aiapiopen corpus organization about usresearchpublishing partnersdata partners   faqcontact proudly built by ai with the help of our collaborators terms of service•privacy policy the allen institute for ai by clicking accept or continuing to use the site, you agree to the terms outlined in our privacy policy, terms of service, and dataset license accept & continue digital health/social media a cardiovascular health and wellness mobile health intervention among church-going african-americans: formative evaluation of the faith! app laprincess brewer , ashok kumbamu , christina smith , sarah jenkins , clarence jones , sharonne hayes , christi a. patten , lisa a. cooper , and lora burke mayo clinic; mayo clinic college of medicine; hue-man partnership; johns hopkins university school of medicine; university of pittsburgh objectives/goals: to evaluate the faith! (fostering african- american improvement in total health) app mhealth lifestyle intervention by using post-intervention feedback obtained from par- ticipants in our intervention pilot study. methods/study population: we used qualitative methods (focus groups) to elicit post-intervention feedback. participants who completed the pilot study were recruited to one of two focus groups. semi-structured focus groups were conducted to explore participants’ views on the app functionality, utility and satisfaction as well as its impact on healthy lifestyle change. sessions were audio-recorded, transcribed verbatim and qualitative data were analyzed by systematic text con- densation thematic analysis. results/anticipated results: nine individuals participated (n = and n = ) in each of the two focus groups. their mean age was . years (sd . ), % were women, and all had at least an education level of some college. six overarching themes emerged from the data: ( ) overall impression, ( ) content usefulness ( ) formatting, ( ) implementation, ( ) impact and ( ) suggestions for improvement. underpinning the themes was a high level of agreement that the intervention facilitated healthy behavioral change through cultural tailoring, multimedia education modules and social networking. among the suggestions for improve- ment were streamlining of app self-monitoring features, personaliza- tion based on individual’s cardiovascular risk and attentiveness to nuanced cultural perspectives. discussion/significance of impact: this formative evaluation found the faith! app mhealth lifestyle intervention had high reported satisfaction and impact on the health-promoting behaviors of african-americans, thereby improving their overall cardiovascular health. the findings provide further support for the acceptability of mhealth interven- tions among african-americans. conflict of interest description: none. all in for health: promoting good health and engaging a health research volunteer community in the hoosier state jessica hall , christine drury, and carmel egan indiana university school of medicine objectives/goals: • to improve and expand health and research literacy throughout indianabysharinghealth-focusedresourcesandresearchoutcomes. • to encourage and increase health research participation throughout indiana by promoting health research opportunities, including clinical studies. methods/study population: • discover and understand community concerns and barriers to good health and clinical research participation by providing a platform for individuals and communities to share their voices. • educate indiana residents on the importance of participating in health research. • engage with the community to meet them where they are (online) and continue to build relationships throughout the state. • promote healthy living for indiana residents by sharing health education and resources from existing state health organizations and initiatives. • develop and maintain the largest statewide database of research volunteers. results/anticipated results: • the anticipated results from this program include engagement of all populations and all communities throughout the state in con- versation and education around good health and health research, aswell as participation inhealth research acrossthe ctsi’s partner organizations. large-scale growth is expected in both the online communityandconsentedvolunteerregistryisexpectedtoinclude and engage racially and ethnically diverse populations, as well as special health populations, such as representatives of rural com- munities,aged,rarediseasesurvivors,andtransgenderindividuals. discussion/significance of impact: • thorough this work, the indiana ctsi has developed a unique program, educating the public about health research and oppor- tunities to participate, while simultaneously supporting research departments with marketing promotion of their efforts, and a ready statewide volunteer community. an app a day: examining clinical evidence for safety and efficacy of diabetes mobile health apps avantika pathak, ms , susan bain , and eunjoo pacifici usc school of pharmacy; usc department of regulatory and quality sciences objectives/goals: mobile health applications are widely used by the public but vary in how they are classified and regulated. this study examines the evidence of the safety and efficacy of mobile medical applications specifically focusing on those that are used to manage dia- betes. methods/study population: to understand the cur- rent regulatory landscape of mobile health applications (mhealth apps) for diabetes, a literature survey was conducted using the pubmed database. top mhealth apps were identified by searching the apple store website using key terms associated with diabetes management applications. a maximum of ten results, when available for each key term, were studied by exploring the fda databases to understand how the products were regulated and if any were subject to recalls. these selected mhealth apps were also searched on clinicaltrials.govtoseeifthere wereongoingorcompletedclinicaltrials and if the trials were designed to include efficacy and safety outcome measures. results/anticipated results: of the mhealth apps for diabetes management that were identified, were regulated. jcts abstract supplement https://www.cambridge.org/core/terms. https://doi.org/ . /cts. . downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core terms of use, available at http://clinicaltrials.gov https://www.cambridge.org/core/terms https://doi.org/ . /cts. . https://www.cambridge.org/core penetration depth measurement of near-infrared hyperspectral imaging light for milk powder sensors article penetration depth measurement of near-infrared hyperspectral imaging light for milk powder min huang , , moon s. kim ,*, kuanglin chao , jianwei qin , changyeun mo , carlos esquerre , , stephen delwiche and qibing zhu key laboratory of advanced process control for light industry, ministry of education, jiangnan university, wuxi , china; huangmzqb@ .com (m.h.); zhuqib@ .com (q.z.) environmental microbial and food safety laboratory, agricultural research service, usda, beltsville, md , usa; kevin.chao@ars.usda.gov (k.c.); jianwei.qin@ars.usda.gov (j.q.) national institute of agricultural science, rural development administration, nongsaengmyeong-ro, wansan-gu, jueonju-si, jeollabuk-do , korea; cymoh @korea.kr school of biosystems and food engineering, university college dublin, dublin , ireland; carlos.esquerre@ucd.ie food quality, agricultural research service, usda, beltsville, md , usa; stephen.delwiche@ars.usda.gov * correspondence: moon.kim@ars.usda.gov; tel.: + - - - ; fax: + - - - academic editor: gonzalo pajares martinsanz received: september ; accepted: march ; published: march abstract: the increasingly common application of the near-infrared (nir) hyperspectral imaging technique to the analysis of food powders has led to the need for optical characterization of samples. this study was aimed at exploring the feasibility of quantifying penetration depth of nir hyperspectral imaging light for milk powder. hyperspectral nir reflectance images were collected for eight different milk powder products that included five brands of non-fat milk powder and three brands of whole milk powder. for each milk powder, five different powder depths ranging from mm– mm were prepared on the top of a base layer of melamine, to test spectral-based detection of the melamine through the milk. a relationship was established between the nir reflectance spectra ( . – . nm) and the penetration depth was investigated by means of the partial least squares-discriminant analysis (pls-da) technique to classify pixels as being milk-only or a mixture of milk and melamine. with increasing milk depth, classification model accuracy was gradually decreased. the results from the -mm, -mm and -mm models showed that the average classification accuracy of the validation set for milk-melamine samples was reduced from . % down to . % as the milk depth increased from mm– mm. as the milk depth increased to mm and mm, model performance deteriorated further to accuracies as low as . % and . %, respectively. the results suggest that a -mm sample depth is recommended for the screening/evaluation of milk powders using an online nir hyperspectral imaging system similar to that used in this study. keywords: penetration depth; hyperspectral imaging; milk powder; pls-da . introduction milk, both a nutritious food in itself and a functional ingredient in other food products, is a complex fluid consisting of fats, proteins, minerals, vitamins, enzymes, carbohydrates and water. however, fluid milk is difficult to transport and store. therefore, milk powders are produced using drying technologies to turn fluid milk into dry milk powder. nonfat milk and whole milk are the two most common milk powders and contribute nutritionally to many food formulations, including reconstituted milk, dairy products, baked goods, confectionery, processed meat products, nutritional beverages and prepared ready-to-eat foods. as an important food ingredient for human sensors , , ; doi: . /s www.mdpi.com/journal/sensors http://www.mdpi.com/journal/sensors http://www.mdpi.com http://www.mdpi.com/journal/sensors sensors , , of and animal food, milk powder safety is a worldwide concern. in recent years, incidents of milk powder adulteration by melamine ( , , -triamino- , , -triazine) to boost apparent protein content caused illnesses and resulted in wide recognition of melamine contamination as a food safety problem. traditional methods of melamine detection in foods involve analytical techniques, such as mass spectrometry and high performance liquid chromatography, are time consuming, expensive and require complicated sample preparation procedures [ , ]. visible (vis) and near-infrared (nir) spectroscopy have been studied as non-destructive methods to detect melamine in milk by several research groups [ , ]; however, spectroscopic assessments with relatively small point-source measurements cannot provide information on the spatial distribution of melamine particles within a food sample. as one of the most promising tools for non-destructive real-time evaluation of food quality and safety in numerous applications, hyperspectral imaging combines features of both imaging and spectroscopy, such that it is capable of not only directly assessing the presence of different components simultaneously, but also locating the spatial distribution of those components in the products under examination [ ]. hyperspectral imaging may be operated in either reflectance or transmittance modes, although reflectance is more commonly used. generally, transmittance mode is used for thinly-prepared samples, allowing light to pass through the samples [ – ], while diffuse reflectance is used for thicker samples in hyperspectral imaging measurements of whole or larger portions of foods, such as apples [ ], peach [ ], mushrooms [ ], cucumbers [ ] and chickens [ ]. hyperspectral reflectance imaging has been used to detect defects and contaminants and to evaluate quality attributes for fruits, vegetables, meats and dairy products. implemented with automated image processing and analysis algorithms, hyperspectral imaging has been demonstrated for effective real-time assessment of the quality and safety attributes of poultry [ ]. for reflectance imaging, the nir light must sufficiently penetrate the food material in order for the intensity of the remitted radiation, as a function of wavelength (or frequency), to have been influenced by the chemical nature of the absorbing compound. light penetration depth is defined as the depth in a sample material at which incident light is reduced by % and will vary with the status of sample, the type of sample and the detection wavelength [ ]. hyperspectral reflectance imaging is usually operated in the vis-nir ( – nm) or nir ( – nm) range. limited research exists for the investigation of the penetration depth in the vis and nir ranges. lammertyn et al. [ ] reported light penetration depth in apples to be up to mm in the – -nm range and between and mm in the – -nm range. qin and lu [ ] found that the light penetration depth in fruit tissue varies depending on the type of fruit, ranging from . mm for plums to . mm for zucchini. most studies reporting light penetration depths were conducted on fruits. further research would provide not only helpful references for thickness determination, but also valuable insight into appropriate sensing configurations, especially for milk powder products [ , ]. fu et al. [ ] coupled the nir hyperspectral imaging technique ( – nm) with spectral similarity analyses to detect melamine mixed into samples of dry milk powder. imaging allowed visualization of the distribution of melamine particles in the milk mixture samples that were prepared at melamine concentrations ranging from . %– . % and presented for imaging in plastic petri dishes. however, it was not examined exactly how many millimeters the nir hyperspectral imaging light could penetrate into the milk powder samples. the objective of this study was to determine the penetration depth of nir hyperspectral imaging light in milk powder for the wavelengths between . nm and . nm. . experimental section . . sample preparation eight different milk powder products were purchased from commercial retailers, including brands of nonfat milk powders (‘valley (n)’, organic valley, la farge, wi, usa; ‘nestle (n)’, nestle, solon, oh, usa; ‘hoosier (n)’, hoosier hill farm, fort wayne, in, usa; ‘now (n)’, now real food, sensors , , of blooming, il, usa; ‘bob (n)’, bob’s red mill, pheasant court milwaukie, or, usa) and three brands of whole milk powders (‘hoosier (w)’, hoosier hill farm, fort wayne, in, usa; ‘nestle (w)’, nestle, glendale, ca, usa; ‘peak (w)’, peak, friesland campina, p. stuyvesantwet , ac leeuwarden, holland). melamine powder ( % purity) was obtained from sigma-aldrich company (m , st. louis, mo, usa). for detecting the light penetration depth of the hyperspectral imaging system, different thicknesses ( mm, mm, mm, mm and mm) of pure milk powders were layered on the top of pure melamine. these milk-melamine samples were prepared in black electroplated aluminum plates precisely machined with square wells ( -mm width and height, -mm depth). each sample was first prepared with a leveled layer of pure melamine in the bottom of the well, followed by a leveled layer of pure milk powder on top of the melamine. the combined thicknesses of the melamine and milk powder layers completely filled the -mm depth of each well, with the milk depth ranging between and mm (i.e., mm-thick milk layer over mm of melamine, mm-thick milk layer over mm of melamine, and so on). figure illustrates the preparation of a sample containing mm of milk over mm of melamine. for each kind of milk powder, three samples were prepared at each of the five milk depths, as well as three samples of pure dry milk ( mm of milk with no melamine underneath) and one sample of pure melamine. with plates prepared for each of the milk powder products, a total of samples were measured for the eight milk products used in this study. sensors , , of blooming, il, usa; ‘bob (n)’, bob’s red mill, pheasant court milwaukie, or, usa) and three brands of whole milk powders (‘hoosier (w)’, hoosier hill farm, fort wayne, in, usa; ‘nestle (w)’, nestle, glendale, ca, usa; ‘peak (w)’, peak, friesland campina, p. stuyvesantwet , ac leeuwarden, holland). melamine powder ( % purity) was obtained from sigma-aldrich company (m , st. louis, mo, usa). for detecting the light penetration depth of the hyperspectral imaging system, different thicknesses ( mm, mm, mm, mm and mm) of pure milk powders were layered on the top of pure melamine. these milk-melamine samples were prepared in black electroplated aluminum plates precisely machined with square wells ( -mm width and height, -mm depth). each sample was first prepared with a leveled layer of pure melamine in the bottom of the well, followed by a leveled layer of pure milk powder on top of the melamine. the combined thicknesses of the melamine and milk powder layers completely filled the -mm depth of each well, with the milk depth ranging between and mm (i.e., mm-thick milk layer over mm of melamine, mm-thick milk layer over mm of melamine, and so on). figure illustrates the preparation of a sample containing mm of milk over mm of melamine. for each kind of milk powder, three samples were prepared at each of the five milk depths, as well as three samples of pure dry milk ( mm of milk with no melamine underneath) and one sample of pure melamine. with plates prepared for each of the milk powder products, a total of samples were measured for the eight milk products used in this study. figure . milk-melamine sample holder (e.g., mm-thick milk and mm-thick melamine layers). the light grey area shows the sample surface ( mm × mm). . . instrument and experiment as shown in the schematic in figure and described in detail in kim et al. [ ], the line-scan nir hyperspectral imaging system consists of an ingaas focal-plane-array (fpa) camera with × pixels (xenics, model xeva- . - , leuven, belgium), an imaging spectrograph (swir hyperspec, headwall photonics, ma, usa) and a -mm zoom lens (optec, model ob-swir / , parabiago, italy), as well as a computer for controlling the camera and acquiring images, two -w dc light sources with fiber optic bundles (dolan jenner, model dc- , boxborough, ma, usa) and a motorized uniaxial stage (velmex, model xn - -m - , bloomfield, ny, usa). the camera array sensor consists of usable pixels in the spectral dimension over a wavelength range of . – . nm, with an average wavelength spacing of . nm. except for the two -w quartz tungsten halogen light sources, the system is entirely housed with an aluminum-framed enclosure. the light is conveyed via the low-oh fiber optic bundles, with each bundle terminating in a mm-long line of fibers encased in a machined aluminum head. the angle of incidence from the line lights is ° (from surface normal), and the distance between line lights and sample surface is mm. for each light source, a mechanical iris was used to allow approximately % of -w light intensity to arrive at the sample surface. the motorized stage moved the samples incrementally across the linear field of view for step-by-step acquisition of line-scan images. mm melamine mm milk figure . milk-melamine sample holder (e.g., mm-thick milk and mm-thick melamine layers). the light grey area shows the sample surface ( mm ˆ mm). . . instrument and experiment as shown in the schematic in figure and described in detail in kim et al. [ ], the line-scan nir hyperspectral imaging system consists of an ingaas focal-plane-array (fpa) camera with ˆ pixels (xenics, model xeva- . - , leuven, belgium), an imaging spectrograph (swir hyperspec, headwall photonics, ma, usa) and a -mm zoom lens (optec, model ob-swir / , parabiago, italy), as well as a computer for controlling the camera and acquiring images, two -w dc light sources with fiber optic bundles (dolan jenner, model dc- , boxborough, ma, usa) and a motorized uniaxial stage (velmex, model xn - -m - , bloomfield, ny, usa). the camera array sensor consists of usable pixels in the spectral dimension over a wavelength range of . – . nm, with an average wavelength spacing of . nm. except for the two -w quartz tungsten halogen light sources, the system is entirely housed with an aluminum-framed enclosure. the light is conveyed via the low-oh fiber optic bundles, with each bundle terminating in a mm-long line of fibers encased in a machined aluminum head. the angle of incidence from the line lights is ˝ (from surface normal), and the distance between line lights and sample surface is mm. for each light source, a mechanical iris was used to allow approximately % of -w light intensity to arrive at the sample surface. the motorized stage moved the samples incrementally across the linear field of view for step-by-step acquisition of line-scan images. sensors , , of sensors , , of figure . schematic of the nir hyperspectral imaging system used to acquire reflectance images of milk powders. fpa, focal-plane-array. in this study, square imaging pixels were achieved by setting the incremental step size to . mm to match the pixel-to-pixel distance along the imaging line, which included pixels. the camera exposure time was set at . ms, and a total of scans were acquired in min. the camera digitized raw energy readings in -bit resolution. finally, each sample’s spectral data were stored as a -bit hyperspectral image cube of dimensions × × containing spatial and spectral data. the hypercube is a three-dimensional image, which represents a d spatial image with x-axis and y-axis coordinate information and z-axis spectral information. . . hyperspectral data analysis . . . image preprocessing in this study, dark current and white reference images were collected to correct the raw reflectance images for wavelength-dependent system responses and heterogeneous dark current in the fpa camera. the dark current image was captured while the lens was covered by a lens cap. an image of an illuminated % diffuse reflectance standard (spectralontm, srt- - , labsphere, nh, usa) was acquired for use as the white reference image. these images were used to calculate the relative reflectance image of each sample, which was calculated by dividing the difference in energy readings between sample and dark current by the difference in energy readings between the white reference and dark current. figure . schematic of the nir hyperspectral imaging system used to acquire reflectance images of milk powders. fpa, focal-plane-array. in this study, square imaging pixels were achieved by setting the incremental step size to . mm to match the pixel-to-pixel distance along the imaging line, which included pixels. the camera exposure time was set at . ms, and a total of scans were acquired in min. the camera digitized raw energy readings in -bit resolution. finally, each sample’s spectral data were stored as a -bit hyperspectral image cube of dimensions ˆ ˆ containing spatial and spectral data. the hypercube is a three-dimensional image, which represents a d spatial image with x-axis and y-axis coordinate information and z-axis spectral information. . . hyperspectral data analysis . . . image preprocessing in this study, dark current and white reference images were collected to correct the raw reflectance images for wavelength-dependent system responses and heterogeneous dark current in the fpa camera. the dark current image was captured while the lens was covered by a lens cap. an image of an illuminated % diffuse reflectance standard (spectralontm, srt- - , labsphere, nh, usa) was acquired for use as the white reference image. these images were used to calculate the relative reflectance image of each sample, which was calculated by dividing the difference in energy readings between sample and dark current by the difference in energy readings between the white reference and dark current. . . . spectral preprocessing the region of interest (roi) for each sample was composed of a square region of , pixels ( ˆ pixels around the image center), selected within the mm ˆ mm sample area (shown in figure ) so as to include only milk powder areas and exclude background (plate) regions for further sensors , , of analyses. spectral preprocessing techniques were used prior to develop a calibration model in the quest for improving the subsequent classification model. common preprocessing methods for nir spectra were used, including standard normal variate (snv), multiplicative scatter correction (msc), extended msc (emsc), normalized (normalized), the common (base ) logarithm (log ), savitzky–golay smooth (smooth) and the savitzky–golay st ( st) and the savitzky–golay nd ( nd) derivative [ ]. . . . development of the classification model after spectral preprocessing, partial least squares discriminant analysis (pls-da) was used to classify each pixel as belonging to either the pure milk class or the milk-melamine mixture class. pls-da, an extension of pls modeling, aims to find the variables and directions in a multivariate space that discriminate the known classes in the calibration set. pls components are computed under the constraint of the maximization of covariance between inputs and outputs. therefore, it can provide a set of orthogonal factors that have the best predictive power from the combinations of different methods with an increased number of variables [ – ]. prior to model development, the triplicate samples (three samples imaged at each depth of milk over melamine and three samples for each pure milk powder product) were divided into two groups, with two samples (comprising , roi pixels) assigned to a model development dataset and one sample (comprising , roi pixels) assigned to a validation dataset for use in evaluating model performance. model performance was compared based on classification accuracies, i.e., the percentage of correctly-classified pixels over the total number of pixels. to better assess the performance of the classification models, calibrations and validations for each depth were run three times [ ]. for each of the different milk-melamine preparations, a separate pls-da model ( -mm model, -mm model, -mm model, -mm model and -mm model) was developed. the number of components chosen for the pls-da models was determined by a contiguous block cross-validation method, in which each block contained the samples from one milk powder product. image processing, selection of the square roi, the spectral preprocessing operation and model development were performed in matlab (r b, mathworks, natick, ma, usa) equipped with the pls toolbox (v. . , eigenvector research, manson, wa, usa). . results and discussion . . hyperspectral spectra figure shows representative mean absorption spectra (calculated from , -pixel roi) of pure melamine and the eight pure milk products (including five nonfat and three whole) obtained from the log preprocessing technique. significant differences, absorption peaks related to the first and second n-h functional group, can be clearly observed between melamine and milk spectra. the mean melamine spectrum had peaks near . and . nm, corresponding to the first overtone of n-h symmetric and anti-symmetric stretching vibration, respectively. the second overtone of n-h stretching vibration is located near . – . nm (centered at . nm). the most significant spectral difference between melamine and milk occurred near . nm, which is attributed to aromatic amine structures [ ] and showed the highest absorption in the melamine spectrum. for the nonfat milk spectra, the spectral patterns of the five different brands are similar, and another similar spectral pattern was observed for the three brands of whole milk from figure . the absorbance of most nonfat milk spectra is lower than that of the whole milk spectra. for the visual difference between the nonfat and whole milk pattern, whole milk spectra have an evident absorption peak around . nm, which is due to the second overtone of c-h stretching vibration constituted by saturated fat structures [ ]. sensors , , of sensors , , of absorption peak around . nm, which is due to the second overtone of c-h stretching vibration constituted by saturated fat structures [ ]. figure . representative mean spectra of pure nonfat (n) and whole (w) milk powders and pure melamine, each calculated from a , -pixel roi. figure . mean roi spectra of samples prepared using (a) ‘valley’ (n) nonfat milk and (b) ‘peak’ (w) whole milk, including pure milk samples and milk-melamine samples with milk depths from – mm (thickness). plots a and c show the full spectra, while plots b and d show the enlarged view of the mean spectra near the . -nm melamine peak. figure . representative mean spectra of pure nonfat (n) and whole (w) milk powders and pure melamine, each calculated from a , -pixel roi. sensors , , of absorption peak around . nm, which is due to the second overtone of c-h stretching vibration constituted by saturated fat structures [ ]. figure . representative mean spectra of pure nonfat (n) and whole (w) milk powders and pure melamine, each calculated from a , -pixel roi. figure . mean roi spectra of samples prepared using (a) ‘valley’ (n) nonfat milk and (b) ‘peak’ (w) whole milk, including pure milk samples and milk-melamine samples with milk depths from – mm (thickness). plots a and c show the full spectra, while plots b and d show the enlarged view of the mean spectra near the . -nm melamine peak. figure . mean roi spectra of samples prepared using (a) ‘valley’ (n) nonfat milk and (b) ‘peak’ (w) whole milk, including pure milk samples and milk-melamine samples with milk depths from – mm (thickness). plots a and c show the full spectra, while plots b and d show the enlarged view of the mean spectra near the . -nm melamine peak. sensors , , of figure shows representative mean spectra for one nonfat milk and one whole milk, valley (nonfat (n)) and peak (whole (w)), including milk-melamine samples for milk depths of – mm (thickness) and for pure milk. among the eight milk products used in this study, ‘peak’ (w) had the strongest absorbance and ‘valley’ (n) had the weakest absorbance. the mean spectra of milk-melamine samples are very similar to the mean spectrum of pure milk; all exhibit no obvious melamine absorption features. this observation suggests that individual pixel-based spectral evaluations (instead of averaged spectra across spatial image areas) may allow better detection of melamine to measure the penetration depth of milk [ ]. in the region of the . -nm melamine absorption peak, the mean absorbance notably decreased as the milk depth increased from mm– mm. for mm– mm, the mean spectra were nearly the same as that for pure milk. the same trends were observed for both nonfat milk ‘valley’ and whole milk ‘peak’. . . discriminant models for milk depth classification the plots in figure compare the validation set classification results (average of three runs) for the pls-da models coupled with specific spectral preprocessing algorithms, for milk-melamine at each of the five different milk depths ( -mm– -mm models). as shown in figure , the different preprocessing algorithms can have a great impact on model accuracy. compared to the other spectra preprocessing algorithms, the nd derivative consistently resulted in low classification accuracy. the reason may be that it has a more prominent spectra shoulder after derivative transformation, but it affects the component number selection of pls-da. the models coupled with the snv and msc algorithm gave better, more robust performance than the normalized model (mean zero, unit variance), which is due to reducing the scattering influence from particle size. since the classification accuracies of the calibration set were better than those of the validation sets, only the classification results of the validation set for milk and milk-melamine samples are presented. table shows the classification results of the validation set based on the pls-da model coupled with the snv preprocessing algorithm for the eight milk powders. the data show the classification accuracy notably decreasing as the milk depth increases from mm– mm. for the -mm model, . %– % of milk samples (pixels) and . %– . % of milk-melamine samples were correctly classified, for overall accuracies of . % and . % for milk and milk-melamine samples, respectively, across the eight milk powders. the -mm model’s highest accuracy was the same as that for the -mm model, while its lowest accuracies were lower at . % for milk and . % for milk-melamine samples. the average misclassification rates were . % for milk and . % for milk-melamine samples across all eight milk powders. although the -mm model achieved an average accuracy . % for milk and . % for milk-melamine samples, the classification accuracies of nonfat milk ‘hoosier’ and ‘nestle’ were lower than %, which is not suitable for melamine detection at lower concentrations. as the milk depth increased to mm and mm, the nearly identical spectra for milk and for milk-melamine (shown in figure ) resulted in deteriorated model performance. the average misclassification rate of the -mm model was greater than %, while about % of the samples were misclassified by the -mm model for some of the milk powders. this means that the -mm and -mm models were invalid for classification. sensors , , of sensors , , of figure . classification comparison of classification results for pls-da models coupled with specific spectral preprocessing algorithms for milk-melamine samples at milk depths from mm to mm (a–e). figure . classification comparison of classification results for pls-da models coupled with specific spectral preprocessing algorithms for milk-melamine samples at milk depths from mm to mm (a–e). sensors , , of table . classification results of the validation set for milk and milk-melamine samples from a -mm– -mm depth using pls-da coupled with the standard normal variate (snv) spectra preprocessing algorithm. classification (%) depth mm mm mm mm mm valley (n) . . . . . hoosier (n) . . . . . nestle (n) . . . . . bob (n) . . . . . milk now (n) . . . . . nestle (w) . . . . . hoosier (w) . . . . . peak (w) . . . . . average . . . . . valley (n) . . . . . hoosier (n) . . . . . nestle (n) . . . . . milk- bob (n) . . . . . melamine now (n) . . . . . nestle (w) . . . . . hoosier (w) . . . . . peak (w) . . . . . average . . . . . figure shows the classification results for two brands of nonfat and whole milk using the pls-da model coupled with the snv spectra preprocessing algorithm. for the same brand milk powder, the classification results of whole milk were slightly higher than those of nonfat milk for the -mm– -mm valid models. sensors , , of table . classification results of the validation set for milk and milk-melamine samples from a -mm– -mm depth using pls-da coupled with the standard normal variate (snv) spectra preprocessing algorithm. classification (%) depth mm mm mm mm mm valley (n) . . . . . hoosier (n) . . . . . nestle (n) . . . . . bob (n) . . . . . milk now (n) . . . . . nestle (w) . . . . . hoosier (w) . . . . . peak (w) . . . . . average . . . . . valley (n) . . . . . hoosier (n) . . . . . nestle (n) . . . . . milk- bob (n) . . . . . melamine now (n) . . . . . nestle (w) . . . . . hoosier (w) . . . . . peak (w) . . . . . average . . . . . figure shows the classification results for two brands of nonfat and whole milk using the pls-da model coupled with the snv spectra preprocessing algorithm. for the same brand milk powder, the classification results of whole milk were slightly higher than those of nonfat milk for the -mm– -mm valid models. figure . comparison of the classification results for the same brand nonfat and whole milk using the pls-da model coupled with the snv spectral preprocessing algorithm for (a) milk and (b) milk-melamine. . conclusions in this study, the penetration depth of near-infrared hyperspectral imaging light ( . – . nm) was investigated for milk powders. five different depths of milk powder, from mm– mm, were investigated for the detection of melamine underneath the milk powder. figure . comparison of the classification results for the same brand nonfat and whole milk using the pls-da model coupled with the snv spectral preprocessing algorithm for (a) milk and (b) milk-melamine. . conclusions in this study, the penetration depth of near-infrared hyperspectral imaging light ( . – . nm) was investigated for milk powders. five different depths of milk powder, from mm– mm, were investigated for the detection of melamine underneath the milk powder. classification models were developed using the pls-da technique for milk and milk-melamine samples prepared using five brands of non-fat milk powder and three brands of whole milk powder. the classification results sensors , , of showed that the classification accuracy gradually decreased as the milk depth increased. for the -mm models, the classification accuracies were higher than % for both milk and milk-melamine samples for all of the milk powders under investigation. it can be concluded that the use of a -mm milk powder depth can be recommended for applying nir hyperspectral imaging for the detection of contaminants in milk powder. in addition, the method described can also be potentially applied to other food powders for penetration depth measurement of nir hyperspectral imaging system. acknowledgments: this work was partially supported by grants from the agenda programs (pj ) and (pj ), rural development administration, korea, the national natural science foundation of china (grant # and ), and the (b ) and qing lan projects of china. the authors would like to thank diane chan of environmental microbial and food safety laboratory, agricultural research service, usda, for reviewing the manuscript. author contributions: m.h., m.s.k, k.c., j.q., c.m and s.d. conceived of and designed the experiments. m.h. and m.s.k. performed the experiments. m.h. and c.e. analyzed the data. m.h. and q.z. wrote the paper. m.s.k. and s.d. edited the paper. conflicts of interest: the authors declare no conflict of interest. references . filigenzi, m.s.; puschner, b.; aston, l.s.; poppenga, r.h. diagnostic determination of melamine and related compounds in kidney tissue by liquid chromatography/tandem mass spectrometry. j. agric. food. chem. , , – . [crossref] [pubmed] . xu, x.; ren, y.; zhu, y.; cai, z.; han, j.; huang, b.; zhu, y. direct determination of melamine in dairy products by gas chromatography/mass spectrometry with coupled column separation. anal. chim. acta , , – . [crossref] [pubmed] . mauer, l.j.; chernyshova, a.a.; hiatt, a.; deering, a.; davis, r. melamine detection in infant formula powder using near- and mid infrared spectroscopy. j. agric. food. chem. , , – . [crossref] [pubmed] . balabin, r.m.; smirnov, s.v. melamine detection by mid-and near-infrared (mir/nir) spectroscopy: a quick and sensitive method for dairy products analysis including liquid milk, infant formula, and milk powder. talanta , , – . [crossref] [pubmed] . wu, d.; sun, d. advanced applications of hyperspectral imaging technology for food quality and safety analysis and assessment: a review—part i: fundamentals. innovative food sci. emerg. technol. , , – . [crossref] . cogdill, r.; hurburgh, c.; rippke, g. single-kernel maize analysis by near-infrared hyperspectral imaging. trans. asae , , – . [crossref] . qin, j.; lu, r. detection of pits in tart cherries by hyperspectral transmission imaging. trans. asae , , – . [crossref] . coelho, p.a.; soto, m.e.; torres, s.n.; sbarbaro, d.g.; pezoa, j.e. hyperspectral transmittance imaging of the shell-free cooked clam mulinia edulis for parasite detection. j. food eng. , , – . [crossref] . huang, m.; wan, x.; zhang, m.; zhu, q. detection of insect-damaged vegetable soybeans using hyperspectral transmittance image. j. food eng. , , – . [crossref] . huang, m.; lu, r. apple mealiness detection using hyperspectral scattering technique. postharvest biol. technol. , , – . [crossref] . lu, r.; peng, y. hyperspectral scattering for assessing peach fruit firmness. biosystems eng. , , – . [crossref] . gowen, a.a.; taghizadeh, m.; o’donnell, c.p. identification of mushrooms subjected to freeze damage using hyperspectral imaging. j. food eng. , , – . [crossref] . ariana, d.p.; lu, r.; guyer, d.e. hyperspectral reflectance imaging for detection of bruises on pickling cucumbers. comput. electron. agric. , , – . [crossref] . chao, k.; yang, c.; kim, m.s.; chan, d. high throughput spectral imaging system for wholesomeness inspection of chicken. appl. eng. agric. , , – . [crossref] . qin, j.; lu, r. measurement of the optical properties of fruits and vegetables using spatially resolved hyperspectral diffuse reflectance imaging technique. postharvest biol. technol. , , – . [crossref] http://dx.doi.org/ . /jf s http://www.ncbi.nlm.nih.gov/pubmed/ http://dx.doi.org/ . /j.aca. . . http://www.ncbi.nlm.nih.gov/pubmed/ http://dx.doi.org/ . /jf m http://www.ncbi.nlm.nih.gov/pubmed/ http://dx.doi.org/ . /j.talanta. . . http://www.ncbi.nlm.nih.gov/pubmed/ http://dx.doi.org/ . /j.ifset. . . http://dx.doi.org/ . / . http://dx.doi.org/ . / . http://dx.doi.org/ . /j.jfoodeng. . . http://dx.doi.org/ . /j.jfoodeng. . . http://dx.doi.org/ . /j.postharvbio. . . http://dx.doi.org/ . /j.biosystemseng. . . http://dx.doi.org/ . /j.jfoodeng. . . http://dx.doi.org/ . /j.compag. . . http://dx.doi.org/ . / . http://dx.doi.org/ . /j.postharvbio. . . sensors , , of . lammertyn, j.; peirs, a.; de baerdemaeker, j.; nicolaï, b. light penetration properties of nir radiation in fruit with respect to non-destructive quality assessment. postharvest biol. technol. , , – . [crossref] . fernández pierna, j.a.; vincke, d.; dardenne, p.; yang, z.; han, l.; baeten, v. line scan hyperspectral imaging spectroscopy for the early detection of melamine and cyanuric acid in feed. j. near infrared spectrosc. , , – . [crossref] . fu, x.; kim, m.s.; chao, k.; qin, j.; lim, j.; lee, h.; garrido-varo, a.; pérez-marín, d.; ying, y. detection of melamine in milk powders based on nir hyperspectral imaging and spectral similarity analysis. j. food eng. , , – . [crossref] . kim, m.s.; chen, y.r.; mehl, p.m. hyperspectral reflectance and fluorescence imaging system for food quality and safety. trans. asae , , – . . rinnan, a.; van den berg, f.; engelsen, s.b. review of the most common pre-processing techniques for near-infrared spectra. trends anal. chem. , , – . [crossref] . barker, m.; rayens, w. partial least squares for discrimination. j. chemom. , , – . [crossref] . westerhuis, j.a.; hoefsloot, h.c.j.; smit, s.; vis, d.j.; smilde, a.k.; van velzen, e.j.j.; van duijnhoven, j.p.m.; van dorsten, f.a. assessment of plsda cross validation. metabolomics , , – . [crossref] . huang, m.; ma, y.; li, y.; zhu, q.; huang, g.; bu, p. hyperspectral image-based feature integration for insect-damaged hawthorn detection. anal. methods , , – . [crossref] . lu, r.; huang, m.; qin, j. analysis of hyperspectral scattering characteristics for predicting apple fruit firmness and soluble solids content. in proceedings of the international conference on spie , sensing for agriculture and food quality and safety, orlando, fl, usa, april . . nagarajan, r.; singh, p.; mehrotra, r. direct determination of moisture in powder milk using near infrared spectroscopy. j. autom. methods manag. chem. , , . [crossref] [pubmed] © by the authors; licensee mdpi, basel, switzerland. this article is an open access article distributed under the terms and conditions of the creative commons by attribution (cc-by) license (http://creativecommons.org/licenses/by/ . /). http://dx.doi.org/ . /s - ( ) -x http://dx.doi.org/ . /jnirs. http://dx.doi.org/ . /j.jfoodeng. . . http://dx.doi.org/ . /j.trac. . . http://dx.doi.org/ . /cem. http://dx.doi.org/ . /s - - - http://dx.doi.org/ . /c ay a http://dx.doi.org/ . /jammc/ / http://www.ncbi.nlm.nih.gov/pubmed/ http://creativecommons.org/ http://creativecommons.org/licenses/by/ . / 	introduction 	experimental section 	sample preparation 	instrument and experiment 	hyperspectral data analysis 	image preprocessing 	spectral preprocessing 	development of the classification model 	results and discussion 	hyperspectral spectra 	discriminant models for milk depth classification 	conclusions for. sci. ( ): – doi: . /forsci/fxz forest science • december fundamental research silviculture effect of prescribed fire on timber volume and grade in the hoosier national forest shannon stanis, jan wiedenbeck, and mike r. saunders low-intensity surface fire is prescribed in eastern north american hardwood stands prior to overstory harvest in order to improve regeneration and recruitment of oak and other fire-tolerant and fire-adapted species. however, this use of prescribed fire potentially can reduce timber value. we inventoried overstory trees (> in. diameter at breast height) in oak-dominated stands with varied prescribed fire histories and aspects in southern indiana. we then documented the extent of prescribed fire damage (i.e., wounds) to overstory trees and quantified both the relative stand volume of timber loss and the proportion of trees that had tree grade reductions because of prescribed fire. generally, as a stand received more prescribed fires, more trees were scarred, the relative volume lost increased, and a higher proportion of trees declined in grade. overall, burned stands experienced less than percent sawtimber volume loss, regardless of the number of prescribed fires and aspect. less than percent of trees, study- wide, had reduced grade because of prescribed fire. grade and volume reductions varied by species, however. our results suggest that prescribed fire has a minor economic impact on standing timber, particularly when timber is harvested within two decades of the first fire. keywords: central hardwoods, bole wounds, usfs tree grades, fire ecology, quercus fire is a disturbance agent that influences many ecosystem processes and shapes the structure and composition of nu-merous biomes throughout the world (bowman et al. , dey and schweitzer ). fires set by native americans were an integral part of most terrestrial ecosystems in eastern north america for thousands of years prior to european colonization (brose et  al. ) and contributed to the dominance of fire- adapted oak (quercus) species in forests throughout the region (abrams ). fire exclusion over the last century has contributed to the lack of advanced oak regeneration and the increased regen- eration of mesic species such as beech and maple (fagus and acer) within many forests (nowacki and abrams , arthur et  al. , brose et al. ). fire can encourage development of advanced oak regeneration (brose et  al. , , ) and is increasingly prescribed by federal and state agencies and non-government organizations for ecosystem restoration (dey and schweitzer ). oak forests without fire are experiencing compositional changes as more mesic species continue to dominate in the midstory, particularly on more productive sites (steiner et  al. ). much of eastern north america’s oak forests are on the verge of transitioning to late successional, mesic forests. this change in overstory composition could have dramatic ecological consequences; oaks are a founda- tional species that many animal species rely upon for both food and habitat (hanberry and nowacki ). loss of oak could also have huge financial ramifications, as these species are used for a vast array of timber products (buehlmann et  al. ) and, for white oak (q. alba) in particular, cooperage for the expanding distilling industry (delany and haynes ). modern natural-resource management frequently aims to meet multiple objectives while balancing tradeoffs (bradford and d’amato ). there are numerous factors that limit the manuscript received september , ; accepted may , ; published online june , . affiliations: shannon stanis (shannonstanis@gmail.com) and mike r.  saunders (msaunder@purdue.edu), department of forestry and natural resources, hardwood tree improvement and regeneration center, purdue university, west state street, west lafayette, in . jan wiedenbeck (jwiedenbeck@ fs.fed.us), usda forest service, northern research station, hardwood lane, suite b, princeton, wv . acknowledgments: the authors would like to thank the staff of the hoosier national forest, namely chris thornton, jeremy kolaks, and ryan otto, for access to gis files and paper records during site selection, access to field sites, use of vehicles, and use of office space in conducting this research. we would like to thank dr. tom schuler of the usda forest service—washington office for his assistance in securing funding and overall support of this project. we would also like to thank rick hovatter and donnie lowther of the usda forest service—northern research station, and don carlson at purdue university for training in grade determi- nation of timber. skye greenler, dave ralston, mike szuter, kristen bellisario, ken kellner, jack vanschaik, destiney priest, john lang, joni willits, caitlin horsch, tom daniel, kelly derolf, and nate elder all assisted in data collection and/or statistical analysis. funding for this project was provided by the usda forest service—northern research station (agreement: -jv- - ), the joint fire sciences program (project: - - - ), and the usda—nifa mcintire-stennis (project: ind ms). published by oxford university press on behalf of the society of american foresters . this work is written by (a) us government employee(s) and is in the public domain in the us. d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch mailto:shannonstanis@gmail.com?subject= mailto:msaunder@purdue.edu?subject= mailto:jwiedenbeck@fs.fed.us?subject= mailto:jwiedenbeck@fs.fed.us?subject= forest science • december implementation of prescribed fire (e.g., short burn windows, smoke concerns, resource allocation, liability, expertise, and reg- ulatory requirements) for ecosystem management. there is also a pervasive concern that prescribed fire will negatively impact timber resources; therefore, many land managers and forestland owners hesitate to use fire (dey and schweitzer , ). previous studies that investigated the effect of periodic, low- intensity prescribed fires on standing hardwood timber volume and value reported minimal losses (stambaugh and guyette , marschall et  al. , wiedenbeck and schuler , knapp et  al. ). however, most previous work focused on the less fire-tolerant red oak (erythrobalanus) group, which was less prev- alent on the landscape prior to fire suppression than the white oak group (lepidobalanus) (abrams ). as white oaks occupy a large volume of the standing sawtimber-sized trees in eastern north america— million bd ft (doyle) in indiana alone (gormanson and kurtz )—the effects of prescribed fire on white oak timber need to be more fully explored. to quantify the effects of prescribed fire on timber volume and quality, we measured prescribed fire damage of overstory trees within oak-dominated hardwood stands in the hoosier national forest (hnf) in southern indiana. stands varied by the number of fires received over   years (zero to five burns) and by aspect. we hypothesized that, as the number of prescribed fires a stand receives increases, the proportion of trees wounded by fire would increase, the relative amount of timber volume lost would increase, and more trees would decline in grade. we further hypothesized that wounding, timber loss, and grade loss are exacerbated in stands with more xeric (i.e., south- and west-facing) than mesic (i.e., north- and east-facing) aspects and that there would be some differences among commercially valuable species groups in fire- induced damage. materials and methods study site the hnf encompasses nearly , ac of southern indiana (figure ). most of the region was cleared for agricultural use in the early s and subsequently abandoned. forests regenerated on these lands, and the hnf was acquired in patches by the united states forest service (usfs) throughout the th century. most of the hnf is second- or third-growth, although very few rem- nant +-year-old stands still exist (maxwell and harley ). all stands chosen for this study were second-growth. the hnf is within the brown county hills and mitchell karst plain sections of the highland rim natural region and the crawford upland and escarpment sections of the shawnee hills natural region (figure ; homoya et  al. ). forests of both sections of the largely unglaciated highland rim natural region are dominated by oak and hickory (carya) species, particularly on ridges and south-facing aspects; ravines and north-facing aspects in both sections include more mesic species such as american beech (f.  grandifolia), northern red oak (q.  rubra), and sugar maple (a. saccharum). in the mitchell karst plain section, post oak (q. stellata) and chinquapin oak (q. muehlenbergii) are common on xeric sites (homoya et al. ). the shawnee hills natural region is mostly unglaciated and primarily composed of upland forests. upper slope positions in this region are dominated by black oak (q.  velutina), white oak, chestnut oak (q.  montana), scarlet oak (q.  coccinea), post oak, pignut hickory (c. glabra), and shagbark hickory (c. ovata). lower slope positions and north-facing aspects are composed of beech, tulip poplar (liriodendron tulipifera), northern red oak, sugar maple, and black walnut (juglans nigra). post and black oaks are more common than chestnut oak on dry sites within the escarpment sec- tion (homoya et al. ). management and policy implications perpetuation of oak-dominated ecosystems in the eastern united states can rely on prescribed fire to create forest-floor conditions conducive to oak re- generation. resistance by some managers to use prescribed fire, for fear of reducing overstory timber value, is often driven by observations and research on wildfire-damaged trees and may be misplaced for prescribed fire. on mesic sites and in mesic regions where prescribed fire intensities are kept low to only regenerate oak stands (i.e., not to thin the overstory), prescribed fire likely reduces a stand’s standing sawtimber volume through wounding by less than percent and has negligible effects on its average tree grade. on more xeric sites and in more xeric regions, sawtimber volume loss could be higher, ex- ceeding percent, but this level of damage likely occurs only after multiple fires. under a goal to promote regeneration, economic damage to timber value from prescribed fire likely is minimal, particularly compared to the costs of regenerating oak artificially as an alternative option. nevertheless, managers concerned about damage to veneer-quality trees may consider taking steps to protect those individuals during prescribed fire treatments or planning to harvest those trees soon (i.e., <  years) after treatment. figure . study sites (+) within the hoosier national forest in southern indiana, usa. sections of the highland rim natural region (light reds) and shawnee hills natural regions (light greens) are also shown (homoya et  al. ). at the scale of this map, study site symbols overlap considerably as sites occurred within only burn units across the hoosier national forest. d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch forest science • december stand selection outside three significant wildfires in the s and s, stands within the hnf had not experienced fire since its establish- ment in the s (j. travis swain, pers. commun., hnf forest silviculturist, usda forest service, january , ). during spring , managers began a prescribed fire program to restore rare, fire-dependent barrens in the forest; the program has since expanded to meet multiple restoration, habitat creation, and regeneration objectives within oak forests and grasslands throughout the forest. for this study, stands had to have received at least one prescribed fire over the past  years, and not occur in those areas with a history of wildfire. most selected stands had received prescribed fire in order to reduce understory and midstory densities of mesic species and to promote advanced oak regeneration. however, some of the stands with the longest prescribed fire histories were adjacent to barrens where fire was being used to encourage warm season grasses and prairie forbs, and to limit the encroachment of woody species. fires were almost exclusively prescribed in the spring burning season from early february to late march. selection of stands was stratified by two factors, burn class and aspect. burn class was the number of prescribed fires a stand re- ceived and categorically assigned as (i.e., controls), , , , or + prescribed fires. aspect was divided into two classes: mesic or xeric. stands with predominantly north to east aspects ( . ° to . ° azimuth) are generally more mesic and somewhat protected from intense fire; stands with predominantly south- to west-facing aspects ( . ° to . ° azimuth) are more xeric and, thus, will generally experience more intense fires (pyne et  al. ). in ad- dition, selected stands had to be merchantable (i.e., mean stand diameter at breast height [dbh] >  in.) and dominated by oak and hickory species in terms of basal area. in sum, we selected stands that had received at least one prescribed fire. we then selected an additional stands within close proximity, on similar aspects, and within the same tree size and species criteria, but without a history of prescribed fire, to serve as controls (i.e., burn class  =  ) for the study. therefore, five to seven stands were selected in each aspect × burn class combination. across all stands, white oak site indices ranged from ~ to ft (base age of  year; schnur ). details on each stand can be found in the supplementary (table s ). field data collection we sampled each stand with variable radius points using a baf prism (i.e., ft ac− ) to inventory living trees > in. dbh at each point. restricted randomization was used to select prism points based on three criteria: ( ) local, neighborhood aspect for at least eight of the points had to correspond to the prevailing aspect for the stand (e.g., a xeric stand had to have a minimum of eight points on south- or west-facing aspects); ( ) points were placed at least ft apart to prevent double sampling of indi- vidual trees; and ( ) points needed a minimum of three “in” trees to be included (i.e., not in a large opening or old landing). points were moved to the nearest location that met these criteria (fewer than instances in the entire study). we recorded species, dbh, us forest service (usfs) tree grade ( , , , and local use; hanks , miller et al. ), merchantable height to a -inch-diameter outside bark (dob) top for each “in” tree, and the presence or ab- sence of any externally visible fire wounding. if a tree had visible wounding that was likely caused by fire (e.g., by presence of char in or near wound), the tree was given two usfs grades, following loomis ( ), where the first grade included all defects, and the second grade ignored any defect likely caused by fire. all wounds beginning below breast height and above stump height (i.e., in.) were characterized by measuring the position on the stem (upslope, sideslope, downslope), type, width at the widest point, start height and total height (in relation to upslope), and depth to the nearest .  inch. all measurements included any decay thought to result from the wound. wounds were only measured to a maximum of ft in height, but notes were taken if the wound extended further up the stem. exterior wounds that occurred only in the stump were not measured, but noted if damage from the wound likely led to cull above stump height (e.g., detectable in- ternal decay through sounding the tree). wound types included: ( ) catfaces (triangle-shaped, open at base of tree), ( ) ovals (oval-shaped, closed at base of tree), ( ) seams, ( ) basal/flutes, and ( ) bark slough (stambaugh and guyette ). catfaces and ovals have distinct wound ribs (thickened rings produced by the cambium in response to injury; smith and sutherland ) when viewed on the cross-section and have asso- ciated depth to their wounds. seams also have distinct wound ribs, but rarely have a visible depth, as many seams are fully overgrown catface or oval wounds. basal/flutes are wounds that occurred on extended butt flare. bark slough was used to indicate when dead bark was disconnected from the bole of the tree, when multiple seams occupied a confined area, and/or when physical damage was evident, but the damage did not meet definitions for the other categories. a standard depth of .  inch was assigned to all wounds without a measurable depth and exhibiting no decay. data compilation species with more than sampled individuals across all sites were grouped into categories based on current local timber markets: hickories (pignut and shagbark hickory); red oaks (northern red, scarlet, and black oak); sugar maple; tulip poplar; white oaks (white and chinquapin [only three instances] oak); and other merchant- able white oaks (post and chestnut oak). species with fewer than sampled individuals across all sites, including many not mer- chantable in local markets, were grouped together into an “other” type; these species were red maple (a.  rubrum), silver maple (a.  saccharinum), dogwood (cornus spp.), persimmon (diospyros virginiana), american beech, ash species (fraxinus spp.), black walnut, sweetgum (liquidambar styraciflua), black gum (nyssa sylvatica), american sycamore (platanus occidentalis), big-toothed and trembling aspen (populus spp.), black cherry (prunus serotina), blackjack oak (q.  marilandica), sassafras (sassafras albidum), and american elm (ulmus americana). stand-level estimates of basal area (ba; ft ac− ), density (tpa; trees ac− ), and quadratic mean diameter (qmd; in.) were calcu- lated for each stand. board foot volume of the first -ft log (v ) and total volume to merchantable height (vt; -in. dob) of each tree were calculated in international quarter-inch scale using equation (wiant ): v = ( . l + . l − . ) + ( . − . l − . l ) × dbh + ( . − . l + . l ) × dbh ( ) d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data forest science • december where l  is the number of -ft logs to the nearest . log and assuming a girard form class of . if a tree was not of sawtimber quality, then volume was set to zero; if the tree did not have a -ft merchantable butt log, v was calculated based on an -ft or -ft log. to estimate the volume loss because of prescribed fire in both v and vt, we first calculated the volume of each wound using formulas consistent with their shape, converted those volumes to board feet, and aggregated volumes for all wounds on each tree. the volume of the tree with fire damage (i.e., v ,fire and vt,fire) was calculated by subtracting the aggregated wound volume from v and vt. volume was also adjusted as needed for sections of the trees determined to be unusable (e.g., swell in the bottom feet). each of these tree volumes (i.e., v , vt, v ,fire, and vt,fire) was converted to a per-acre basis and then summed to calculate stand-level values. finally, we corrected for differences in overall stocking among all stands by calculating the relative total stand volume loss (per- centage) in each stand as (equation  ): rvx = ( vx − vx,f ire ) /vx × ( ) where x corresponds to either the bottom -ft log volume ( ) or the total merchantable volume (t). to check for species-specific differences in vulnerability to fire, we calculated rv and rvt for two additional subsets of the data: ( ) for merchantable species only (all species groups besides the “other” type) and ( ) for the white oak group only. to estimate the grade loss because of prescribed fire, we calcu- lated the percentage of trees that had differing usfs tree grades (i.e., grade accounting for fire damage and grade ignoring fire damage) per acre in each stand. this was calculated for all overstory trees, only merchantable species, and only white oak. statistical analyses to investigate the potential impact of prescribed fire on stand structure, we performed separate analyses of variance (anova) on the influence of burn class, aspect, and interaction between these variables on stand-level ba, tpa, and qmd. to test for prescribed fire effects on tree wounding, timber volume loss, and tree grade, we conducted an anova on the percentage of trees wounded by fire, on both rv and rvt, and on the average percentage of trees per acre that changed grade, respectively, in each stand with burn class and aspect as main effects. stands were the experimental unit. anova tests used type iii sums of squares because of the unbalanced data structure. all tests were repeated for the three datasets—all species, merchantable species, and white oak only. arcsin transformations were used on wound proportions, relative volume loss, and grade loss data to normalize the data and reduce heteroscedasticity. tukey honestly significant difference (hsd) tests were performed to determine significant trends between factor levels. analyses were conducted in r . . (r core team ), using the car (fox and weisberg ) and agricolae (de mendiburu and simon ) packages for some tests. all analyses used an α = . . results sample profile overall, , trees were sampled with prism plots across the stands. stands were distributed nearly equally across aspect classes (mesic: ; xeric: ). basal area in most stands was dominated by white oak ( ), although some stands were composed of mixtures of oak and hickory species ( ). a few stands (five) were composed of a diverse mixture of mesic species, with oaks and hickories comprising a minor component (supplementary, table s ). overall, there were species inventoried representing the seven species groups, with white oak comprising . percent of inventoried stems. red oaks ( . percent), hickories ( . percent), and the other white oaks ( . percent) were the next most frequently inventoried species groups (table ). across all treatments, , wounds were measured, of which . percent occurred in stands with a history of prescribed fire (burn classes , , , +). for all treatments, seams were the most common type of wound followed by bark sloughing/multiple seams and catfaces (table ). catfaces had the largest average de- fect volume per wound, .  ±  . bd ft (mean ± standard error), whereas seams had the lowest at .  ±  . bd ft. the average de- fect for all wound types was .  ±  . bd ft. all wound types, on average, ended below a height of .  ±  . in. and were .  ±  . in. wide and .  ±  . in. deep (table ). wounds were found on . percent of all trees sampled with prism plots ( , / , ), . percent of merchantable trees ( , / , ), and . percent of white oaks ( / , ). if wounded, trees typically had one or two wounds; only . percent ( ), . percent ( ), and . percent ( ) of all sample trees, merchantable species, and white oaks, respectively, had three or more wounds. consequently, wounds only led to grade reductions for . percent of all sampled trees ( ), . percent of merchant- able species ( ; table ), and . percent of white oaks ( ). stand-level patterns overstory bas ranged from to ft ac− across the stands, with an average of   ±  . ft ac− (table a ). density of stems > in. dbh (den) ranged from to stems ac− and averaged  ±  stems ac− (table a ). qmd ranged from . to . in. and averaged .  ±  . in. (table a ). neither ba, den, nor qmd varied significantly with either burn class, aspect, or table . number of trees inventoried across stands on the hoosier national forest, partitioned by species group to stands with mesic (north- or east-facing) and xeric (south- or west-facing) aspects. species groupa aspect total mesic xeric white oak , , red oak hickory other white oak tulip poplar sugar maple other total , , , aspecies groups were defined as white oak—white (quercus alba) and chinquapin oak (q.  muehlenbergii); red oak—northern red (q.  rubra), scarlet (q.  coccinea) and black oak (q.  velutina); hickory—pignut (carya glabra) and shagbark hickory (c. ovata); other white oak—post (q. stellata) and chestnut oak (q. mon- tana); tulip poplar—tulip poplar (liriodendron tuliperfera); and sugar maple— sugar maple (acer saccharum). the “other” species group contains rarely sampled (n <  ) and/or nonmerchantable species; see methods for more details. d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data forest science • december their interaction (ba: f ,  =  . , p = . ; den: f ,  =  . , p = . ; qmd: f ,  =  . , p = . ) (see also table a ). wounding generally, the percentage of trees per acre wounded increased as the number of prescribed fires increased (all species: f ,  =  . , p = . ; merchantable species: f ,  =  . , p = . ; white oak: f ,  =  . , p = . ), but did not vary across aspects (p > . for all species groups; table a ). control stands with no prescribed fire history (i.e., burn class )  had an average percentage of trees per acre wounded of percent. prescribed fire increased this per- centage to percent for burn classes and +; burn classes and were intermediate in wounding (figure ). tukey hsd tests in- dicated that wounding was significantly different only for stands in burn classes and as compared to unburned control stands (all species: p = . and . ; merchantable species: p = . and . ; white oak: p = . and . , for three and four prescribed fires, respectively); stands with fewer burns were not significantly different from one another or from the control (p > . ; figure ). the interaction between burn class and aspect for percentage of trees per acre with wounds was not significant (see table s ). relative volume lost prescribed fire caused relative volume loss, although the re- sponse was often weak and highly variable. across all species, rvt was < . percent and rv was < . percent in stands with three or fewer prescribed fires, whereas stands with four or more prescribed fires experienced an average rvt and rv of .  ±  . percent and .  ±  . percent, respectively (figure a, b). rvt and rv exceeded percent in only two stands, both on xeric aspects that had been burned four or more times. results from anovas (table s ) in- dicated that burn class strongly influenced both rvt (f ,  =  . , p < . ) and rv (f ,  =  . , p < . ). the interaction be- tween burn class and aspect strongly influenced rvt (f ,  =  . , p  = . ) for all species, but less so for rv (f ,   =  . , p  = . ); this interaction was a result of the aforementioned xeric sites that had been burned four or more times. aspect had no direct in- fluence on rvt or rv (p > . ; table s ). for merchantable species, relative volume results largely mir- rored those for all species (figure c, d). aspect did appear to influence merchantable species rvt and rv , although it was only significant for the latter (rvt: f ,   =  . , p  = . ; rv : f ,  =  . , p = . ; table s ). likewise, the interaction between burn class and aspect strongly influenced rv of merchantable spe- cies (f ,  =  . , p = . ; table s ). white oak relative volume loss also mirrored the trends observed for all species, although relative losses were half those values (figure e, f ). only one stand, lying on a xeric aspect and with four burns, had rvt and rv exceeding percent for white oak; the re- mainder were below . percent for rvt and . percent for rv . only burn class significantly affected rvt (f ,  =  . , p < . ) or rv (f ,  =  . , p < . ; table s ), although tukey hsd tests suggested that the xeric sites with four burns were significantly different from the controls (figure e, f ). this result may be an ar- tifact of the exceptionally high rvt and rv observed in the one stand mentioned earlier. grade change prescribed fire led to grade loss in most stands. the percentage of trees per acre that changed grade averaged .  ±  . percent (range: – . percent) for all species combined, .  ±  . percent (range: – . percent for merchantable species, and .   ±  . percent (range: – . percent) for white oak. the average percentage of trees per acre that changed grade was less than percent for stands with three or fewer prescribed fires for all species, merchantable species, and only white oak; the grade change was higher, however, in stands with four or more burns, nearly . percent for both all species and merchantable species groups, but only . percent for white oak (table ). grade loss was influenced by burn class (f ,   =  . , p  < . ) and the interaction of burn class and aspect for all spe- cies (f ,   =  . , p  = . ; table s ). results were similar for merchantable species. burn class was the only significant factor explaining changes in grade for white oak, however (f ,  =  . , p = . ; table s ). stands receiving four or more burns experienced table . mean (standard error) height, width, depth, and defect volume for individual wounds by type across all sampled trees. wound type n heighta (in.) width (in.) depthb (in.) volumec (bd ft) volume range (bd ft) seam . ( . ) . ( . ) . ( . ) . ( . ) . – . bark slough/multiple seams . ( . ) . ( . ) . ( . ) . ( . ) . – . catface . ( . ) . ( . ) . ( . ) . ( . ) . – . oval . ( . ) . ( . ) . ( . ) . ( . ) . – . basal/flutes . ( . ) . ( . ) . ( . ) . ( . ) . – . all types , . ( . ) . ( . ) . ( . ) . ( . ) . – . note: individual wounds could be caused, in whole or part, by prescribed fire (i.e., on burn class >  ) or only by other agents (i.e., burn class =  ). range in defect volume by wound type as observed across all sample trees is also provided. aheight was defined as the height to the top of the defect on the upslope side of the tree. height is an indication of what length of the butt log was affected by the wound. bclosed seams and other wounds were assumed to have a minimal depth of . in. cdefect volume included any visible decay that resulted from the wound. it was calculated consistent with its general geometric shape of the wound (e.g., catfaces as a triangular prisms, seams as rectangular prisms). table . classification matrix of usfs tree grades for sampled trees of merchantable species measured across stands of the hoosier national forest (n =  , ), both considering (i.e., observed) and ignoring fire damage. observed grade grade ignoring fire damage local use , – – – – – – local use note: values on the diagonal are the number of trees where grade did not change because of fire damage (i.e., observed grade = grade ignoring fire damage); values below the diagonal are the number of trees where grade declined because of fire damage (i.e., observed grade < grade ignoring fire damage). d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data forest science • december significantly more grade loss in all species than all other burn classes (all p  < . ). however, for merchantable species, burn class + stands were only different from those receiving no or one prescribed fire (p = . and p = . , respectively). burn class + stands only differed from control stands for white oak (p = . ). the significance of the interaction term in both all species and merchantable species models was largely due to higher observed grade loss on xeric burn class + sites than on mesic burn class + sites (all species: xeric =  .  ±  . percent, mesic =  .  ±  . percent; merchantable species: xeric =  .  ±  . percent, mesic =  .  ±  . percent); all other burn classes did not differ greatly between aspects (< percent difference). for white oak, a difference between xeric and mesic burn class + sites was also observed, but the difference was not as pronounced (xeric =  .  ±  . percent, mesic =  .  ±  . percent). much of this difference was a result of one site where . percent of the white oak had grade loss. discussion most trees measured in this study either did not get damaged by prescribed fire or were able to compartmentalize (sensu shigo ) and heal wounds quickly. although we observed wounds that led to cull, these wounds were not common. most observed wounds were either seams or the generic group of bark sloughing/ multiple seams, with distinct woundwood ribs that touched and had closed over the wound (table ). wound closure helps protect the tree from pathogens, rot, and additional mechanical wounds figure . average percentage of trees per acre wounded within each burn class for (a) all species, (b) merchantable species, and (c) white oak only. burn class refers to the number of prescribed fires received over the past  years. bars indicate ± standard error; within each panel, and different letters indicate significantly different means at α =  . with the tukey honestly significant difference test. d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch forest science • december (shigo , dey and schweitzer ) and can happen as quickly as one year after a fire, depending on scar width (stambaugh et al. ). most measurable volume loss occurred with catfaces, basal/ flutes, and ovals, all of which are open and vulnerable to decay. since these wounds have measurable depth, we observed the highest average volume (board feet) in defect per wound for these wound types (table ). closed wounds, on the other hand, can easily hide deeper pockets of decay that may extend into the cant during lumber milling, so we suspect our estimates of volume loss in trees with these defects may be conservative; destructive sam- pling and lumber recovery studies would be needed to confirm our estimates. figure . relative total stand volume lost (rv t ; a, c, and e) and relative volume lost from the bottom -ft log (rv ; b, d, and f) by burn class and aspect for all species (panels a and b), merchantable species (c and d), and white oak only (e and f). burn class refers to the number of prescribed fires received over the past  years. bars indicate ± standard error; different letters indicate significantly different means within each subfigure at α =  . with the tukey honestly significant difference test. in anova models, the interaction between burn class and aspect was significant (p < . ) only for rv t of all and of merchantable species (see table s ). table . mean percentage (se) of trees per acre that changed grade because of prescribed fire damage for all species (f ,   =  . , p < . ), merchantable species (f ,  =  . , p = . ), and only white oak (f ,  =  . , p = . ). burn class all species merchantable species white oak . ( . ) a . ( . ) a . ( . ) a . ( . ) ab . ( . ) ab . ( . ) abc . ( . ) bc . ( . ) bc . ( . ) ab . ( . ) bc . ( . ) bc . ( . ) bc + . ( . ) c . ( . ) c . ( . ) c note: letters indicate significant differences (α =  . ) between burn classes within each species group using tukey’s honestly significant different tests. d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data forest science • december marschall et al. ( ), in a lumber recovery study, determined that wounds below . in. in height would have minimal impact on the value loss  years after the initial prescribed fire in stands that receive multiple fires. considering both that our observed wound heights exceeded this threshold, averaging about . in. in height (table ), and that residency time (length of time since a wound was inflicted) was likely longer than   years in some stands (see table s ), our average volume defect for all wound types remained relatively low, only .  ±  . bd ft (table ). therefore, individual wounds usually had minimal impact on stumpage value for a given tree; only when wounds were numerous and spread across mul- tiple faces was there a potential to significantly affect value through change in volume or grade. multiple wounds were not common in our study; only . percent of sample trees had three or more wounds, and usfs tree grade changed on only . percent. stand-level patterns overall, prescribed fire did not influence overstory structure at the hnf. ba, den, and qmd of the stands did not differ significantly among similar stands with increasing numbers of prescribed fire on any aspect. other studies in the missouri ozarks also found no impact of prescribed fire on ba, den, and qmd in oak–pine and oak stands (stevenson , knapp et al. ). fire behavior of prescribed fires in most eastern closed canopy, oak-dominated forests commonly follows anderson fuel model (anderson ); this behavior leads to little mortality (< – percent) for trees greater than in. dbh and declines as tree sizes (and bark thicknesses) increase (hutchinson et al. , kinkead et al. , dey and schweitzer ). therefore, much more in- tensive prescribed fires, likely with average flame heights exceeding ft, more typical of anderson fuel model- , and/or mechanical thinning may be required to reduce overstory ba significantly for woodland or barren restoration efforts should that be the goal of managers (brose et al. ). wounding we examined the impacts of increasing number of prescribed fires on wounding rather than treating prescribed fire as a binary variable (i.e., fire or no fire) as was done in most prior studies. we found that, whereas prescribed fire wounds trees, it does not wound all trees in a stand, an observation made in several earlier studies (kinkead et  al. , smith and sutherland , stevenson et  al. ). as hypothesized, the percentage of trees wounded by prescribed fire generally increased as the number of prescribed fires increased, but the shape of the response (i.e., linear versus a threshold) was not clear because of the high variability seen among the stands; only stands with three or more prescribed fires were statistically different from unburned, control stands. this trend was the same for all species combined, for all merchantable species, and white oak alone. therefore, an expansion of this study to more stands to refine these relations is likely warranted. notably, trees within unburned, control stands also demonstrated a fair amount of damage ( percent of all trees per acre), as damage can be caused by many environmental factors (e.g., frost, wildlife interactions, falling limbs), not just fire. in stands with three or more prescribed fires, the average per- centage of trees per acre wounded was percent. plots burned pe- riodically since in the missouri ozarks had similar wounding rates at . percent (knapp et  al. ). stevenson ( ) also found a similar response, but a higher proportion of trees wounded on xeric aspects ( percent) and mesic aspects ( percent) in stands with histories of one to nine prescribed fires in the missouri ozarks. most low- to moderate-intensity fires burn with variable severity, creating a mosaic of areas with no to high fuel consumption (pyne et al. ). as a result, many trees may escape injury in a single burn, although over multiple burns the likelihood of injury, and resulting damage, would increase. thick bark also reduces the likelihood of in- jury, by insulating the cambium from lethal fire temperatures, even in areas with higher fuel loads (hengst and dawson ). regardless, more trees may have internal damage than can be seen externally, as fire wounds, especially small ones, can compartmentalize and heal quickly after a fire (stambaugh et al. ). aspect did not heavily influence the proportion of trees wounded, even though it has been previously reported that more xeric, southern-facing aspects have higher rates of scarring from prescribed fire (stevenson et al. , kinkead et al. ). aspect influences fire behavior through altered solar radiation, air temper- ature, and fuel moisture (pyne et al. ); therefore, we expected a stronger relation between aspect and wounding. on these highly dissected sites, local topography in the immediate neighborhood of individual trees could have either dampened or increased local fire intensity and severity, thereby masking the influence of aspect on wounding at the stand level. other environmental factors that influence fire behavior, such as seasonality of fire and method of ig- nition, could have additionally confounded the influence of aspect (pyne et al. ). future studies should stratify on more of these environmental variables in stand selection and, potentially, analyze data at the plot or tree level so as to better understand how fire be- havior affects tree damage in eastern oak forests. relative volume lost our results were consistent with previous studies of prescribed fire’s effects on standing timber or lumber volume. as residence time, the length of time since the inception of a fire scar, increases, more rot and decay can enter the wood, leading to cull and larger defect volumes (stambaugh and guyette , dey and schweitzer ). in a lumber recovery study, marschall et al. ( ) reported a reduction of . percent in red oak log volume across all samples taken from stands with three to four prescribed fires over a period of   years; this reported volume falls between those losses that we measured in merchantable species for stands with three and with four or more burns (figure ). using data collected from . -ft butt logs of fire-damage trees on a single site, stambaugh and guyettte ( ) developed a volume-reduction model that predicted a loss of – percent after   years, depending on dbh and tree scar sizes. across all stands that had prescribed fire, rvtot was very low ( . percent), and only two stands had merchantable volume losses of over percent. unlike what we hypothesized, xeric aspects did not strongly increase prescribed fire damage, except on the most frequently burned sites (figure ). instead, there was a positive re- lation between damage and number of prescribed fires. in another study from the missouri ozarks, knapp et  al. ( ) reported much higher, stand-level volume losses, . percent, for trees that had received prescribed fire every four years since . however, most of the volume for these trees occurred in the -ft butt log; our hnf study trees were substantially taller and could produce d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch https://academic.oup.com/forestscience/article-lookup/doi/ . /forsci/fxz #supplementary-data forest science • december as many as three additional -ft logs. given that prescribed fire damage is exclusive to the butt log, this likely explains our lower relative volume loss. in addition, our study sites are considerably moister and of higher site quality, and generally have a lower fire intensity than the missouri ozarks. for example, an expansion of this study to the mark twain national forest has measured volume reductions of two- to threefold reported here (d. mann, unpubl. data), much more in agreement to the study by knapp et al. ( ). these fire-caused volume losses can be reduced with some atten- tion and care during harvesting and milling. trees can be optimally bucked (sessions , sessions et  al. ) to exclude excessively scarred areas, and log lengths can be shortened to produce higher- grade material; at the mill, boards can be cut with seams on the edges, allowing enough clear wood throughout the rest of the board to maintain the grade. furthermore, shallow fire scars from recent fires and scars on large trees are often removed in the slab during timber processing, leading to no realized volume deductions in the lumber (loomis , stambaugh and guyette , marschall et al. , wiedenbeck and schuler ). further lumber recovery studies are needed to relate observed fire-damage volume losses at harvest to the eventual economic value of the milled lumber. grade change prescribed fire did not greatly impact tree quality at the stand level, even in the cases where the residence time was  years and more than four prescribed fires have been applied. study-wide, less than percent, on average, of a burned site’s trees received enough damage to reduce the usfs tree grade. as hypothesized, the number of prescribed fires increased grade loss, but similar to wounding, the slope of the response was unclear because of high variability among sites. only in stands with four or more prescribed fires did grade reductions differ from controls, exceeding percent of overstory stems. white oak trees were particularly resistant to grade change, having grade reductions of approximately one-half the rate of all other species (table ). fire frequently wounds just one face of the tree, the leeward side of the tree (gutsell and johnson ). trees could frequently keep their grade because the process of grading trees ignores the face with the most defects and specifies that the grade is based on the best ft in the bottom -ft log (hanks ). in our study, most fire damage on trees was isolated to only one face and within the bottom -ft section of the bole—both portions that can be de- fective yet ignored while grading if other portions of the tree stem are of a higher quality. thus, grade only declined for trees that had excessive cull because of fire damage that exceeded the cull limits for each grade or where fire significantly damaged two or more faces of the tree. fire also degraded a tree if the tree otherwise had many defects (e.g., dead limbs) and the best face was that which received fire damage; the additional fire wounds shifted the best face to a more defect-laden face. our findings should not be extrapolated to veneer tree quality and value, since veneer grades are more stringent than the highest usfs grade requirements. past research has found minimal effects of prescribed fire on overstory tree quality for similar reasons. for example, wiedenbeck and schuler ( ) found that percent of trees had grade change or scale volume deduction after two prescribed fires, but per- cent of these trees were red maple, a known fire-intolerant species. the remaining trees that changed grade were more fire-tolerant spe- cies, percent of all trees, and similar to that found in our study ( percent). the only wound type to affect timber quality in the wiedenbeck and schuler ( ) study was large catfaces, of which we had very few (only instances of catfaces with a defect of > bd ft across , trees). however, stambaugh and guyette ( ) reported a much higher rate, approximately percent of oak trees (black, scarlet, and white oak), that changed grade because of prescribed fire damage after two fires in the missouri ozarks. conclusion in southern indiana, prescribed fire can potentially wound in- dividual trees and significantly reduce their timber volume and quality; in practice, however, losses are minimal when aggregated to the stand level. generally, the percentage of trees wounded increases with the number of prescribed fires a stand receives, but even then, not all trees in a stand are wounded. most overstory trees in mature, oak–hickory stands have relatively thick bark that insulates them from fire damage. for trees that are wounded, particularly oak spe- cies, the ability to compartmentalize and heal wounds minimizes fire’s impact on overstory tree volume and value, even after  years and more than four prescribed fires. nevertheless, continued use of prescribed fire over multiple decades would likely accumulate defects and lead to higher losses (knapp et al. ). for example, pole- and small sawtimber-sized trees are more susceptible to damage and decay, especially if left in the stand for long periods of time before harvest (dey and schweitzer , ). these results indicate that prescribed fire can be used in con- junction with other forest-management strategies to meet mul- tiple objectives (dey and schweitzer ). for example, if fire is used within a shelterwood regime and conducted less than  years before final overstory harvest, oak regeneration may be strongly promoted, whereas losses to timber resources can be minimized. multiple applications of fire are likely needed, and, once regen- eration is present, timing between burns should exceed   years (brose et  al. ). fire can also be helpful in combination with midstory removal of mesic species to allow more light to the forest floor and reduce the litter layer depth for successful oak regenera- tion (dey and schweitzer ), although placement of slash from the midstory removal near boles of trees can be problematic by increasing local fire intensity and causing more fire damage (brose and van lear ). regardless, further research is still needed to weigh the economic benefits of promoting oak regeneration against the economic damage to residual timber across a broad range of forest types, sites, and differing fire prescriptions. supplementary materials supplementary data are available at forest science online. literature cited abrams,  m.d. . where has all the white oak gone? bioscience ( ): – . anderson,  h.e. . aids to determining fuel models for estimating fire behavior. usda forest service gen. tech. rep. gtr-int- , intermountain forest and range, experiment station, ogden, ut. p. bowman, d.m.j.s., j. balch, p. artaxo, w.j. bond, m.a. cochrane, c.m. d’antonio, r. defries, et al. . the human dimension of fire regimes on earth. j. biogeogr. ( ): – . d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch forest science • december bradford,  j.b., and a.w.  d’amato. . recognizing trade-offs in multi-objective land management. front. ecol. environ. ( ): – . brose, p.h., and d.h. van lear. . effects of seasonal prescribed fires on residual overstory trees in oak-dominant shelterwood stands. south. j. appl. for. ( ): – . brose, p.h., d.h. van lear, and r. cooper. . using shelterwood harvests and prescribed fire to regenerate oak stands on productive up- land sites. forest ecol. manage. ( – ): – . brose, p.h., d.c. dey, r.j. phillips, and t.a. waldrop. . a meta- analysis of the fire-oak hypothesis: does prescribed burning promote oak reproduction in eastern north america? for. sci. ( ): – . brose,  p.h., d.c.  dey, and t.a.  waldrop. . the fire-oak litera- ture of eastern north america: synthesis and guidelines. usda forest service gen. tech. rep. gtr-nrs- , northern research station, newtown square, pa. p. buehlmann,  u., m.  bumgardner, and d.  alderman. . recent developments in us hardwood lumber markets and linkages to housing construction. curr. for. rep. ( ): – . carmean,  w.h., j.t.  hahn, and r.d.  jacobs. . site index curves for forest species in the eastern united states. usda forest service gen. tech. rep. gtr-nc- , north central forest experiment station, st. paul, mn. p. delany,  e., and m.  haynes. . industry on tap: breweries. usdc bureau of labor statistics report, spotlight on statistics (december ), philadelphia, pa. p. available online at www.bls.gov/ spotlight/ /industry-on-tap-breweries/home.htm; last accessed august , . de mendiburu, f., and r. simon. . agricolae—ten years of an open source statistical tool for experiments in breeding, agriculture and bi- ology. peerj preprints :e v . dey,  d.c., and c.j.  schweitzer. . timing fire to minimize damage in managing oak ecosystems. p. – in proceedings of the th biennial southern silvicultural research conference, holley, a.g., k.f. conner, and j.d. haywood (eds.). usda forest service gen. tech. rep. egtr-srs- , southern research station, asheville, nc. dey,  d.c., and c.j.  schweitzer. . a review on the dynamics of prescribed fire, tree mortality and injury in managing oak natural communities to minimize economic loss in north america. forests ( ): . fox, j., and s. weisburg. . an r companion to applied regression. nd ed. sage publishing, thousand oaks, ca. p. gormanson,  d.d., and c.m.  kurtz. . forests of indiana. usda forest service resource update fs- , northern research station, newtown square, pa. p. gutsell,  s.l., and e.a.  johnson. . how fire scars are formed: coupling a disturbance process to its ecological effect. can. j. for. res. ( ): – . guyette, r.p., j. marschall, m. stambaugh, and a. stevenson. . validating time dependent tree defect models and determining lumber yield changed for prescribed fire injured oak trees. final report mcc- - - . missouri department of conservation. p. hanks, l.f. . hardwood tree grades for factory lumber. usda forest service res. paper rp-ne- , northeastern forest experiment station, upper darby, pa. p. hare, r.c. . contribution of bark to fire resistance of southern trees. j. for. ( ): – . harmon, m.e. . survival of trees after low-intensity surface fires in the great smoky mountains national park. ecology ( ): – . hengst,  g.e., and j.o.  dawson. . bark properties and fire resist- ance of selected tree species from the central hardwood region of north america. can. j. for. res. ( ): – . homoya, m.a., d.b. abrell, j.r. aldrich, and t.w. post. . the natural regions of indiana. proc. indiana acad. sci. : – . hutchinson,  t.f., e.k.  sutherland, and d.a.  yaussy. . effects of repeated prescribed fires on the structure, composition, and regeneration of mixed-oak forests in ohio. forest ecol. manage. ( – ): – . kinkead,  c.s., m.c.  stambaugh, and j.m.  kabrick. . mortality, scarring, and growth in an oak woodland following prescribed fire and commercial thinning in the ozark highlands. forest ecol. manage. ( ): – . knapp,  b.o., j.m.  marschall, and m.c.  stambaugh. . effects of long-term prescribed burning on timber value in hardwood forests of the missouri ozarks. p. – in proceedings of the th central hardwood forest conference, kabrick,  j.m., d.c.  dey, b.o.  knapp, d.r.  larsen, s.r.  shifley, and h.e.  stelzer (eds.). usda forest service gen. tech. rep. gtr-nrs-p- , northern research station, newtown square, pa. loomis, r.m. . predicting the losses in sawtimber volume and quality from fires in oak–hickory forests. usda forest service res. paper rp-nc- , north central forest experiment station, st. paul, mn. p. marschall, j.m., r.p. guyette, m.c. stambaugh, and a.p. stevenson. . fire damage effects on red oak timber product value. forest ecol. manage. : – . maxwell,  j.t., and g.l.  harley. . dendroclimatic reconstructions from multiple co-occurring species: a  case study from an old-growth deciduous forest in indiana, usa. int. j. climatol. ( ): – . miller, g.w., l.f. hanks, and h.v. wiant jr. . a key for the forest service hardwood tree grades. north. j. appl. for. ( ): – . nowacki,  g., and m.d.  abrams. . the demise of fire and “mesophication” of forests in the eastern united states. bioscience ( ): – . pyne, s.j., p.l. andrews, and r.d. laven. . introduction to wildland fire. john wiley and sons, new york. p. r core team. . r: a  language and environment for statistical com- puting. r foundation for statistical computing, vienna, austria. available online at www.r-project.org; last accessed august , . schnur,  g.l. . yield, stand, and volume tables for even-aged upland oak forests. usda tech. bull. . usda, washington, dc. p. sessions, j. . making better tree-bucking decisions in the woods. j. for. ( ): – . sessions, j., e. olsen, and j. garland. . tree bucking for optimal stand value with log allocation constraints. for. sci. ( ): – . shigo,  a.l. . compartmentalization: a  conceptual framework for understanding how trees grow and defend themselves. annu. rev. phytopathol. : – . smith, k.t., and e.k. sutherland. . fire-scar formation and com- partmentalization in oak. can. j. for. res. ( ): – . smith,  k.t., and e.k.  sutherland. . terminology and bi- ology of fire scars in selected central hardwoods. tree ring res. ( ): – . stambaugh,  m., and r.  guyette. . chapter : prescribed fire effects on the wood quality of three common oaks in the ozark region, q.  coccinea, q.  velutina, q.  alba. p. in prescribed fire effects on the wood quality of oak (quercus sp.) and shortleaf pine (pinus echinata), guyette,  r., m.  stambaugh, a.  stevenson, and r.m.  muzika (eds.). final report. missouri department of conservation, west plains, mo. available online at http://faculty.missouri.edu/~stambaughm/mtrl/ timberquality.html; last accessed may , . stambaugh,  m.c., k.t.  smith, and d.c.  dey. . fire scar growth and closure rates in white oak (quercus alba) and the implications for prescribed burning. for. ecol. manage. : – . steiner, k.c., b.s. stein, and j.c. finley. . a test of the delayed oak dominance hypothesis at mid rotation in developing upland stands. for. ecol. manage. : – . d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch http://www.bls.gov/spotlight/ /industry-on-tap-breweries/home.htm http://www.bls.gov/spotlight/ /industry-on-tap-breweries/home.htm http://www.r-project.org http://faculty.missouri.edu/~stambaughm/mtrl/timberquality.html http://faculty.missouri.edu/~stambaughm/mtrl/timberquality.html forest science • december stevenson, a.p. . effects of prescribed burning in missouri ozark up- land forests. m.sc. thesis, university of missouri-columbia, columbia, mo. p. stevenson, a.p., r. muzika, and r.p. guyette. . fire scars and tree vigor following prescribed fires in missouri ozark upland forests. p. – in proceedings, th central hardwood forest conference, jacobs, d.f., and c.h.  michler (eds.). usda forest service gen. tech. rep. gtr- nrs-p- , northern research station, newtown square, pa. sutherland,  e.k., and k.t.  smith. . resistance is not futile: the response of hardwoods to fire-caused wounding. p. – in proceedings: workshop on fire, people, and central hardwoods landscape, d.a.  yaussy (ed.). usda forest service gen. tech. rep. gtr-ne- , northeastern research station, newtown square, pa. wiant,  h.v. . formulas for mesavage and girard’s volume tables. north. j. appl. for. : . wiedenbeck, j.k., and t.m. schuler. . effects of prescribed fire on the wood quality and marketability of four hardwood species in the central appalachian region. p. – in th central hardwood forest conference proceedings, groninger,  j.w., e.j.  holzmueller, c.k.  nielsen, and d.c.  dey (eds.). usda forest service gen. tech. rep. gtr-nrs-p- , northern research station, newtown square, pa. d ow nloaded from https://academ ic.oup.com /forestscience/article-abstract/ / / / by guest on m arch wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ vashistha et al. bmc cancer , : http://www.biomedcentral.com/ - / / research article open access current and recent clinical trials for perioperative systemic therapy for muscle invasive bladder cancer: a systematic review vishal vashistha , david i quinn , tanya b dorff and siamak daneshmand * abstract background: although muscle invasive bladder cancer (mibc) is increasing in incidence, treatment has largely remained limited to radical cystectomy with or without cisplatin-based neoadjuvant and/or adjuvant chemotherapy. we reviewed the current and recent clinical trials evaluating perioperative chemotherapy, targeted therapy, and novel therapeutic regimens for mibc patients undergoing radical cystectomy. methods: an overview of perioperative mibc management was conducted initially using medline. the clinical trials registry and medline were further searched specifically for perioperative mibc chemotherapy, targeted therapy, and other novel therapeutic approaches. trials involving non-perioperative management, operative management other than radical cystectomy, multiple tumors, or purely superficial or metastatic disease were excluded from selection. these criteria were not specifically fulfilled for mtor inhibitor and immune therapy trials. only phase iii chemotherapy and phase ii targeted therapy trials found in the clinical trials registry were selected. medline searches of specific treatments were limited to january to january whereas the clinical trials registry search had no timeline. systematic medline searches had no phase restrictions. trials known by the authors to fulfill search criteria but were not found via searches were also selected. results: twenty-five trials were selected from the clinical trials registry including phase iii chemotherapy trials, phase ii targeted therapy trials, immune therapy trials, mammalian target of rapamycin (mtor) inhibitor trial, and gene and vaccine therapy trials. nine trials have been completed and have been terminated early or withdrawn. nine trials have data available when individually searched using medline and/or google. systematic searches of medline separately found trials in the past years. two phase iii chemotherapy trials were selected based on knowledge by the authors. no phase iii trials of targeted therapy have been registered or published. conclusions: new trials are currently being conducted that may revolutionize mibc treatment preceding or following cystectomy. head-to-head phase iii trials of perioperative chemotherapy and further phase ii and phase iii trials of targeted therapy and other therapeutic approaches are necessary before the current cisplatin-based perioperative chemotherapy paradigm is altered. keywords: bladder cancer, radical cystectomy, perioperative chemotherapy, perioperative targeted therapy, mtor inhibitor, immune therapy, gene therapy, vaccine therapy * correspondence: daneshma@med.usc.edu department of urology, usc/norris comprehensive cancer center, usc institute of urology, eastlake abe, suite , los angeles, ca , usa full list of author information is available at the end of the article © vashistha et al.; licensee biomed central. this is an open access article distributed under the terms of the creative commons attribution license (http://creativecommons.org/licenses/by/ . ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. the creative commons public domain dedication waiver (http://creativecommons.org/publicdomain/zero/ . /) applies to the data made available in this article, unless otherwise stated. mailto:daneshma@med.usc.edu http://creativecommons.org/licenses/by/ . http://creativecommons.org/publicdomain/zero/ . / vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / background bladder cancer is the th most common cancer in men and th most common cancer in women in the united states [ ]. in total, , new cases are diagnosed each year with , deaths annually [ ]. though rarer worldwide compared to other cancers, bladder cancer is also increasing in incidence globally as the use of tobacco products becomes more prevalent in developing nations [ ]. in the united states, about % of these tumors invade past both the bladder submucosa and mucosa and are therefore defined as muscle invasive bladder cancer (mibc) [ ]. though mibc is prevalent both in the united states and internationally, treatment options for mibc have remained essentially unchanged for the past years. traditionally, radical cystectomy has been the mainstay, but as evidence accumulates for the benefit of periopera- tive therapy, neoadjuvant or adjuvant cisplatin-based chemotherapy have become more valid options for mibc patients [ - ]. despite this success and that of other tri- als, only - % of patients will receive neoadjuvant chemotherapy, though its prevalence has been increasing over time [ ]. the number of medical oncologists recom- mending perioperative chemotherapy to their patients has increased as almost % of oncologists recently sur- veyed have suggested perioperative chemotherapy to their mibc patients, albeit those physicians were well-versed in bladder cancer treatment while many patients fail to have access to such medical oncologists for socieoeconomic reasons [ ]. to continue the increasing trend of utilizing perioperative chemotherapy, clinical studies should ad- dress the overarching issues of predicting which patients will more likely benefit from chemotherapy, identifying particular chemotherapy regimens for specific patient subsets, and developing more efficacious non-cisplatin based regimens for patients who are cisplatin-ineligible. in addition, there is a desperate need for continued explor- ation of novel therapeutic treatment options to help modernize the perioperative management of mibc. as of now, perioperative mibc clinical research is mainly focused on selecting a more efficacious cisplatin-based regi- men using head-to-head trials with a limited number of studies addressing regimens for cisplatin-ineligible patients. novel therapeutic approaches including targeted therapy, mammalian target of rapamycin (mtor) inhibitors, im- mune therapy, and gene and vaccine therapy are now being evaluated in early phase trials with cautious optimism. understanding the rationale and outcomes of these current and recent trials is imperative for clinicians and investiga- tors to continue to encourage patient participation in these research efforts and to design future studies that enhance our ability to offer personalized treatment for mibc pa- tients. the following is a systematic review of the current and recent perioperative clinical trials conducted worldwide in the management of mibc for patients undergoing radical cystectomy with an evaluation of specific areas that could benefit from future trials. methods data sources two separate databases were used to explore current and recent clinical trials for the perioperative management of mibc. initially, a medical literature analysis and retrieval system online (medline) search through pubmed was completed for a general overview of the lit- erature. this was followed by a search of all perioperative trials for mibc using the clinical trials registry online with the search phrase of “bladder cancer”. all trials se- lected for further review from the clinical trials registry were independently searched using medline and google for published results. the initial overview of medline abstracts and the clinical trials registry search were used to develop individual search equations based on the different perioperative treatment classes for mibc. subse- quently, medline was systematically searched for peri- operative trials using these search equations. all search equations for medline and the clinical trials registry are listed in table . lastly, any upcoming trials that are currently unregistered with the clinical trials registry but were known by the authors to have been presented at any major urology and/or oncology conference were included. no new human subject data was studied requiring approval from an institution’s ethics board. study selection trials involving non-perioperative management, periopera- tive treatment for multiple tumors in addition to bladder cancer, operative management other than radical cystec- tomy, unconfirmed operative management, measured out- comes solely of serum or urine biomarker concentration, or purely metastatic or non-invasive disease were excluded from selection. the exclusion criteria of “unconfirmed operative management” was not strictly met for trials obtained from the clinical trials registry involving rapa- mycin (mtor) inhibition and immune therapy due to the current limited number of trials in those fields. clinical trials of all phases were reviewed but only current and recent phase iii trials of perioperative chemo- therapy and phase ii trials of perioperative targeted therapy found in the clinical trials registry were selected and further discussed. trials of all phases from other peri- operative treatment options were selected and evaluated. published medline clinical trials selected from the sys- tematic review of each treatment class were limited to data published between january and january . no phase restrictions were placed on these medline systematic searches. published medline trial data ob- tained from independently searching individual trials table current and recent perioperative clinical trial systematic searches for mibc* search category database search equation used trials selected/reviewed broad overview of all published abstracts medline (“bladder cancer” or “urothelial cancer” or “urothelial cell carcinoma” or “transitional cell carcinoma”) and (“bladder resection” or “radical cystectomy”) and (“perioperative chemotherapy” or “adjuvant chemotherapy” or “neoadjuvant chemotherapy” or “targeted therapy” or “biologic” or “immunotherapy” or “gene” or “vaccine” ) abstracts reviewed^ all registered clinical trials with the nih clinical trials registry bladder cancer / # trials of neoadjuvant chemotherapy medline ((neoadjuvant) and (chemotherapy or cisplatin or gemcitabine or carboplatin or methotrexate or vinblastine or doxorubicin or adriamycin or paclitaxel or ifosfamide or abraxane)) and (bladder cancer) and (radical cystectomy) / trials of adjuvant chemotherapy medline ((adjuvant) and (chemotherapy or cisplatin or gemcitabine or carboplatin or methotrexate or vinblastine or doxorubicin or adriamycin or paclitaxel or ifosfamide or abraxane)) and (bladder cancer) and (radical cystectomy) / trials of egfr inhibitors medline (egfr inhibitor or cetuximab or erlotinib or gefitinib or genistein or her- inhibitor or lapatinib or mgah or panitumumab or trastuzumab) and (bladder cancer or radical cystectomy) / trials of vegf inhibitors medline (vegf inhibitor or bevacizumab or ramucirumab or trebananib or ziv aflibercept) and (bladder cancer or radical cystectomy) / trials of other tyrosine kinase inhibitors medline (tyrosine kinase inhibitor or dasatinib or pazopanib or sorafenib or sunitinib) and (bladder cancer or radical cystectomy) / trials of mtor inhibitors medline (mtor inhibitor or everolimus or sirolimus or temsirolimus or rapamycin) and (bladder cancer or radical cystectomy) / trials of immune therapy medline (immune therapy or interferon-alpha) and (bladder cancer or radical cystectomy) and (phase) / trials of gene and vaccine therapy medline (vaccine or gene therapy) and (bladder cancer or radical cystectomy) and (phase) / nih = national institute of health. medline = medical literature analysis and retrieval system online. *medline clinical trial abstracts were only reviewed dating from january to january . trials from the clinical trials registry online and the medline broad overview of the literature had no date restrictions. ^no trials were individually selected from the medline broad overview of the literature. treatment classes and specific drugs were taken from this medline search and from the trials search of the clinical trials registry to develop the systematic search equations for each individual treatment class. #this result includes phase iii perioperative chemotherapy trials, phase ii targeted therapy trials, and clinical trials of all phases for mtor inhibitors, immune therapy, and gene and vaccine therapy. vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / found on the clinical trials registry was reported regard- less of publication date. results the current treatment paradigm for mibc management is conveyed in figure . outcomes from meta-analyses of previous phase iii perioperative chemotherapy trials are reported in table . the current and recent perioperative clinical trials for mibc are identified below based on drug class. the classes include perioperative chemotherapy (table ), targeted therapy (table ), mtor inhibition (table ), immune therapy (table ), gene therapy (table ), and vaccine therapy (table ). perioperative chemotherapy trials the initial medline review of abstracts identified a meta-analysis for neoadjuvant chemotherapy of phase iii trials, an updated meta-analysis for adju- vant chemotherapy of phase iii trials, and a systematic re- view of neoadjuvant and adjuvant chemotherapy trials prior to september [ , , ]. a neoadjuvant meta-analysis by winquist et al., the original adju- vant meta-analysis by the advanced bladder cancer meta-analysis corporation prior to the update, and a neoadjuvant and adjuvant meta-analysis abstract by tjokrowidjaja et al. were also reviewed but were not results of the original medline search [ , , ]. the results of the neoadjuvant and adjuvant meta-analyses are described in table . individual current and recent phase iii neoadjuvant and adjuvant chemotherapy trials are de- tailed in table . the clinical trials registry search found a total of neoadjuvant chemotherapy trials. these included phase iii trial, phase ii trials, and phase trial. the phase iii trial is a randomized control trial of high dose methotrexate, viblastine, doxorubicin, and cisplatin mvac (hd-mvac) compared to gemcitabine and cisplatin (gc) at the university hospital, rouen. this study is also accepting mibc patients for adjuvant consideration. an additional phase iii trial investigated by the southwest oncology group has recently been registered involving a comparison of dose dense mvac (dd-mvac) and gc, figure flow-chart of current management paradigm for patients with mibc. rct = randomized controlled trial. vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / though this was not found in our search [ ]. neither phase iii trial has available results. the systematic review of medline of neoadjuvant chemotherapy identified eligible trials out of results. two of these publications referred to one single-arm study of gemcitabine with carboplatin that resulted in safe, tol- erable uptake of this regimen with a comparable efficacy to cisplatin-based regimens [ , ]. two other trials were subset studies of previous large phase iii trials and are not described in table [ , ]. the review also identified a phase ii neoadjuvant erlotnib study that is discussed further below as a targeted therapy [ ]. lastly, a recently published phase iii neoadjuvant mvac trial by the japan clinical oncology group, which was more recent than our search timeline criteria, reported an improved overall survival (os) with the neoadjuvant arm but was not statis- tically significant and is included in table [ ]. a total of adjuvant chemotherapy trials were found in the clinical trials registry. this included adjuvant-only phase iii trials, aforementioned neoadjuvant and adju- vant trial from the university hospital, rouen, and phase ii trials. one of these adjuvant-only phase iii trials, a national cancer institute trial of adjuvant mvac with gemcitabine compared to progression-triggered mvac with gemcitabine, was withdrawn prior to enroll- ment. a google and medline search for each of the table meta-analyses of perioperative chemotherapy phase iii trials for mibc study treatment type year number of trials patients therapy os (hr) dfs (hr) tjokrowidjaja et al. [ ] neoadjuvant + adjuvant only abstract available . . ( . – . ) ( . – . ) tjokrowidjaja et al. [ ] neoadjuvant only abstract available . . ( . – . ) ( . – . ) advanced bladder cancer- meta-analysis corporation [ ] neoadjuvant platinum-based . . ( . – . ) ( . – . ) winquist et al. [ ] neoadjuvant ^ cisplatin-based . insufficient data for analysis+ ( . – . ) tjokrowidjaja et al. [ ] adjuvant only abstract available . . ( . – . ) ( . – . ) leow et al. [ ] (update of the meta-analysis below) adjuvant # cisplatin-based . . ( . – . ) ( . – . ) advanced bladder cancer meta-analysis corporation [ ] adjuvant * cisplatin-based . . ( . – . ) ( . – . ) os: overall survival. dfs: disease free survival. hr: hazard ratio. ^ trials had hr data for os and only trials had hr data for dfs. +less than half the number of patients assessed were available for dfs results, and therefore, no statistical pooling analysis was undertaken. # trials had hr data for os and only trials had hr data for dfs. * trials had hr data for os and only trials had hr data for dfs. table current and recent perioperative chemotherapy phase iii trials for mibc institution regimen setting status/results trial id university hospital, rouen hd-mvac vs. gc neoadjuvant or adjuvant recruiting participants. nct southwest oncology group dd-mvac vs. gc neoadjuvant recruiting participants [ ]. japan clinical oncology group mvac vs. immediate rc neoadjuvant completed. total subjects. os increased by % in mvac arm ( subjects) but was not statistically significant. improved pt rate with mvac [ ]. ftrc gc vs. immediate rc adjuvant completed. total subjects. no difference between adjuvant treatment ( ) and control arms for year os. failed to reach accrual goal, low study power [ ]. nct cairo university gc + radiotherapy vs. radiotherapy alone adjuvant completed. total subjects. improvement in -month dfs from % to % in chemoradiotheraphy ( ) group but failed to be statistically significant [ ]. nct eastern cooperative oncology group mvac vs. pca adjuvant completed. results unreported. nct sun yat-sen university intraarterial gc adjuvant recruiting participants. nct association of urogenital oncology adjuvant vs. progression-triggered gemcitabine adjuvant completed. total subjects. no statistically significant difference between adjuvant or progression-triggered gemcitabine for dfs or os. failed to reach accrual goal [ ]. nct eortc adjuvant vs. progression triggered mvac + gemcitabine adjuvant closed early due to poor accrual nct ftrc = fondazione c.n.r./regione toscana “g. monasterio”, pisa, italy. eortc = european organisation of research and treatment of cancer. gc = gemcitabine + cisplatin. mvac = methotrexate + vinblastine + doxorubicin + cisplatin. rc = radical cystectomy. pca = paclitaxel + carboplatin. hd = high dose. dd = dose dense. os = overall survival. dfs = disease free survival. vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / table current and recent perioperative targeted therapy phase ii trials for mibc institution regimen receptor target setting results/status trial id university of north carolina erlotinib egfr neoadjuvant completed. well tolerated. % ( / ) histologic downstaging preoperatively [ ]. university of north carolina erlotinib egfr neoadjuvant + adjuvant ongoing nct mdacc erlotinib egfr neoadjuvant ongoing nct dendreon dn - her r adjuvant ongoing. updated results: % of patients had her expression in primary tumor, % in lymph nodes. apc activation for all patients who have received infusions of drug [ ]. nct medical university of south carolina bevacizumab + gc and bevacizumab + paclitaxel# vegfr-a neoadjuvant + adjuvant ongoing. preliminary results: % ( / ) postoperative complication rate. % ( / ) downstaging preoperatively [ ]. nct mdacc bevacizumab + mvac vegfr-a neoadjuvant ongoing nct ftrc sorafenib + gc vegfr + pdgfr* neoadjuvant unknown. nct us oncology sunitinib + gc vegfr + pdgfr* neoadjuvant terminated early due to patient toxicity. only mibc patients studied for neoadjuvant therapy [ ]. mskcc sunitinib + gc vegfr + pdgfr* neoadjuvant completed early due to limited study accrual. pt rate was low with combination [ ]. nct cccc sunitinib vegfr + pdgfr* neoadjuvant completed, results unreported nct hoosier oncology group dasatinib bcr/abl^ neoadjuvant completed. well tolerated. follow-up pathology study: sfk expression downregulated in % ( / ) of patients [ , ]. nct university of michigan sunitinib vegfr + pdgfr* adjuvant terminated due to poor accrual nct hoosier oncology group sunitinib + gc vegfr + pdgfr* neoadjuvant terminated due to patient toxicities nct mdacc = m.d. anderson cancer center. ftrc = fondazione c.n.r./regione toscana “g. monasterio”, pisa, italy. mskcc = memorial sloan-kettering cancer center. cccc = case comprehensive cancer center. gc = gemcitabine and cisplain. mvac = methotrexate, vinblastine, doxorubicin, cisplatin. sfk = src family kinase. apc: antigen presenting cell. *sorafenib and sunitinib target multiple receptor tyrosine kinases (rtks) in addition to the aforementioned receptors. ^dasatinib targets multiple tyrosine kinases. #this trial includes neoadjuvant bevacizumab with gc followed by surgery and adjuvant bevacizumab and paclitaxel. vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / remaining trials found results for of these studies. cognetti et al. at fondazione c.n.r./regione toscana evaluated adjuvant gc vs. cystectomy alone and found no difference in os, but the study failed to achieve its accrual goal [ ]. a phase iii trial at the cairo university by table current mtor inhibitor and immune therapy trials fo institution regimen setting tr uthscsa sirolimus neoadjuvant ph tis hoosier oncology group everolimus +/− paclitaxel unspecified^ ph no university of washington sirolimus + gc unspecified^ ph pa mdacc inf-alpha prior to biopsy ph tre gc = gemcitabine + cisplatin. uthscsa = the university of texas health science ce turbt = transurethral resection of bladder tumor. *table does not include trials for multiple solid tumors that may include bladder ca ^unspecified regimens can be given in the neoadjuvant, adjuvant, locally recurrent zaghloul et al. of gc with radiotherapy did show an increase in disease free survival (dfs) but failed to reach statistical significance [ ]. the association of urogenital oncology’s phase iii trial of adjuvant gemcitabine com- pared to disease progression-triggered gemcitabine for r mibc* ial description status trial id ase trial for preoperative treatment. sue evaluated before and after treatment. recruiting nct ase ii trial for cisplatin ineligible patients. t a perioperative specific study. recruiting nct ase i/ii trial for disease of any stage. tients likely will undergo cystectomy. ongoing nct ase trial with turbt conducted following atment. radical cystectomy may follow. ongoing nct nter at san antonio. mdacc = m.d. anderson cancer center. ncer as a subset. , or metastatic setting. table current and recent perioperative gene and vaccine therapy trials for mibc* institution therapy trial description status/outcomes trial id uppsala university gene phase /iia intravesical instillation of adcd l that may improve anti-tumor immune response. cystectomy followed phase trial. completed. patients with limited complications. successful gene transfer detected in biopsies [ ]. nct celldex therapeutics vaccine phase ii neoadjuvant and adjuvant administration of cdx- vaccine with chemotherapy for bladder tumors expressing b-hcg protein for patients undergoing radical cystectomy. terminated due to lack of enrollment. nct mskcc and nci vaccine phase adjuvant administration of ny-eso- peptide vaccine with bcg + sargramostim for tumors expressing ny-eso- or lage- antigens. patients previously underwent radical cystectomy. competed. results unreported. nct *table does not include trials for multiple solid tumors that may include bladder cancer as a subset. mskcc = memorial sloan-kettering cancer center. nci = national cancer institute. vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / cisplatin-ineligible patients found no statistically signifi- cant outcomes [ ]. the systematic review of medline of adjuvant chemo- therapy identified eligible studies out of results. one study is the aforementioned phase iii trial conducted by cognetti et al. and is included in the updated meta- analysis by leow et al. [ ]. one phase ii trial showed benefit using adjuvant gc or mvac compared to obser- vation alone following cystectomy [ ]. three selected tri- als were neoadjuvant studies including small single-arm study evaluating the efficacy and tolerability of a split- dosed cisplatin regimen for patients with renal impair- ment that was not found with the neoadjuvant chemo- therapy trials systematic review [ - ]. perioperative targeted therapy trials a total of phase ii targeted therapy trials were identi- fied using the clinical trials registry as detailed in table . no phase iii trials met study eligibility criteria. seven were neoadjuvant trials, were adjuvant trials, and were both neoadjuvant and adjuvant trials. of these trials, adju- vant trial of sunitinib by the university of michigan was terminated due to poor accrual and neoadjuvant trial of sunitinib with gc by the hoosier oncology group was terminated due to patient toxicities. five trials included combinations with chemotherapy with trial being the neoadjuvant sunitinib trial that was terminated. a google and medline search of each clinical trial found results for studies. preliminary results of a neoadjuvant trial of bevacizumab and gc followed by adjuvant bevacizumab and paclitaxel at the medical university of south carolina have shown a high surgical complication rate though further results are pending [ ]. a neoadjuvant trial of sunitinib and gc at memorial-sloan kettering cancer center failed to show a significant rate of pt disease at surgery [ ]. a trial of neoadjuvant dastanib by the hoosier oncology group was well tolerated and further showed a marked decrease in tumor phosphory- lated src family kinase (sfk) expression levels but failed to show changes in cell proliferation [ , ]. the independent medline search of epidermal growth factor receptor (egfr) inhibitors found eligible study out of results. this was a published phase ii study in- volving neoadjuvant erlotinib at the university of north carolina conveying erlotinib to be well-tolerated with a substantial pre-surgical downstaging rate [ ]. our medline search of vascular epithelial growth factor (vegf) inhibitors provided eligible trials out of results. our search of other possible tyrosine kinase inhibitors provided eligible trial out of results. this was a neoad- juvant sunitinib with gc study published by us oncology that was terminated early due to patient toxicities [ ]. perioperative mtor inhibitor trials a total of trials were selected from the clinical trials registry involving mtor inhibitors. these trials included a phase neoadjuvant sirolimus trial by the university of texas health science center at san antonio, a phase ii everolimus with paclitaxel and carboplatin (pca) trial by the hoosier oncology group, and a phase i/ii sirolimus with gc trial by the university of washington. the phase ii trial by the hoosier oncology group is not a periopera- tive specific study. the phase i/ii trial by the university of washington includes patients that are likely to undergo cystectomy but will be further evaluated for surgery based on clinical judgment. the university of washington phase i/ii trial also involves non-invasive and metastatic disease. our medline search found eligible trials out of abstracts. perioperative immune therapy trials one immune therapy trial was selected from the clinical trials registry as described in table . this was a phase i trial of interferon-alpha being administered prior to vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / transurethral resection of bladder tumor for patients who are potential candidates for radical cystectomy at the m.d. anderson cancer center. the trial was not a perioperative specific study. the medline search found eligible trial out of ab- stracts. this study was also found in the gene and vaccine therapy systematic review and is further discussed below and included in table . perioperative gene and vaccine therapy trials phase i/ii gene therapy trial was found using the clinical trials registry. this was a completed study of the adcd l vaccine by uppsala university, which was also found in the perioperative immune therapy medline search. phase i vaccine therapy trial and phase ii vac- cine therapy trial were found on the clinical trials registry. the phase trial sponsored by memorial sloan kettering cancer center and the national cancer institute involving the adjuvant administration of ny-eso- following radical cystectomy has been completed but the results remain unreported. the phase ii trial sponsored by celldex thera- peutics involving the neoadjuvant administration of a vaccine for beta-human chorionic gonadotropin (bhcg) expressing mibc tumors was terminated due to a lack of enrollment. the gene therapy and vaccine therapy trials are detailed in table . our medline search for both gene therapy and vaccine therapy found eligible trial out of results. this was the published results of the adcd l immunogene adminis- tered as intravesical therapy prior to cystectomy conveying tolerable uptake of the treatment [ ]. discussion thus far, the perioperative medical management of mibc has largely been limited to platinum-based chemotherapy regimens, particularly involving cisplatin as described in figure . our systematic review identifies the current clinical trials evaluating perioperative chemotherapy regimens, targeted therapy, and other novel treatment options. below, we address the benefits and limitations of these trials as well as highlight the need for further particular studies. neoadjuvant chemotherapy trials though neoadjuvant chemotherapy has been proven beneficial, further research is necessary towards selecting a particular regimen based on patient demographics and tumor specifics particularly for patients who are cisplatin- ineligible. the meta-analysis conducted by the advanced bladder cancer meta-analysis collaboration found a % risk reduction in mortality or a % increase in -year os with the use of neoadjuvant chemotherapy prior to radical cystectomy [ ]. over % of patients re- ceived cisplatin-based regimens [ ]. clinicians often substitute regularly-dosed cisplatin for split-dosed cis- platin or carboplatin for patients with medical comorbidi- ties, mainly renal impairment. on the contrary, both the european association of urology (eau) and the society of urologic oncology recommend proceeding directly to radical cystectomy if patients cannot tolerate cisplatin- based treatment [ , ]. winquist et al. similarly found a % risk reduction in a meta-analysis of trials with a % risk reduction or a . % increase in os in the trials for patients undergoing a combination chemother- apy regimen [ ]. all trials were cisplatin-based [ ]. lastly, tjokrowidjaja et al. found a % risk reduction in the most recent meta-analysis, though only the ab- stract data is available and details of the patient population and trials used have yet to be published [ ]. interest- ingly, no direct comparisons between neoadjuvant chemo- therapy regimens have been completed in a phase iii trial. our review identified new phase iii trials including hd-mvac compared to gc at the university hospital, rouen and dd-mvac compared to gc by the southwest oncology group and recently published phase iii trial of mvac by the japan clinical oncology. the university hospital, rouen trial is both a neoadjuvant and adjuvant study with a goal of total patients. the southwest oncology group study is referred to as the co-expression extrapolation (coxen) trial because tumor messenger rna, dna, and stem cell biomarkers will be evaluated prior to administering mvac or gc [ ]. choi et al. of the southwest oncology group has previously shown that mibc with protein properties to that of p -mutated breast cancers have increased chemoresistance to cisplatin [ ]. additionally, other genomic characteristics of mibc may help clinicians select an appropriate chemotherapy regimen. therefore, the coxen trial can offer information as to which patients may benefit from chemotherapy and whether mvac or gc is more appropriate based on tumor genetics. overall, both the university hospital, rouen and the southwest oncology group trials will offer insight into the safety profiles and efficacy of the two regimens when directly compared in a prospective randomized fashion. neither of these trials compares chemotherapy regimens to a control group of cystec- tomy without neoadjuvant treatment. lastly, kitamura et al. of the japanese clinical oncology group found a % increase in os for patients undergoing neoadju- vant mvac compared to cystectomy-alone but did not quite reach statistical significance (p = . ) [ ]. this study was closed early due to poor accrual and had a total of patients in the neoadjuvant arm and patients in the radical cystectomy-only arm [ ]. though the results were not statistically significant, this recently published trial offers continued cautious sup- port for a cisplatin-based regimen in the neoadjuvant setting. vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / addressing the role of neoadjuvant chemotherapy for cisplatin-ineligible patients has been difficult. only phase iii neoadjuvant trial from has been conducted for carboplatin-based treatment [ ]. our medline re- view of neoadjuvant chemotherapy identified a single arm prospective trial of gemcitabine and carboplatin leading to a . % ( / ) pt stage at surgery and an . % os rate with a median follow-up of months [ , ]. % of these patients were ruled cisplatin-ineligible [ , ]. the study was hampered by a single arm design at a single in- stitution with a shortened follow-up timeframe [ , ]. moreover, phase ii carboplatin-based regimen trials have showed either no change in efficacy or worse efficacy than cisplatin-based regimens with varying levels of toxicity [ , ]. furthermore, our adjuvant chemotherapy med- line review identified a phase ii neoadjuvant split-dosed cisplatin study for mibc patients [ ]. the treatment regimen was well-tolerated with of patients achiev- ing complete response to neoadjuvant treatment while out of the patients with chronic kidney disease had fur- ther decline in renal function [ ]. the study was limited as only patients underwent radical cystectomy while others preferred organ-sparing or palliative treatment due to age or comorbidities, but nevertheless, the split-dosed cisplatin regimen offers a possible alternative to traditional cisplatin-based therapy for patients who have renal im- pairment [ ]. to our knowledge, no phase iii split-dosed cisplatin trials have been conducted, thus reaffirming the eau and the society of urologic oncology’s stance for preceding directly to cystectomy for cisplatin-ineligible patients [ , ]. the role of future trials in neoadjuvant chemotherapy should continue to assess particular regimens based on individual patient demographics and tumor specifics. randomized, multi-center phase iii trials for split-dosed cisplatin and carboplatin are also needed for possible evidence-based alternatives for patients who are unable to tolerate a traditional cisplatin-based regimen. though meta-analysis data has proven its efficacy, several ques- tions still need to be investigated to improve the benefit of neoadjuvant chemotherapy on a personalized level. adjuvant chemotherapy trials adjuvant chemotherapy for mibc has recently been shown to be beneficial though a specific regimen based on patient characteristics has yet to be discerned. the updated meta-analysis by leow et al. reported a % risk reduction in mortality with adjuvant chemotherapy [ ]. all trials involved cisplatin-based regimens [ ]. in total, only patients were used in the meta-analysis compared to the patients and patients included in the most recent neoadjuvant chemotherapy meta- analyses [ , , ]. additionally, the results suggest that adjuvant chemotherapy may play a greater role in patients with lymph node positive (n+) disease [ ]. the up- date by leow et al. added trials and updated trial to the original trials included in the meta-analysis by the advanced bladder cancer meta-analysis corporation [ , ]. the original meta-analysis was limited by a low study power with only patients included [ ]. in a separate analysis, tjokrowidjaja et al. reported a % risk reduction in mortality with patients, which is consist- ent with the findings of leow et al. [ , ] the limited number of patients in all meta-analyses leads to difficulty towards recommending adjuvant therapy to all mibc pa- tients post-cystectomy. in fact, the eau has yet to recom- mend adjuvant chemotherapy [ ]. recent commentary on the eau guidelines by sternberg et al. as well as the perioperative chemotherapy summary from the society of urologic oncology annual meeting suggest that this is purely because the data on adjuvant chemo- therapy is insufficient [ , ]. our review identified phase iii trials registered with the clinical trials registry including completed head-to- head trial, trials involving gc, and comparing a traditional adjuvant delivery with a disease progression- triggered schedule. the head-to-head trial is a comparison of mvac with pca completed by the eastern cooperative oncology group though the results have yet to be reported. to our knowledge, this is the first completed periopertive chemotherapy trial directly comparing two chemotherapy regimens. this trial can offer insight into taxane-based regimens compared to traditional cisplatin- based treatments. pca has been studied in the past as a single-arm trial of patients showing tolerable patient uptake with a -year os of . % [ ]. a head-to-head comparison with a traditional regimen will shed light on the efficacy of a taxane-based regimen and may prove its value for cisplatin-ineligible patients. three of the adjuvant trials selected by our review involve the use of gc alone, gc with radiotherapy, or gc delivered via the intrarterial route. cognetti et al. of fondazione c.n.r./regione toscana evaluated adjuvant gc at tumor relapse compared to cystectomy alone and found a -year os hazard ratio (hr) of . in the treatment arm but failed to reach statistical significance (p = . ) [ ]. only patients were included and the study failed to reach its accrual goal forcing the study to be closed early [ ]. this study is included in the meta-analysis by leow et al., and to our knowledge, is the first completed adjuvant phase iii trial of gc alone, which is a common regimen for neoadjuvant therapy [ ]. an- other phase iii trial at the cairo university by zaghloul et al. compared adjuvant gc with radiotherapy with postoperative radiotherapy alone and showed an increase in -month dfs from +/− % to +/− % in the chemotherapy arm but failed to be statistically significant (p = . ) [ ]. chemoradiation has been proven effective vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / previously as a bladder-sparing modality for mibc in a pooled-analysis of phase ii and phase iii trials, but this publication by zaghloul et al. is the first phase iii trial evaluating chemoradiation in the post-cystectomy setting [ ]. though the findings were not statistically significant, the combination of adjuvant gc with radiotherapy im- proved dfs, which suggests that adjuvant chemoradiation should be studied more extensively in the future [ ]. lastly, a current phase iii study by sun yat-sun university is evaluating the use of – cycles of intrarterial gc for adjuvant use within – weeks following cystectomy. previously, a retrospective trial comparing patients receving intraarterial gc with patients solely undergo- ing cystectomy has recently been published by jiang et al. [ ] the data from the retrospective study is promising as the hr for -year survival dropped to . (p = . ) in the adjuvant treatment arm with the most common adverse effect being transient myelosuppresion ( %) [ ]. a phase iii trial of such a delivery method will shed better insight into the intraarterial options for adjuvant chemotherapy. the last two adjuvant chemotherapy trials identified by our review investigate the role of disease progression- triggered delivery of adjuvant chemotherapy compared to a traditional adjuvant delivery. the association of uro- genital oncology phase iii trial of adjuvant gemcitabine compared to disease progression-triggered gemcitabine for cisplatin-ineligible patients showed an increase in dfs (hr . ), cancer specific survival (hr . ), and -year os (hr . ), but failed to show statistical significance for any of these outcomes (p = . , p = . , p = . , respectively) [ ]. of note, the trial closed early and in- cluded patients for analysis rather than the planned patients [ ]. nonetheless, the trial offered an alterna- tive to cisplatin-based treatment with a novel delivery time. lastly, the eastern organisation for research and treatment of cancer phase iii trial of adjuvant mvac vs. disease progression-triggered mvac was withdrawn prior to enrollment. currently, evidence fails to suggest that a non-cisplatin based regimen or a progression-triggered treatment is beneficial in the adjuvant setting. neoadjuvant chemotherapy has thus far been more thoroughly studied than adjuvant chemotherapy as highlighted by the larger number of patients included in the neoadjuvant meta-analyses. regardless, similar to the current need in neoadjuvant chemotherapy clinical re- search for mibc, more head-to-head phase iii adjuvant chemotherapy trials are necessary to understand the benefits and drawbacks of particular regimens with the hopes of offering a certain combination based on patient and tumor specifics. additionally, further trials particu- larly studying n+ patients can offer insight into the benefit of adjuvant chemotherapy. one previous phase ii trial conveyed that gc is a tolerable adjuvant regimen for n+ disease while a phase iii trial included in the updated meta-analysis by leow et al. concluded that a cisplatin-based adjuvant regimen can particularly benefit patients with > pt and/or n+ disease [ , , ]. focusing phase iii trials on this particular group of pa- tients may define a specific role for adjuvant chemother- apy. finally, current shortcomings of adjuvant trials are likely due to poor accrual forcing studies to close early. clinicians and investigators need to continue to encour- age patients to participate in these trials to improve our understanding of adjuvant chemotherapy. perioperative targeted therapy trials thus far, the role of targeted therapy in the perioperative management of mibc is convoluted because of the lim- ited data from clinical studies. no active phase iii trials of perioperative targeted therapy were identified by our sys- tematic review. our review identified currently active or recently completed phase ii trials. eight of these studies are purely neoadjuvant with other trials evaluating a combination of neoadjuvant and adjuvant therapy. this discrepancy between the number of neoadjuvant and advjuvant targeted therapy trials may be attributed to neo- adjuvant chemotherapy being considered beneficial for a much longer time and with larger phase iii studies than adjuvant chemotherapy. five trials include targeted ther- apy combination treatment with chemotherapy. for of these trials, the chemotherapy selected for combination treatment include traditional cisplatin-based regimens, which is expected as cisplatin-based regimens dominate the options for perioperative chemotherapy for mibc. of note, the dendreon adjuvant study of dn - , which is designed to stimulate an immune response against the human epidermal growth factor receptor (her ) is discussed below as an immune therapy. erlotinib may be considered the best candidate for a fu- ture phase iii trial of targeted therapy. unlike other tar- geted therapies, erlotinib is mainly being evaluated alone without combination chemotherapy. a phase ii trial of neoadjuvant erlotinib with clinical t patients at the university of north carolina led to ( %) patients being downstaged to < pt at cystectomy without significant tox- icity [ ]. since this trial was not a combination treatment trial with chemotherapy, the results convey that erlotinib alone can possibly offer neoadjuvant intervention. two ac- tive erlotinib trials are currently being investigated by the university of north carolina and m.d. anderson cancer center, which will further offer insight on the safety and ef- ficacy of erlotinib alone for neoadjuvant mibc. multiple phase ii trials of bevacizumab with chemother- apy are assessing the role of the vegf inhibitor in peri- operative mibc management. a phase ii trial by the medical university of south carolina evaluating bevacizu- mab with gc in the neoadjuvant setting and bevacizumab with paclitaxel in the adjuvant setting reported a % vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / downstaging rate with a % postoperative complication rate for patients in [ ]. all patients underwent neoadjuvant therapy but only patients underwent adju- vant treatment, and therefore, the effect of paclitaxel with bevacizumab cannot yet be predicted [ ]. postoperative complications included enterovesical fistula, ileus, and pel- vic abscess [ ]. further patient accrual for the study will help describe whether the downstaging rate will be benefi- cial enough to use bevacizumab despite its high surgical complication risk once more results are reported. in addition, bevacizumab is currently being evaluated with mvac by m.d. anderson cancer center, which will further offer outcomes of the added treatment. further information is needed currently for bevacizumab before a phase iii trial can be undertaken. five of our reviewed phase ii targeted therapy trials involve sunitinib, but of these studies by the university of michigan and the hoosier oncology group have been withdrawn due to poor accrual or patient toxicities, respectively. a trial of sunitinib with gc by us oncology was also terminated early due to a hematologic toxicity rate of % despite lower dosing of gc to offset early ad- verse effects [ ]. only mibc cases were evaluated prior to radical cystectomy, but % ( / ) patients did have complete pathologic response to treatment [ ]. the same regimen was studied at memorial sloan-kettering cancer center and was hampered by a low pt rate of % ( / ), which is a slightly decreased rate to that of gc or mvac neoadjuvant therapy alone, thus suggesting a lim- ited efficacy of the added sunitinib [ , , ]. case com- prehensive cancer center has recently completed a trial of neoadjuvant sunitinib alone, which will hopefully assess whether sunitinib can have a role monotherapy, but the results remain unreported. neoadjuvant sunitinib has not been proven beneficial at this time either as a monother- apy or as a combination therapy and evidence does not currently warrant a phase iii trial. our review also identified trials involving dasatinib and sorafenib as possible perioperative targeted therapies. a phase ii trial of dasatanib monotherapy by the hoosier oncology group of patients was tolerated by % with % undergoing hematologic toxicity [ ]. the follow-up pathology results conveyed a decrease in phosphorylated sfk expression in % ( / ) of patients but failed to show a change in cell proliferation or apoptosis while not identifying a downstaging rate [ ]. phosphorylated sfk expression is a common and specific histologic manifestation of bladder cancer [ ]. further investiga- tion is required to understand how dastanib’s effect on phosphorylated sfk expression can alter patient out- comes and if this actually has any clinical role towards treating mibc. lastly, our review found a sorafenib with gc study at fondazione c.n.r./regione toscana, though the status of this study remains unreported. the next step towards considering targeted therapy as a legitimate option for the perioperative management of mibc would be a phase iii trial. the most feasible op- tion would be to await the results of the current erloti- nib clinical trials at the university of north carolina and the m.d. anderson cancer center. based on the findings from these studies in addition to the published results from a previous university of north carolina neoadjuvant erlotinib trial, erlotinib alone could be a candidate for a phase iii trial [ ]. further phase ii trials of neoadjuvant dasatinib addressing tumor staging fol- lowing treatment could lead to a phase iii trial. as of now, an appropriate regimen for a phase iii trial of chemotherapy and targeted therapy would be difficult to elucidate based on the current data. perioperative mtor inhibitor trials the mtor protein serves as a regulator for cell growth and proliferation based on stimuli stemming from nutri- ent availability and energy processes within the cell [ ]. the upregulation of this protein has been shown in solid tumors and hematologic malignancies, and therefore, in- hibition of the mtor protein can limit malignancy growth for patients refractory to other treatments [ ]. consequently, the role of mtor inhibitors is now being evaluated in bladder cancer. everolimus has been shown to affect bladder cancer lines in vitro and in mouse models [ ]. additionally everolimus has been combined with cisplatin to decrease cell proliferation in vitro [ ]. temsirolimus and everolimus have been studied as phase ii trials in metastatic urothelial cancer [ , ]. however, no perioperative specific trials for mibc have been completed. our review identified phase neoadjuvant trial of sir- olimus at the university of texas health science center at san antonio. this trial is of particular importance be- cause it can offer evidence that mtor inhibitors can be safely tolerated in patients undergoing a major cystectomy operation. additionally, despite failing to fulfill criteria as being perioperative specific studies, we selected an everolimus with paclitaxel trial by the hoosier oncology group and a sirolimus with gc trial by the university of washington. these trials were selected because of the lack of current available mtor inhibitor trials. the everolimus with paclitaxel phase ii trial is a first-line therapy for inva- sive bladder cancer for patients who are not candidates for cisplatin-based treatment. the sirolimus with gc trial is a phase i/ii study for patients who are likely to undergo cystectomy, thus it can be considered a neoadjuvant study, but surgical intervention is not definitive based on the clinical trials registry information. the results from these studies may promote pure perioperative studies of mtor inhibitors in the future. vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / further perioperative phase i and phase ii trials of mtor inhibitors for mibc are imperative to encourage their use. as of now, no real clinical data can define the role of mtor inhibitors for mibc patients undergoing cystectomy. perioperative immune therapy trials immune therapy has yet to be thoroughly studied in blad- der cancer. a large phase ii trial has been previously con- ducted evaluating interferon-alpha b with the bacillus calmette–guérin (bcg) vaccine in non-invasive bladder cancer [ ]. interferon-alpha- b has further been studied independently for non-invasive bladder cancer as intrave- sical administration for patients who failed bcg therapy [ ]. however, no trials have been completed examining immune therapy for mibc. despite not fulfilling criteria for being a perioperative specific study, we selected a trial of interferon-alpha for advanced bladder cancer at m.d. anderson cancer center. evaluating the tolerability of interferon-alpha in advanced bladder cancer can help define the safety profile of the treatment. the trial was selected due to the scarcity of im- mune therapy trials for mibc. in addition, an ongoing trial reported in table by dendreon of dn - , an immune stimulant against her , includes the use of culturing per- ipheral blood mononuclear cells with recombinant antigen ba following -week cycles of leukopharesis and in- fuses this product back into the patient’s bloodstream [ ]. ba is a fusion protein of her , a protein that has intracellular and extracellular elements of her , and granulocyte-macrophage colony-stimulating factor, which enhances activation of antigen presenting cells (apc’s). measured responses of the trial include serum apc activa- tion, particularly ba and her specific antibody proliferations [ ]. thus far, patients have been en- rolled in this trial with % of these patients having some level of her expression in the primary tumor and % in lymph node samples. thirty patients have undergone the full infusion cycles with apc activation evident after each infusion but responses were greater at infusions and than at infusion [ ]. further long-term results can describe the actual effect of dn - on clinical outcomes. on an overarching note, the administration of dn - can be considered a mixture of targeted therapy with im- mune therapy considerations. given the heterogeneity of bladder cancer biology, and the clear need for less toxic, non-cisplatin based therapies and therapies for patients in whom neoadjuvant chemotherapy has failed, the fact that this phase ii trial actually met its accrual goal should be celebrated as enthusiasm for rational targeted therapy by investigators for mibc. moving forward, perioperative specific trials of interferon-alpha and a further phase iii trial of dn - are necessary to understand their efficacy and tolerability among patients undergoing cystectomy. the data for im- mune therapy is currently too limited to suggest its benefit in mibc. perioperative gene and vaccine therapy trials gene and vaccine therapy trials for mibc are being evalu- ated for patient tolerability. a phase i/iia trial of adcd l by uppsala university used an intravesical administration of an adenovirus vector to upregulate the human cd lig- and in mibc patients scheduled for cystectomy [ ]. only out of patients experienced a serious adverse effect postoperatively, and the histologic samples showed in- creased immune activation and decreased cell proliferation with successful gene transfer [ ]. though only an early phase trial, the study conveyed that gene therapy can be well tolerated prior to cystectomy. in addition to the adcd l trial, our review identified two other current studies in gene and vaccine therapy. celldex therapeutics terminated a phase ii trial of neoad- juvant and adjuvant cdx- with chemotherapy for mibc patients with newly diagnosed muscle-invasive bladder cancer (the n-able trial). the study was ter- minated due to poor accrual. cdx- is a monoclonal antibody targeting the mannose receptor and bhcg, a common couplet in epithelial tumors [ ]. cdx- had been studied previously in a phase i trial of multiple tumors showing tolerability of the vaccine and increased immune response [ ]. the cdx- phase ii trial of- fered a novel combination of vaccine therapy and chemo- therapy, and such a treatment regimen can be emulated in future studies. the trial’s failure to accrue its patient goal may be a reflection of a current lack of faith in vaccine or immune therapy for mibc. furthermore, memorial sloan-kettering cancer center and the national cancer institute have completed a phase i vaccine trial of ny- eso- , bcg, and sargramostim in patients who recently underwent cystectomy for bladder cancer and express ny-eso- and/or lage- antigens histologically. ny- eso- and lage- are antigens expressed by testicular germ cells and a limited number of tumors; however, pre- vious results indicate that almost % of high grade urothelial carcinomas express one or both of these anti- gens, and thus, a vaccine for either may prove beneficial in bladder cancer [ ]. previously, sharma et al. have demon- strated tolerable uptake of the vaccine with bcg and granulocyte macrophage colony-stimulating factor for patients that underwent cystectomy for mibc [ ]. ny-eso- specific antibody responses were found in / patients with cd t-cell responses being found in / patients [ ]. the new phase i study that has been com- pleted had a project accrual of – patients and will shed more light on the tolerability of the vaccine for mibc patients. the ny-eso- vaccine has been used tested previously in patients with non-small cell lung vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / cancer, esophageal carcinoma, or prostate adenocarcin- oma and was well-tolerated [ ]. the clinical outcomes were mixed but it was noted that regulatory t cells in the blood may limit the antibodies produced by the vaccine and suppressing this immune response can possibly im- prove the vaccine’s effectiveness [ ]. this is one of the shortcomings of the vaccine that requires further study. nonetheless, identifying eligible patients for the ny-eso- vaccine based on tumor specific antigens is a reflection of mibc research moving towards personalized medicine. the safety profile of gene and vaccine therapy has yet to be elucidated for any specific treatment. adcd l is a promising immune-stimulating agent, but a future phase ii trial is necessary to further assess tolerability. cdx- theoretically would have benefit in bhcg expressing bladder cancers; however, no bladder cancer specific trial has been completed thus far. lastly, ny- eso- and lage- offer targets for peptide vaccine therapy, but results from the current ny-eso- vaccine trial and future clinical studies are required to assess the tolerability and efficacy of the treatment. expanding areas for future clinical trials the cancer genome atlas (tcga) research network is a widespread effort developed by the national cancer insti- tute to collect specimens of several cancers with the hopes of analyzing the genetics and molecular biology of tumors to identify common mutations and targets for treatment [ ]. recently, tcga published data on mibc speci- mens [ ]. mutations in genes were found to be statisti- cally significant, reinforcing current targets for clinical trials and suggesting new areas for therapeutic studies [ ]. using the data from the consortium to identify need for future study may prove beneficial towards constructing new trials. limitations of our systematic review all searches were limited to the clinical trials registry, medline, and google. any abstract or current clinical trial unregistered or unreported by the aforementioned databases may have been overlooked unless specifically known of through recent conferences. however, our searches detailed an extensive review of multiple databases for perioperative mibc trials. phase i and ii trials of perioperative chemotherapy were not substantially discussed above because of the wide array of literature on the topic currently. phase iii peri- operative chemotherapy subset studies of previous larger phase iii trials were not substantially discussed because of overlap with trials included in past meta-analyses. only non-subset phase iii trials were evaluated in detail because of their possible changes in clinical management of perioperative mibc in the near future. similarly, phase i trials of targeted therapy were not closely addressed in comparison to phase ii trials. conclusion although the perioperative management of mibc has been limited to platinum-based chemotherapy, several novel treatments are being evaluated for tolerability and efficacy. phase iii trials directly comparing chemotherapy regimens are now being studied, which will hopefully classify different regimens for specific patients. targeted therapy with monoclonal antibodies is the closest non- chemotherapy treatment to being implemented, but no phase iii trial of targeted therapy has been conducted. further trials in all treatment modalities are required to address the large need for improved perioperative options in mibc management. abbreviations mibc: muscle invasive bladder cancer; medline: medical literature analysis and retrieval system online; mvac: methotrexate, vinblastine, adriamycin (doxorubicin), and cisplatin; gc: gemcitabine and cisplatin; pca: paclitaxel and carboplatin; cmv: cisplatin, methotrexate, vinblastine; hd: high dose; dd: dose dense; sfk: src family kinase; egfr: epidermal growth factor receptor; vegf: vascular epithelial growth factor; her r: human epidermal growth factor receptor ; bcg: bacillus calmette–guérin; bhcg: beta-human chorionic gonadotropin. competing interests the authors declare that they have no competing interests. authors’ contributions conception and design: sd and vv; acquisition of data: vv, sd, and dq; analysis and interpretation of data: all authors; manuscript drafting: vv and sd; manuscript revising: all authors; final approval of this version: all authors. all authors read and approved the final manuscript. acknowledgements we would like to thank debra l. zynger, m.d. of the department of pathology at the ohio state wexner medical center for her contributions to our systematic review. author details department of internal medicine, cleveland clinic foundation, cleveland, oh, usa. division of oncology, usc/norris comprehensive cancer center, usc institute of urology, los angeles, ca, usa. department of urology, usc/norris comprehensive cancer center, usc institute of urology, eastlake abe, suite , los angeles, ca , usa. received: september accepted: december published: december references . siegel r, ma j, zou z, jemal a: cancer statistics, . ca cancer j clin , : – . . ploeg m, aben kk, la kiemeney world: the present and future burden of urinary bladder cancer in the world. j urol , : – . . bassi p, pappagallo gl, cosciani s, lembo a, anselmo g, sperandio p, signorelli g, di tonno f, hurle r, lavelli d, seren m, piazza n, faggiano l, pagano f: neoadjuvant mvac chemotherapy of invasive bladder cancer results of a multicenter phase iii trial. j urol , : . . grossman hb, natale rb, tangen cm, speights vo, vogelzang nj, trump dl, de vere white rw, sarosdy mf, wood dp jr, raghavan d, crawford ed: neoadjuvant chemotherapy plus cystectomy compared with cystectomy alone for locally advanced bladder cancer. n engl j med , : – . vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / . griffiths g, hall r, sylvester r, raghavan d, parmar mk: international phase iii trial assessing neoadjuvant cisplatin, methotrexate, and vinblastine chemotherapy for muscle-invasive bladder cancer: long-term results of the ba trial. j clin oncol , : – . . sherif a, holmberg l, rintala e, mestad o, nilsson j, nilsson s, malmström pu, nordic urothelial cancer group: neoadjuvant cisplatinum based combination chemotherapy in patients with invasive bladder cancer: a combined analysis of two nordic studies. eur urol , : – . . bono av, benvenuti c, gibba a: adjuvant chemotherapy in locally advanced bladder cancer: final analysis of a controlled multicentre study. acta urol ital , : – . . freiha f, reese j, torti fm: a randomized trial of radical cystectomy versus radical cystectomy plus cisplatin, vinblastine and methotrexate chemotherapy for muscle invasive bladder cancer. j urol , : – . discussion – . . stadler wm, lerner sp, groshen s, stein jp, shi sr, rghavan d, esrig d, steinberg g, wood d, klotz l, hall c, skinner dg, cote rj: phase iii study of molecularly targeted adjuvant therapy in locally advanced urothelial cancer of the bladder based on p status. j clin oncol , : – . . lehmann j, franzaring l, thuroff j, wellek s, stockle m: complete long- term survival data from a trial of adjuvant chemotherapy vs control after radical cystectomy for locally advanced bladder cancer. bju int , : – . . otto t, goebell pj, rubben h: perioperative chemotherapy in advanced bladder cancer—part ii: adjuvant treatment. onkologie , : – . . paz-ares lg, solsona e, esteban e, saez a, gonzalez-larriba j, anton a, hevia m, de la rosa f, guillem v, bellmunt j: randomized phase iii trial comparing adjuvant paclitaxel/gemcitabine/cisplatin (pgc) to observation in patients with resected invasive bladder cancer: results of the spanish oncology genitourinary group (sogug) / study [abstract]. j clin oncol , : . . skinner dg, daniels jr, russell ca, lieskovsky g, boyd sd, nichols p, kern w, sakamoto j, krailo m, groshen s: the role of adjuvant chemotherapy following cystectomy for invasive bladder cancer: a prospective comparative trial. j urol , : – . discussion – . . zaid hb, patel sg, stimson cj, resnick mj, cookson ms, barocas da, chang ss: trends in the utilization of neoadjuvant chemotherapy in muscle- invasive bladder cancer: results from the national cancer database. urology , : – . . apolo ab, kim jw, bochner bh, steinberg am, bajorin df, kelly kw, agarwal pk, koppie tm, kaag mg, quinn di, vogelzang nj, sridhar ss: examining the management of muscle invasive bladder cancer (mibc) by medical oncologists in the united states. urol oncol. in press. . tjokrowidjaja a, lee c, stockler mr: does chemotherapy improve survival in muscle-invasive bladder cancer (mibc)? a systematic review and meta-analysis (ma) of randomized controlled trials (rct) [abstract]. j clin oncol , : . . advanced bladder cancer (abc) meta-analysis collaboration: neoadjuvant chemotherapy in invasive bladder cancer: update of a systematic review and meta-analysis of individual patient data. eur urol , : – . . winquist e, kirchner ts, segal r, chin j, lukka h, genitourinary cancer disease site group, cancer care ontario program in evidence-based care practice guidelines initiative: neoadjuvant chemotherapy for transitional cell carcinoma of the bladder: a systematic review and meta-analysis. j urol , : – . . leow jj, martin-doyle w, rajagopal ps, patel cg, anderson em, rothman at, cote rj, urun y, chang sl, choueiri tk, bellmunt j: adjuvant chemotherapy for invasive bladder cancer: a updated systematic review and meta-analysis of randomized trials. eur urol , : – . . advanced bladder cancer (abc) meta-analysis collaboration: adjuvant chemotherapy in invasive bladder cancer: a systematic review and meta-analysis of individual patient data advanced bladder cancer (abc) meta-analysis collaboration. eur urol , : – . discussion – . . kumar pal s, southwest oncology group: examining the pre-metastatic niche in muscle invasive bladder cancer. [http://www.swog.org/visitors/ fall gpmtg/plenary/ -mcconkey-plenaryi-october- .pdf] . kitamura h, tsukamoto t, shibata t, masumori n, fujimoto h, hirao y, fujimoto k, kitamura y, tomita y, tobisu k, niwakawa m, naito s, eto m, kakehi y, urologic oncology study group of the japan clinical oncology group: randomised phase iii study of neoadjuvant chemotherapy with methotrexate, doxorubicin, vinblastine and cisplatin followed by radical cystectomy compared with radical cystectomy alone for muscle-invasive bladder cancer: japan clinical oncology group study jcog . ann oncol , [epub ahead of print]. . cognetti f, ruggeri em, felici a, gallucci m, muto g, pollera cf, massidda b, rubagotti a, giannarelli d, boccardo f, study group: adjuvant chemotherapy with cisplatin and gemcitabine versus chemotherapy at relapse in patients with muscle-invasive bladder cancer submitted to radical cystectomy: an italian, multicenter, randomized phase iii trial. ann oncol , : – . . zaghloul ms, khaled hm, lotayef m, william hm, nazmy attia g: a controlled prospective randomized trial of adjuvant chemoradiotherapy following radical cystectomy in advanced high risk bladder cancer [abstract]. j clin oncol , : . . lehmann j, kuehn m, fischer c, volkmer b, von rundstedt f, albers p, becht e, bannowsky a, hofmann r, langbein s, retz m, stoeckle m: randomized phase iii study of adjuvant versus progression-triggered treatment with gemcitabine (g) after radical cystectomy (rc) for locally advanced bladder cancer (labc) in patients not suitable for cisplatin-based chemotherapy (cbc) (auo-trial ab / ) [abstract]. j clin oncol , : . . pruthi rs, nielsen m, heathcote s, wallen em, rathmell wk, godley p, whang y, fielding j, schultz h, grigson g, smith a, kim w: a phase ii trial of neoadjuvant erlotinib in patients with muscle-invasive bladder cancer undergoing radical cystectomy: clinical and pathological results. bju int , : – . . dinney cp, fisher mb, navai n, o’donnell ma, cutler d, abraham a, young s, hutchins b, caceres m, kishnani n, sode g, cullen c, zhang g, grossman hb, kamat am, gonzales m, kincaid m, ainslie n, maneval dc, wszolek mf, benedict wf: phase i trial of intravesical recombinant adenovirus mediated interferon-α b formulated in syn for bacillus calmette-guérin failures in nonmuscle invasive bladder cancer. j urol , : – . . chaudhary ub, golshayan ar, brisendine a, kraft as, clarke h, keane te: phase ii trial of neoadjuvant cisplatin, gemcitabine, and bevacizumab followed by radical cystectomy (rc) in patients with muscle-invasive transitional cell carcinoma (tcc) of the bladder [abstract]. j clin oncol , : . . galsky md, hahn nm, powles t, hellerstedt ba, lerner sp, gardner ta, yu m, o’rourke m, vogelzang nj, kocs d, mckenney sa, melnyk am jr, hutson te, rauch m, wang y, asmar l, sonpavde g: gemcitabine, cisplatin, and sunitinib for metastatic urothelial carcinoma and as preoperative therapy for muscle-invasive bladder cancer. clin genitourin cancer , : – . . balar av, iyer g, apolo ab, regazzi am, garcia-grossman ir, pendse d, ostrovnaya i, chou jf, bochner b, dalbagni g, herr hw, milowsky mi, bajorin df: phase ii trial of neoadjuvant gemcitabine (g) and cisplatin (c) with sunitinib in patients (pts) with muscle-invasive bladder cancer (mibc) [abstract]. j clin oncol , : . . hahn nm, daneshmand s, posadas em, koch mo, bihrle r, foster r, masterson ta, cheng l, liu z, breen t, fleming mt, lance r, ryan cw, corless cl, galsky md, alva as, mitchell c, shen ss, lerner sp: a phase ii trial of neoadjuvant dasatinib (neo-d) in muscle-invasive urothelial carcinoma of the bladder (miucb): hoosier oncology group gu - trial [abstract]. j clin oncol , : . . knudsen b, hahn nm, daneshmand s, posadas em, muhanty s, lerner sp, sonpavde g: biologic activity of dasatinib administered as neoadjuvant therapy preceding radical cystectomy (rc) for muscle-invasive bladder cancer (mibc) [abstract]. j clin oncol , : . . malmström pu, loskog as, lindqvist ca, mangsbo sm, fransson m, wanders a, gårdmark t, tötterman th: adcd l immunogene therapy for bladder carcinoma–the first phase i/iia trial. clin cancer res , : – . . meeks jj, bellmunt j, bochner bh, clarke nw, daneshmand s, galsky md, hahn nm, lerner sp, mason m, powles t, sternberg cn, sonpavde g: a systematic review of neoadjuvant and adjuvant chemotherapy for muscle-invasive bladder cancer. eur urol , : – . . ohyama c, hatakeyama s, yoneyama t, koie t: neoadjuvant chemotherapy with gemcitabine plus carboplatin followed by immediate radical cystectomy for muscle-invasive bladder cancer. int j urol , : – . . koie t, ohyama c, hashimoto y, hatakeyama s, yamamoto h, yoneyama t, kamimura n: efficacies and safety of neoadjuvant gemcitabine plus https://tcga-data.nci.nih.gov/tcga/ https://tcga-data.nci.nih.gov/tcga/ vashistha et al. bmc cancer , : page of http://www.biomedcentral.com/ - / / carboplatin followed by immediate cystectomy in patients with muscle- invasive bladder cancer, including those unfit for cisplatin: a prospective single-arm study. int j clin oncol , : – . . rosenblatt r, sherif a, rintala e, wahlqvist r, ullén a, nilsson s, malmström pu, nordic urothelial cancer group: pathologic downstaging is a surrogate marker for efficacy and increased survival following neoadjuvant chemotherapy and radical cystectomy for muscle-invasive urothelial bladder cancer. eur urol , : – . . scosyrev e, ely bw, messing em, speights vo, grossman hb, wood dp, de vere white rw, vogelzang nj, trump dl, natale rb, tangen cm, crawford ed, thompson im: do mixed histological features affect survival benefit from neoadjuvant platinum-based combination chemotherapy in patients with locally advanced bladder cancer? a secondary analysis of southwest oncology group-directed intergroup study (s ). bju int , : – . . hussain sa, palmer dh, lloyd b, collins si, barton d, ansari j, james nd: a study of split-dose cisplatin-based neo-adjuvant chemotherapy in muscle-invasive bladder cancer. oncol lett , : – . . witjes ja, compérat e, cowan nc, de santis m, gakis g, lebret t, ribal mj, van der heijden ag, sherif a: eau guidelines on muscle-invasive and metastatic bladder cancer: summary of the guidelines. eur urol , : – . . apolo ab, grossman hb, bajorin d, steinberg g, kamat am: practical use of perioperative chemotherapy for muscle-invasive bladder cancer: summary of session at the society of urologic oncology annual meeting. urol oncol , : – . . choi w, porten s, kim s, willis d, plimack er, hoffman-censits j, roth b, cheng t, tran m, lee il, melquist j, bondaruk j, majewski t, zhang s, pretzsch s, baggerly k, siefker-radtke a, czerniak b, dinney cp, mcconkey dj: identification of distinct basal and luminal subtypes of muscle- invasive bladder cancer with different sensitivities to frontline chemotherapy. cancer cell , : – . . abol-enein h, el-mekresh m, el-baz m, ghoneim ma: neo-adjuvant chemotherapy in the treatment of invasive transitional bladder cancer: a controlled prospective randomised study. br j urol , : . . grivas pd, hussain m, hafez k, daignault-newton s, wood d, lee ct, weizer a, montie je, hollenbeck b, montgomery js, alva a, smith dc: a phase ii trial of neoadjuvant nab-paclitaxel, carboplatin, and gemcitabine (acag) in patients with locally advanced carcinoma of the bladder. urology , : – . . smith dc, mackler nj, dunn rl, hussain m, wood d, lee ct, sanda m, vaishampayan u, petrylak dp, quinn di, beekman k, montie je: phase ii trial of paclitaxel, carboplatin and gemcitabine in patients with locally advanced carcinoma of the bladder. j urol , : – . . sternberg cn, bellmunt j, sonpavde g, siefker-radtke ao, stadler wm, bajorin df, dreicer r, george dj, milowsky mi, theodorescu d, vaughn dj, galsky md, soloway ms, quinn d, international consultation on urologic disease-european association of urology consultation on bladder cancer : icud-eau international consultation on bladder cancer : chemotherapy for urothelial carcinoma-neoadjuvant and adjuvant settings. eur urol , : – . . bamias a, deliveliotis c, aravantinos g, kalofonos c, karayiannis a, dimopoulos ma, hellenic cooperative oncology group: adjuvant chemotherapy with paclitaxel and carboplatin in patients with advanced bladder cancer: a study by the hellenic cooperative oncology group. j urol , : – . . mak rh, hunt d, shipley wu, efstathiou ja, tester wj, hagan mp, kaufman ds, heney nm, zietman al: long-term outcomes in patients with muscle- invasive bladder cancer after bladder-preserving combined-modality therapy: a pooled analysis of rtog , , , , , and [abstract]. j clin oncol , : . . jiang l, zhang z, dong p, li y, yao k, liu z, han h, qin z, yao m, zhou f: efficacy of radical cystectomy plus adjuvant intraarterial chemotherapy with gemcitabine and cisplatin on locally advanced bladder cancer. chin med j (engl) , : – . . gofrit on, stadler wm, zorn kc, lin s, silvestre j, shalhav al, zagaja gp, steinberg gd: adjuvant chemotherapy in lymph node positive bladder cancer. urol oncol , : – . . paz-ares lg, solsona e, esteban e, saez a, gonzalez-larriba j, anton a, hevia m, de la rosa f, guillem v, bellmunt j: randomized phase iii trial comparing adjuvant paclitaxel/gemcitabine/cisplatin (pgc) to observation in patients with resected invasive bladder cancer: results of the spanish oncology genitourinary group (sogug) / study [abstract]. j clin oncol , , lba . . kaneko g, kikuchi e, matsumoto k, obata j, nakamura s, miyajima a, oya m: neoadjuvant gemcitabine plus cisplatin for muscle-invasive bladder cancer. jpn j clin oncol , : – . . cornu m, albert v, hall mn: mtor in aging, metabolism, and cancer. curr opin genet , : – . . chiong e, lee il, dadbin a, sabichi al, harris l, urbauer d, mcconkey dj, dickstein rj, cheng t, grossman hb: effects of mtor inhibitor everolimus (rad ) on bladder cancer cells. clin cancer res , : – . . pinto-leite r, arantes-rodrigues r, palmeira c, colaço b, lopes c, colaço a, costa c, da silva vm, oliveira p, santos l: everolimus combined with cisplatin has a potential role in treatment of urothelial bladder cancer. biomed pharmacother , : – . . gerullis h, eimer c, ecke th, georgas e, freitas c, kastenholz s, arndt c, heusch c, otto t: a phase ii trial of temsirolimus in second-line metastatic urothelial cancer. med oncol , : – . . milowsky mi, iyer g, regazzi am, al-ahmadie h, gerst sr, ostrovnaya i, gellert ll, kaplan r, garcia-grossman ir, pendse d, balar av, flaherty am, trout a, solit db, bajorin df: phase ii study of everolimus in metastatic urothelial cancer. bju int , : – . . joudi fn, smith bj, o’donnell ma, national bcg-interferon phase investigator group: final results from a national multicenter phase ii trial of combination bacillus calmette-guérin plus interferon alpha- b for reducing recurrence of superficial bladder cancer. urol oncol , : – . . bajorin df, sharma p, gomella lg, plimack er, o’donnell ph, hoffman- censits jh, flaig tw, quinn di, sims rb, locker m, sheikh na, devries t, lerner sp: neuact, a phase ii, randomized, open-label trial of dn - : updated analysis of her expression, immune responses, product parameters, and safety in patients with surgically resected her + urothelial cancer [abstract]. j clin oncol , : . . morse ma, bradley da, keler t, laliberte rj, green ja, davis ta, inman ba: cdx- : a novel vaccine under study as treatment for muscle-invasive bladder cancer. expert rev vaccines , : – . . morse ma, chapman r, powderly j, blackwell k, keler t, green j, riggs r, he lz, ramakrishna v, vitale l, zhao b, butler sa, hobeika a, osada t, davis t, clay t, lyerly hk: phase i study utilizing a novel antigen-presenting cell- targeted vaccine with toll-like receptor stimulation to induce immunity to self-antigens in cancer patients. clin cancer res , : – . . sharma p, gnjatic s, jungbluth aa, williamson b, herr h, stockert e, dalbagni g, donat sm, reuter ve, santiago d, chen yt, bajorin df, old lj: frequency of ny-eso- and lage- expression in bladder cancer and evidence of a new ny-eso- t-cell epitope in a patient with bladder cancer. cancer immun , : . . sharma p, bajorin df, jungbluth aa, herr h, old lj, gnjatic s: immune responses detected in urothelial carcinoma patients after vaccination with ny-eso- protein plus bcg and gm-csf. j immunother , : – . . gnjatic s, altorki nk, tang dn, tu sm, kundra v, ritter g, old lj, logothetis cj, sharma p: ny-eso- dna vaccine induces t-cell responses that are suppressed by regulatory t cells. clin cancer res , : – . . the cancer genome atlas. [https://tcga-data.nci.nih.gov/tcga/] . cancer genome atlas research network: comprehensive molecular characterization of urothelial bladder carcinoma. nature , : – . doi: . / - - - cite this article as: vashistha et al.: current and recent clinical trials for perioperative systemic therapy for muscle invasive bladder cancer: a systematic review. bmc cancer : . https://tcga-data.nci.nih.gov/tcga/ 	abstract 	background 	methods 	results 	conclusions 	background 	methods 	data sources 	study selection 	results 	perioperative chemotherapy trials 	perioperative targeted therapy trials 	perioperative mtor inhibitor trials 	perioperative immune therapy trials 	perioperative gene and vaccine therapy trials 	discussion 	neoadjuvant chemotherapy trials 	adjuvant chemotherapy trials 	perioperative targeted therapy trials 	perioperative mtor inhibitor trials 	perioperative immune therapy trials 	perioperative gene and vaccine therapy trials 	expanding areas for future clinical trials 	limitations of our systematic review 	conclusion 	abbreviations 	competing interests 	authors’ contributions 	acknowledgements 	author details 	references gefitinib plus celecoxib in chemotherapy-naïve patients with stage iiib/iv non-small cell lung cancer: a phase ii study from the hoosier oncology group original article gefitinib plus celecoxib in chemotherapy-naı̈ve patients with stage iiib/iv non-small cell lung cancer a phase ii study from the hoosier oncology group anuj agarwala, md,* william fisher, md,†‡ daniel bruetman, md,†§ john mcclean, md,†� david taber, md,†¶ michael titzer, md,†# beth juliar, md,* menggang yu, phd,* tim breen, phd,* lawrence h. einhorn, md,*† and nasser hanna, md*† background: gefitinib, an inhibitor of the epidermal growth factor receptor (egfr) pathway, has single agent activity in non-small cell lung cancer (nsclc). preclinical studies demonstrate significant interactions between the egfr and cyclooxygenase (cox- ) pathways and that simultaneous inhibition may have benefits over egfr inhibitors alone. methods: eligibility criteria: chemotherapy-naive, stage iiib (with pleural effusion) or iv nsclc, eastern cooperative oncology group performance status (ps) – . patients were treated with gefitinib mg po daily plus celecoxib mg po every hours. cycles consisted of -day treatment and continued until unaccept- able toxicity or progression of disease. the primary objective was to evaluate the overall response rate; secondary objectives included estimation of progression free survival, overall survival, and to assess the toxicity of this regimen. results: from january to november , patients were enrolled: male/female / ; median age years (range, – ); % had adenocarcinoma; eastern cooperative oncology group ps / / ; stage iiib/iv / . two patients died of interstitial lung disease due to treatment. there were three additional deaths during treatment that were not considered treatment related. two additional patients discontinued treatment due to adverse events (elevated liver enzymes). select grade / toxicities included: pneumonitis ( %), hepatic ( %), diarrhea ( %), and skin ( %). response rate was % ( % ci, – %), median progression free survival and overall survival were . ( % ci, . – . months) and . months ( % ci, . – . months), respectively. all responders were females with adenocarcinoma, two were remote or never smokers and three were former smokers. conclusion: gefitinib plus celecoxib in an unselected population of chemotherapy naive patients with advanced nsclc and a ps of – has a lower response rate and overall efficacy compared with historical controls of combination chemotherapy. key words: non-small cell lung cancer, gefitinib, celecoxib. (j thorac oncol. ; : – ) lung cancer is the leading cause of cancer-related death inboth men and women in the united states, with , projected deaths in the year . approximately % of these patients will have non-small cell lung cancer (nsclc) type and the majority will present with advanced disease. modest gains in survival time and quality of life have resulted from the use of chemotherapy in patients with advanced nsclc. – a multicentered trial from the united kingdom randomizing patients to supportive care versus platinum- based chemotherapy demonstrated a small, but statistically significant, survival advantage for the use of chemotherapy. unfortunately, a plateau in survival times has been reached as comparable survival results are reported with several differ- ent regimens. furthermore, although survival gains are usu- ally modest with chemotherapy, many patients experience grade or toxicities, the extent of which differs based upon the regimen used. novel strategies and therapeutics with safer side effect profiles are greatly needed for this patient population. gefitinib is an inhibitor of the epidermal growth factor receptor (egfr) intracellular tyrosine kinase. inhibition of this signal transduction cascade can affect tumor growth, angiogenesis, and survival of lung cancer cell lines. phase ii trials of gefitinib in previously treated patients with advanced nsclc reported response rates of . % and . %, respec- tively. , substantially more patients achieved stable disease for at least months, so that the disease control rate was approximately to %, comparable to what is achieved with chemotherapy in this setting. at a dose of mg daily, gefitinib was generally well tolerated with few instances of grade or toxicity. *indiana university, †hoosier oncology group, indianapolis, indiana; ‡medical consultants, p.c., muncie, indiana; §center for cancer care at goshen health system, goshen, indiana; �medical & surgical special- ists, llc, galesburg, illinois; ¶northern indiana cancer research con- sortium, south bend, indiana; and #oncology-hematology associates of sw indiana, evansville, indiana. disclosure: the authors declare no conflict of interest. address correspondences to: nasser hanna, md, barnhill drive rt , indianapolis, in . e-mail: nhanna@iupui.edu trial data were reported in part at the annual meeting of the american society of clinical oncology. copyright © by the international association for the study of lung cancer issn: - / / - journal of thoracic oncology • volume , number , april cyclooxygenase- (cox- ) is also believed to play an important role in the pathogenesis of some malignancies. investigators at indiana university characterized the fre- quency of cox- expression by immunoblot analysis in primary lung cancers from frozen tumor tissue and matched normal adjacent tissue. cox- protein was expressed more frequently and at higher levels in adenocarcinomas and squa- mous cell carcinomas of the lung when compared with the normal adjacent tissue. preclinical data has shown that cox- inhibitors inhibit both in vitro and in vivo growth of human cancer lines, demonstrate antiangiogenic activity, in- creased cellular apoptosis, and decreased cellular prolifera- tion. – furthermore, significant intracellular interactions between the egfr pathway and cox- pathway have been reported in preclinical studies, suggesting that simultaneously inhibiting both targets may provide enhanced benefits. – therefore, based on the desire to develop more effective and safer regimens for patients with advanced nsclc, this study was designed to assess the efficacy of gefitinib, an egfr inhibitor, with celecoxib, a cox- inhibitor, in chemotherapy- naive patients with stage iiib or iv nsclc. patients and methods patients were enrolled from participating sites of the hoosier oncology group, a community-based cooperative group. eligible patients met the following criteria: histologic or cytologic evidence of nsclc, stage iiib disease with pleural effusion and/or positive supraclavicular nodes or stage iv disease, chemotherapy-naive, the presence of mea- surable disease per the recist criteria, an eastern cooper- ative oncology group performance status of or at baseline, adequate baseline hematologic function (absolute neutrophil count � /mm , platelet count � , /mm , hemoglobin level � g/dl), hepatic function (total bilirubin level � mg/dl, aspartate aminotransferase and alkaline phosphatase � . � upper limits of normal), and renal function (serum creatinine level � mg/dl or a calculated creatinine clearance � ml/min). radiotherapy was al- lowed, provided the radiated area was not the only site of measurable disease and was completed � days before study registration. patients with brain metastases were eligi- ble if the brain metastases were adequately treated, asymp- tomatic, and clinically stable for at least weeks. patients were excluded if they had received prior anti-egfr therapy, had a prior malignancy, except for adequately treated basal cell or squamous cell skin cancer, in situ cervical cancer, gleason �grade organ confined prostate cancer or other cancers for which the patient has been disease-free for � years; history of severe or uncontrolled systemic disease (e.g., unstable or uncompensated respiratory, hepatic, or renal disease), active infections, history of neuropathic keratopa- thy, current treatment for ocular inflammation or corneal ulceration, or evidence of clinically active interstitial lung disease (patients with chronic stable radiographic changes who were asymptomatic were not excluded). pregnant or lactating women were excluded and women of childbearing potential and sexually active males had to be willing to use contraception during and for months after completion of protocol therapy. all patients gave written informed consent and hipaa authorization for release of protected health information. the local institutional review boards approved the protocol before enrollment at each site. all patients were treated with gefitinib mg po daily plus celecoxib mg po every hours. the study was stopped after we failed to meet our endpoints from stage i to go on to stage ii. each cycle consisted of a -day time period and treatment continued until disease progression or intoler- able toxicities occurred. physical exams and toxicity evalua- tions were completed at baseline and before each cycle and disease assessment was undertaken at baseline and prior to every other cycle. toxicities were graded using the common toxicity criteria version . . day treatment of a new cycle was administered only when the absolute neutrophil count was � /l and platelets � , /l. patients who re- quired greater than a -day interruption in therapy due to toxicity were taken off study. there were no dose reductions allowed for gefitinib. patients with grade or skin toxicity, diarrhea, stomatitis and/or any other significant toxicity thought to be related to gefitinib were treated with supportive care and gefitinib could be held up to days. treatment was restarted when the patient recovered to baseline or �grade . if patients presented with an acute worsening or new onset of respiratory symptoms such as dyspnea, cough, and fever, gefitinib therapy was interrupted and evaluation for intersti- tial lung disease (ild) was initiated. if ild was confirmed, gefitinib was permanently discontinued. if a patient devel- oped any grade or toxicity attributable to celecoxib, treatment with celecoxib was interrupted until symptoms resolved to a grade or less toxicity and then celecoxib was resumed at a % dose reduction. gastrointestinal bleeding attributable to celecoxib would have required the patient be taken off study. a maximum -day treatment interruption was allowed with only one dose reduction allowed per patient. of note, a study amendment was made on december , , stopping the use of celecoxib after evidence was shown correlating the drug with increased risk of serious cardiovascular events. this amendment was made during the interval between the first and second stages while the data was being analyzed to see if we would proceed with the second stage. this amendment would have only affected those still on study as no further patients were accrued for the study. statistical considerations the primary objective of this multiinstitutional phase ii study was to estimate the overall response rate of the com- bination of gefitinib plus celecoxib in this patient population. secondary objectives included estimation of progression free survival, overall survival, and to assess the toxicity of this regimen. patients were to be entered in two stages. if � of the initial patients achieved an objective response, the study would be terminated. if � of the initial patients achieved an objective response, accrual would continue to a total patients. a one-sided statistical test was to be carried out at the % alpha level. the null hypothesis would be rejected if the lower bound of a one-sided % confidence interval constructed around the observed response rate is higher than the historical response rate of %. the hypoth- journal of thoracic oncology • volume , number , april gefitinib plus celecoxib in nsclc copyright © by the international association for the study of lung cancer esis that the response rate is below % would be rejected in favor of the alternative that the true response rate is higher than % if at least of evaluable patients respond. if the true response rate is % or higher, this design was expected to provide better than % power to confirm that the true response rate is no less than %. for progression-free survival and overall survival, a lower % one-sided confi- dence interval was constructed around the observed median (either median time to progression or death). therapy was to be considered having exceeded the current therapeutic standard at the % confidence level if the lower bound of this confidence interval was higher than and months, respectively. response was assessed using the recist criteria, and all responses were confirmed a minimum of weeks after the initial response was recorded. duration of response is defined among responders from earliest date of response to date of progression or death, with those continuing to respond at last data collection as censored. to be assigned a status of stable disease, measurements on imaging could neither have had sufficient shrinkage to qualify for partial response nor suffi- cient criteria to show progressive disease. this had to have been shown at least once after study entry at a minimum of to weeks. survival end point is defined as time to death with surviving patients at last data collection as censored. progres- sion-free survival is defined as time to death or date of progression with surviving patients at last data collection as censored. follow-up time end point is date of last data collection with censoring at death. results from january to november , eligible patients were entered onto the study. two patients were simultaneously consented from separate sites on the same day. therefore, we elected to allow both patients to enroll. this increased enrollment by patient for the first stage of the study. the study did not meet its predefined criteria of an observation of at least % response rate to continue beyond the first stage. patient demographics and disease characteris- tics are summarized in table . the majority of patients were female and former smokers; adenocarcinoma was the predomi- nant histologic type. all but two patients had stage iv disease with eastern cooperative oncology group performance status of for patients and for patients. the median number of completed cycles was (range, – ). there were seven pa- tients who stopped celecoxib permanently in december , when the amendment to do so took effect. toxicity toxicity data is summarized in table for all patients (n � ). two patients died as a result of toxicity. one female patient, with no prior radiotherapy, developed interstitial lung disease and died to weeks after initiating treatment. of note, she had pneumonia months before entering the study. the second patient who died due to treatment also died of interstitial lung disease. the patient was a male who had prior radiotherapy to the chest and back completed in september and started protocol therapy in october . after weeks, he had stable disease but had increasing radiation- markings. he became dyspneic and was given oxygen and steroids for presumed pneumonitis. his symptoms worsened and he died in january , just months after starting protocol therapy. another patient with known prior history of congestive heart failure developed symptoms of lethargy and respiratory distress with signs of rhonchi and rales on phys- ical examination. chest radiograph confirmed volume over- load and despite diuretic and aerosol treatments the patient died; however, this was not considered treatment related. there were two other patients who died while on treatment, one from guillain-barre and another who died suddenly from a presumed pulmonary embolus, not considered to be treat- table . patient demographics and disease characteristics count percent sex female male histology adenocarcinoma squamous cell carcinoma other ecog performance status stage iiib iv smoking history never smoked former smoker current smoker unknown ecog, eastern cooperative oncology group. table . combined grade and toxicities toxicity count percent alt, sgpt (serum glutamic pyruvic transaminase) atelectasis cardiac diarrhea dizziness dyspnea fatigue hyperkalemia hypotension mucositus non-neutropenic infection pneumonitis/pulmonary infiltrates pulmonary/upper respiratory renal skin vomiting weight loss agarwala et al. journal of thoracic oncology • volume , number , april copyright © by the international association for the study of lung cancer ment related. two patients discontinued treatment before disease progression due to elevated liver enzymes. twelve patients ( %) had at least one grade or toxicity. three patients ( %) had grade diarrhea. there were no instances of grade or rash, gastrointestinal bleeds, myocardial infarctions, or cerebrovascular accidents. efficacy all patients (n � ) were considered evaluable for response, progression-free survival, and overall survival. five patients achieved a partial response ( %, % ci, – %), whereas an additional patients ( %, % ci, – %) had stable disease; the remaining patients ( %, % ci, – %) had progressive disease. the overall disease control rate (partial response plus stable disease) was % ( % ci, – %). median duration of response among five respond- ers was . months (range of response, . – . months.) characteristics of the five patients with a partial response are summarized in table . as of this analysis, of patients were alive. the median progression-free survival and overall survival times were . ( % ci, . – . months) and . months ( % ci . – . months), respectively (figure and ). the -year survival estimate is % ( % ci, – %). discussion to our knowledge, this is the only published phase ii study evaluating the combination of gefitinib plus celecoxib in good performance status patients with advanced nsclc treated in the first line. our reported response rate with this combination was similar to the single agent activity of ge- fitinib alone in a similar patient population, and thus we failed to meet our primary goal and did not continue the study onto its second stage. – furthermore, this regimen seems less active than many combination chemotherapy regimens in a similar patient population. , , in addition, despite having a performance status of or at study entry, the median survival time of only . months is substantially lower than historical controls with chemotherapy as first line therapy ( – months) for this patient population on most recent studies. although five patients did achieve a partial response, including some with durable responses, it is unclear what the contribution of celecoxib was for these patients. other studies have combined egfr-tyrosine kinase inhibitors with disappointing results. gadgeel et al. had a response rate of % when combining gefitnib with celecoxib in patients with platinum refractory nsclc. o’byrne et al. combined gefitinib with rofecoxib in patients with platinum- pretreated relapsed nsclc resulting in a % response rate. another study combining erlotinib with celecoxib in patients with relapsed nsclc resulted in an % response rate. several studies evaluating single agent gefitinib in a chemonaive patients have been reported. niho et al. reported a response rate of % and median survival of . months with gefitinib as first-line treatment in advanced nsclc in an asian population. responders on this trial were generally female patients with adenocarcinoma, similar to observations on other studies, including the current study. , our data suggests that females who never smoked may be a patient population who would be best treated with an egfr inhibitor upfront. whether sequencing of therapy with an egfr in- hibitor as first-line therapy followed by chemotherapy second line would affect overall outcomes compared with the reverse order in this patient population remains unknown. such a question would be worthy of testing in a randomized trial. although four randomized trials failed to demonstrate a survival difference when treating patients with chemotherapy alone versus chemotherapy plus an egfr inhibitor, subset analyses on at least two of these trials suggested that in never smokers a survival advantage may still be realized with combined chemotherapy and egfr blockade. – this strat- egy is being tested in a randomized study by the calgb in which patients receive erlotinib alone versus chemotherapy plus erlotinib. in the second-line setting, several studies have com- bined a chemotherapy agent with a cox- inhibitor, suggest- ing some subsets may benefit from the addition of a cox- inhibitor. for example, although csiki et al. showed no difference in survival by combining celecoxib with docetaxel compared with docetaxel alone as second line therapy, their data suggested that patients with the greatest proportional decline in urinary pge-m levels experienced survival pro- longation when compared with those with no change or an increase in pge-m ( . versus . versus months). nugent et al. report in a phase ii study of docetaxel and celecoxib in previously treated patients with nsclc that the figure . kaplan-meier curve for overall survival. table . characteristics of responding patients sex age orig. description pks/yrr smoking status f lung adenocarcinoma never smoked f lung adenocarcinoma quit � mo ago but � yr f lung adenocarcinoma quit � mo ago but � yr f lung adenocarcinoma quit � mo ago but � yr f bone adenocarcinoma hasn’t smoked in � yr journal of thoracic oncology • volume , number , april gefitinib plus celecoxib in nsclc copyright © by the international association for the study of lung cancer addition of celecoxib may prolong time to disease progres- sion. gasparini et al. evaluated the combination of cele- coxib with weekly paclitaxel and measured circulating vas- cular endothelial growth factor and reported the subset of patients who responded to the combination of celecoxib and paclitaxel also showed decreased levels of serum vascular endothelial growth factor. pulmonary toxicity, in particular ild, remains a con- cern for a small percentage of patients treated with egfr inhibitors. the study by niho et al. reported a % rate of ild in their studied japanese population. other trials sug- gest much lower rates. , in the current study, of patients ( %) developed ild, both of whom died. there were only four patients whom experienced grade and toxicity of dyspnea. continued exploration of treating unselected patient populations with targeted agents such as gefitinib or cele- coxib are unlikely to significantly improve outcomes. in addition to the potential biomarkers mentioned to determine celecoxib sensitivity, several predictors of gefitinib efficacy have been reported. these include the presence of mutation in the egfr tyronine kinase domain, increased gene copy number of egfr band on fish analysis, lack of kras mutations, and lack of pten loss among others. – in conclusion, this study combining gefitinib with cele- coxib failed to demonstrate improved outcomes when com- pared with historical controls of chemotherapy in a similar patient population. as such, we do not recommend pursuit of this combination in an unselected patient population. the value of adding celecoxib to gefitinib remains in doubt, although worthy of further pursuit if a select patient popula- tion can be identified. acknowledgments support for this trial was provided by a grant from astrazeneca pharmaceuticals lp and pfizer inc. references . jemal a, siegel r, ward e, et al. cancer statistics, . ca cancer j clin ; : – . . non-small cell lung cancer collaborative group. chemotherapy in non- small cell lung cancer: a meta-analysis using updated data on individual patients from randomized clinical trials. br med j ; : – . . souquest pj, chauvin f, boissel jp, et al. polychemotherapy in ad- vanced non small cell lung cancer: a meta-analysis. lancet ; : – . . marino p, pampallona s, preatoni f, et al. chemotherapy vs. supportive care in advanced non-small cell lung cancer: results of a meta-analysis of the literature. chest ; : – . . grilli r, oxman a, julian j. chemotherapy for advanced non-small cell lung cancer: how much benefit is enough? j clin oncol ; : – . . stephens r, fairlamb d, gower n, et al. the big lung trial (blt): determining the value of cisplatin-based chemotherapy for all patients with non-small cell lung cancer (nsclc). preliminary results in the supportive care setting. proc am soc clin oncol ; : a (abstract # ). . schiller j, harrington d, belani cp, et al. comparison of four chemo- therapy regimens for advanced non-small cell lung cancer. n engl j med ; : – . . fukuoka m, yano s, giaccone g, et al. multi-institutional randomized phase ii trial of gefitinib for previously treated patients with advanced non-small cell lung cancer. j clin oncol ; : – . . kris mg, natale rb, herbst rs, et al. efficacy of gefitinib, an inhibitor of the epidermal growth factor receptor tyrosine kinase, in symptomatic patients with non-small cell cancer: a randomized trial. jama ; : – . . sweeney c, marshall m, barnard d, et al. cyclooxygenase expression in primary cancers of the lung and bladder compared to normal adjacent tissue. cancer detect prev ; : – . . hida t, kozaki k, ito h, et al. significant growth inhibition of human lung cancer cells both in vitro and in vivo by the combined use of a selective cyclooxygenase inhibitor, jte- and conventional anti- cancer agents. clin cancer res ; : – . . masferrer j, leahy k, koki a, et al. antiangiogenic and antitumor activities of cyclooxygenase inhibitors. cancer res ; : – . . qadri s, wang j, redmond k, et al. the role of cox- inhibitors in lung cancer. ann thorac surg ; : – . . howe l, subbaramaiah k, patel j, et al. celecoxib, a selective cyclo- oxygenase inhibitor, protects against human epidermal growth factor receptor (her- )/neu-induced breast cancer. cancer res ; : – . . niki t, kohno t, iba s, et al. frequent co-localization of cox- and laminin- gamma chains at the invasive front of early-stage lung adenocarcinomas. am j pathol ; : – . . kinoshita t, takahashi y, sakashita t, et al. growth stimulation and induction of epidermal growth factor receptor by overexpression of cyclooxygenase and in human colon carcinoma cells. biochem biophys acta ; : – . . coffey r, hawkey c, damstrup l, et al. epidermal growth factor receptor activation induces nuclear targeting of cyclooxygenase- , ba- solateral release of prostaglandins, and mitogenesis in polarizing colon cancer cells. proc natl acad sci ; : – . . solomon sd, mcmurray jjv, pfeffer ma, et al. cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. n engl j med ; : – . . niho s, kubota k, goto k, et al. first-line single agent treatment with gefitinib in patients with advanced non-small cell lung cancer: a phase ii study. j clin oncol ; : – . . kris m, sandler a, miller v, et al. cigarette smoking history predicts sensitivity to erlotinib: results of a phase ii trial in patients with bronchioalveolar carcinoma (bac). proc am soc clin oncol ; : s (abstract # ). . west h, franklin w, gumerlock p, et al. gefitinib (zd ) therapy for advanced bronchioloalveolar lung cancer (bac): southwest oncology group (swog) study so . j clin oncol ; : – . . johnson b, lucca j, rabin m, et al. preliminary results from a phase ii study of the epidermal growth factor receptor tyrosine kinase inhibitor erlotinib in patients � years of age with previously untreated ad- vanced non-small cell lung carcinoma. proc am soc clin oncol ; : s (abstract # ). . spigel d, hainsworth j, burkett e, et al. single-agent gefitinib in patients with untreated advanced non-small cell lung cancer and poor performance status: a minnie pearl cancer research network phase ii trial. clin lung cancer ; : – . . sandler a, nemunaitis j, denham c, et al. phase iii trial of gemcitabine plus cisplatin versus cisplatin alone in patients with locally advanced or metastatic non-small-cell lung cancer. j clin oncol ; : – . . kelly k, crowley j, bunn p, et al. randomized phase iii trial of paclitaxel plus carboplatin versus vinorelbine plus cisplatin in the treatment of patients with advanced non-small-cell lung cancer: a southwest oncology group trial. j clin oncol ; : – . . gadgeel sm, ruckdeschel jc, heath ei, et al. phase ii study of gefitinib, an epidermal growth factor receptor tyrosine kinase inhibitor (egfr- tki), and celecoxib, a cyclooxygenase- (cox- ) inhibitor, in patients with platinum refractory non-small cell lung cancer (nsclc). j thorac oncol ; : – . . o’byrne kj, danson s, dunlop d, et al. combination therapy with gefitinib and rofecoxib in patients with platinum-pretreated relapsed non-small cell lung cancer. j clin oncol ; : – . . fidler mj, argiris a, patel jd, et al. gastronintestinal hemorrhage in advanced non-small cell lung cancer (nsclc) patients treated with agarwala et al. journal of thoracic oncology • volume , number , april copyright © by the international association for the study of lung cancer erlotinib and celecoxib. j clin oncol ; : . asco meeting abstracts. . herbst r, prager d, hermann r, et al. tribute: a phase iii trial of erlotinib hydrochloride (osi- ) combined with carboplatin and pac- litaxel chemotherapy in advanced non-small-cell lung cancer. j clin oncol ; : – . . giaccone g, herbst r, manegold c, et al. gefitinib in combination with gemcitabine and cisplatin in advanced non-small-cell lung cancer: a phase iii trial—intact i. j clin oncol ; : – . . herbst r, giaccone g, schiller j, et al. gefitinib in combination with paclitaxel and carboplatin in advanced non-small-cell lung cancer: a phase iii trial—intact . j clin oncol ; : – . . gatzemeier u, pluzanska a, szczesna a, et al. phase iii study of erlotinib in combination with cisplatin and gemcitabine in advanced non-small-cell lung cancer: the tarceva lung cancer investigation trial. j clin oncol ; : – . . csiki i, morrow jd, sandler a, et al. targeting cyclooxygenase- in recurrent non-small cell lung cancer: a phase ii trial of celcoxib and docetaxel. clin cancer res ; : – . . nugent fw, mertens wc, graziano s, et al. docetaxel and cyclooxy- genase- inhibition with celecoxib for advanced non-small cell lung cancer progressing after platinum-based chemotherapy: a multicenter phase ii trial. cancer j ; : – . . gasparini g, meo s, comella g, et al. the combination of the selective cyclooygenase- inhibitor celecoxib with weekly paclitaxel is a safe and active second-line therapy for non-small cell lung cancer: a phase ii study with biological correlates. j thorac cardiovasc surg ; : – . . inoue a, saijo y, maemondo m, et al. severe acute interstitial pneu- monia and gefitinib. lancet ; : – . . lynch tj, bell dw, sordella r, et al. activating mutations in the epidermal growth factor receptor underlying responsiveness of non- small cell lung cancer to gefitinib. n engl j med ; : – . . hirsch fr, varella-garcia m, mccoy j, et al. increased epidermal growth factor receptor gene copy number detected by fluorescence in situ hybridization associates with increased sensitivity to gefitinib in patients wish bronchioalveolar carcinoma subtypes: a southwest on- cology group study. j clin oncol ; : – . . pao w, wang ty, riely gj, et al. kras mutations and primary resistance of lung adenocarcinomas to gefitinib or erlotinib. plos med ; :e . journal of thoracic oncology • volume , number , april gefitinib plus celecoxib in nsclc copyright © by the international association for the study of lung cancer 	gefitinib plus celecoxib in chemotherapy-naı¨ve patients with stage iiib/iv non-small cell lung cancer: a phase ii study from the hoosier oncology group 	patients and methods 	statistical considerations 	results 	toxicity 	efficacy 	discussion 	acknowledgments 	references journal of virology, oct. , p. - copyright © american society for microbiology vol. , no. prit.ted in u.s.a. biophysical, biological, and cytochemical features of a virus associated with transplantable hamster tumors w. a. stenback, g. l. van hoosier, jr., and j. j. trentin division of experimenztal biology, baylor unziversity college of medicinie, houstoni, texas received for publication june a virus resembling type c murine leukemia viruses, which is associated with transplantable hamster tumors, was partially characterized with respect to certain biological, biophysical, and cytochemical features. as determined by electron microscopy, high concentrations of the virus appeared in the blood plasmas of tumor-bearing hamsters. hamsters inoculated with virus concentrates did not show gross evidence of disease, and preliminary attempts to infect various hamster cells in tissue culture were unsuccessful. a line of cells from a virus-containing tumor which had been established in tissue culture released large numbers of the virus into the supernatant fluids by budding. the buoyant density peak of virus concentrates in potassium tartrate density gradients was . g/cm by ultraviolet absorption and electron microscopic analysis. acridine orange staining and nuclease digestion methods have indicated that the virus is probably a ribonucleic acid virus. previously, we reported the presence of virus particles in several transplantable hamster tumors as revealed by electron microscopy ( ). one of the two types of particles found resembles type c murine leukemia virus, the first report of such an agent in hamsters. one of our transplantable tumors (d- ), a malignant lymphoma, con- sistently contained many budding type c particles, as revealed by electron microscopy, and a second type particle which appears to be morphologically identical with the particle first described by bernhard and tournier ( ) and recently studied by thomas et al. ( ). blood plasma pellets from animals bearing this tumor contain large numbers of mature and immature type c particles but no "bernhard" type particles. the present report is concerned primarily with studies of the type c virus. materials and methods in vivo virus. groups of twenty-five - to -week-old hamsters received transplants of about mm of tumor subcutaneously by trocar. this tumor, a malig- nant lymphoma, grows rapidly and usually kills within to days. at about days, the bloods of the moribund animals were withdrawn by cardiac puncture into syringes rinsed with , units of heparin per ml, and were pooled in a chilled container in an ice bath. the blood cells were removed by low- speed centrifugation, and the plasma was stored at c presented in part before the american association for cancer research, atlantic city, n.j., april . until concentration of the virus was performed (within day). tissue-culture virus. cells obtained by trypsinization from a freshly excised d- tumor were placed in eagle's minimal essential medium supplemented with %c fetal calf serum in -oz ( . liter) bottles and were incubated under % co at c. after approxi- mately month, with minimal disturbance and feeding only once a week, a luxuriant growth of cells was evi- dent. these cells were subsequently serially passed at approximately - to -day intervals. thin sections of cell pellets and pellets of supernatant fluids showed large numbers of mature and immature type c par- ticles, many in the process of budding from the cyto- plasmic membranes. the "bernhard" type particles, present in the original tumor, were not evident in the cultured cells. virus concentrationz. virus-rich hamster plasmas or supernatant tissue culture fluids were centrifuged at , x g for min and were then filtered through a thin layer of celite ( ). the virus was then pelleted from the clarified fluids at , rev/min for hr in the no. rotor of the spinco model l ultracentrifuge or was collected on a cushion of potassium tartrate (density, . g/cm ) in the sw . rotor at , rev/min for hr; the supernatant fluids were dis- carded. density gradient centrifugation. virus pellets were gently resuspended in about . ml of single strength phosphate-buffered saline (pbs) with a potter-elve- hjem homogenizer ( ). the material was layered over linear( . to . g/ml) potassium tartrate gradients ( ) preformed with the aid of a layering device ( ), and was centrifuged at , rev/min for hr in the sw- rotor of a spinco model l ultracentrifuge. the o n a p ril , a t c a r n e g ie m e l l o n u n iv l ib r h ttp ://jvi.a sm .o rg / d o w n lo a d e d fro m http://jvi.asm.org/ stenback, van hoosier, and trentin gradient was then sampled either by collecting four- drop fractions after piercing the bottom of the tube or, for ultraviolet (uv) analysis, by utilizing a top-un- loading piercing unit (buchler instruments, inc., fort lee, n.j.). for further purification, the fluid above the opalescent band near the center of the gradient was removed into a syringe equipped with a -inch, - gauge needle bent at right angles. the band was then similarly removed with a fresh syringe and was sub- mitted to a second cycle in a potassium tartrate gradi- ent after dialysis against single strength pbs. densities were determined by weighing -/aliter samples (i ) or by measurement of the refractive index ( ). electronz microscopy. low-speed cell pellets and high-speed virus pellets were fixed in % phosphate- buffered glutaraldehyde for hr or, if necessary, were stored in this solution for several days at c before processing as previously described ( ). briefly, the pellets were postfixed in osmium tetroxide, dehy- drated in graded alcohols, and embedded in araldite resin. thin sections were double-stained with uranyl acetate and lead citrate and were examined in the siemens elmiskop ia electron microscope. cytochemical staininlg technique. the virus band from the second cycle of density gradient cen- trifugation was dialyzed against single strength pbs and then was pelleted at , rev/min for i hr. small droplets of the concentrated virus were placed on the center of small cover slips and were allowed to dry thoroughly in air. after fixation in carnoy's fixa- tive, acridine orange fluorochrome staining and nuclease digestion procedures were performed accord- ing to the methods of mayor ( ). fluorescence microscopy. the leitz ortholux fluorescence microscope equipped for dark-field illumination was used. for observation of acridine orange-stained virus materials, the bg- (heat) and bg- (blule) filters and the og- (orange) eye piece barrier filter were employed. results lack of serological cross-reactivity with known murine viruses. to help rule out the possibility that the agent might be a murine leukemia virus inadvertently introduced into our hamster colony, frozen, virus-rich tumor samples and sera from tumor-bearing hamsters and sera from normal control hamsters were sent to the labora- tories of janet hartley and w. rowe, national institute of allergy and infectious diseases, to whom we are indebted for testing these materials. no specific cross-reactivity was found in the complement-fixing (cf) test with their murine leukemia group-reactive rat tumor antigen and antiserum reagents ( ). attempts to demonstrate biological activity. high-speed pellets from the blood plasma of d- tumor-bearing hamsters contained large numbers of type c virus particles as revealed by electron microscopic examination of thin sections (fig. ). these particles consist of an electron-dense nucleoid surrounded by a loose membranous envelope. virus samples concentrated on a cushion of potassium tartrate and dialyzed over- night against pbs were inoculated into the follow- ing hamster cells in tissue culture: hamster embryo, hamster mixed spleen-thymus cells, adeno- induced tumor cells, and bhk- cells. no cytopathic effects (cpe) were seen, and thin sections of cell pellets examined by electron microscopy showed no evidence of type c viral replication, although the "bernhard" type parti- cles were observed in the bhk- cells, as re- ported by others ( , ), as well as in the spleen- thymus cells and adeno- tumor cells. super- natant fluid pellets from the above cells contained type c particles only on the first medium change, most probably representing the inoculum. newborn and weanling hamsters were inocu- lated with cell-free virus concentrated from tumor-bearing hamster plasma. tumors were found in two animals; because of the low in- cidence and advanced age of the animals at the time of autopsy, the relationship of the tumors to the inoculum is uncertain. membranous glomerulonephritis was also observed in the older animals coming to autopsy. the possible etiological relationship between the virus and the kidney lesions is under investigation. virus production in tissue culture. immature type c virus particles were seen budding from cytoplasmic membranes of the d- tumor cells in tissue culture (fig. ). these immature particles were observed at a frequency of about to per standard mesh electron microscope grid square, and the frequency remained the same between and days in culture. when tissue- cultured cells were inoculated into weanling hamsters, tumors were produced in four of four hamsters within days; these tumors also con- tained budding type c virus particles in thin sec- tions at a frequency of to per mesh grid square. the "bernhard" type particles were not evident in the d- cells in tissue culture nor in the tumors that we obtained from these cells. pellets of supernatant fluids from cell cultures contained many type c particles in thin sections (fig. ). equilibrium density gradient centrifugation. in potassium tartrate gradient centrifugation of virus-containing materials, two light-scattering bands were produced (fig. a). the upper band was a diffuse, opalescent band with a -nm absorption peak at a density of . g/cm , whereas the lower band was flocculent with a uv absorption peak at a density of . g/cm (fig. ). control materials contained only the lower band (fig. b). when the two bands were col- lected and pelleted for thin sectioning after j. virol. o n a p ril , a t c a r n e g ie m e l l o n u n iv l ib r h ttp ://jvi.a sm .o rg / d o w n lo a d e d fro m http://jvi.asm.org/ fig. . tlhin section throughl a high-speed pellet of pooled blood plasmas from tumor-bearing hamsters. many type c virus particles may be seen throughlout the section. the bar in all micrographs represents nm. ** s ! *j * .e x' as*' .: szr er a ..r. ~~~ . _i . t * ._ t w w ' . : * w~~~~~~~ s \ x, #ts ' s es . .a i ei fig. . immature type c virus particle budding from the cytoplasmic membrane of a d- tumor cell in tissue culture. .''l' ; " . . jf! . v.:+ . : ... *l .: : -,, j o n a p ril , a t c a r n e g ie m e l l o n u n iv l ib r h ttp ://jvi.a sm .o rg / d o w n lo a d e d fro m http://jvi.asm.org/ stenback, van hoosier, and trentin fig. . thini section through a high-speed pellet of -day supernataiit fluids from d- tumor cells growing in tissue culture. {l. virol. o n a p ril , a t c a r n e g ie m e l l o n u n iv l ib r h ttp ://jvi.a sm .o rg / d o w n lo a d e d fro m http://jvi.asm.org/ vol. , virus associated with transplantable hamster tumors dialysis against single strength pbs, the upper band (density, . g'cm ) was found to contain type c particles in large numbers (fig. ). no virus particles were associated with the lower l band, and the nature of its amorphous con- stituents is unknown. when the upper band was collected and submitted to a second cycle of density gradient centrifugation, a lower flocculent band was still produced but to a lesser degree. acridine orange staining. when fixed and stained with acridine orange, the partially purified virus concentrate appeared brilliant orange-red . under ihe fluorescence microscope. prior treat- ment with ribonuclease prevented the red fluo- rescence, whereas treatment with deoxyribonu- ecd clease or buffer alone had no effect. these results are consistent with a single-stranded ribonucleic acid (rna)-containing virus ( ). a discussion lack of a system for assaying biological activity has limited the extent to which the characteristics of this virus, which is associated with trans- plantable hamster tumors, can be studied. i i density . . fig. . photographic reproductionz ofa tracing made duiring uv absorptionz anialysis of a potassium tartrate dentsity gradient runz similar to that shiown in fig. a. the initial virus pellet was derived from tumor-bearing hamster plasma. the peak at the . g/cm denisity level conitainied the viruis. the . g/cm density peak consisted ofamorphouis material devoid of virus. nevertheless, some important characteristics have been determined by physical and cytochem- ical means. these characteristics include: (i) possession of a loose sac-like envelope, (ii) possession of high water content as evidenced by morphological alteration in concentrated potas- sium tartrate, (iii) particle formation by budding from the plasma membrane with a lack of cpe, and (iv) cytochemical evidence that the nucleic acid of this virus is rna. all of these charac- teristics are common to oncogenic rna viruses a b_ ( ). the buoyant density of . g/cm may be slightly less than that of known leukemia viruses ( ); however, this density was determined by fig. . distributiont of light-scatterinzg zontes of physical means rather than by assay of biological virus-conttaininlg anid conitrol materials int linlear potas- activity. possible destruction of biological ac- sium tartrate density graidientts. resuispended viruis pel- tivity by the potassium tartrate has not been lets were layered over preformed gradientts anid were centtrifuiged at , x g for hr in the sw- ruled out; however, there is no reason to assume rotoofa beckman spinco ultracentrifue. () vu. that brief exposure of this virus to the salt shouldrotor of a beckmaii spitico lultraceittrifltge. (a) virutsbeaymrdtienlthnscexoues band, derived from tissute ctultuire fluiids of d- tutmor be any more detrimental than such exposure is cells, is visible nzear the middle of the tutbe. (b) conitrol to other similar viruses ( , ). fluids from hamster embryo cells were devoid of virius. there is no evidence that murine agents j. .. .__ . .- j _., o n a p ril , a t c a r n e g ie m e l l o n u n iv l ib r h ttp ://jvi.a sm .o rg / d o w n lo a d e d fro m http://jvi.asm.org/ stienback, van hoosier, and trentin .j: .! t$.* * . *e ii ,. t ^, . gy' se* :' i l .. s ,... sf. ve ..... , ';.e, t' +:. 's- .~~~ ~ ~ frf ... tiv= i _s|-f- fig. . thlini sectioln through a high-speed pellet ofa . g/cm density band isolated from a potassiuim tartrate denisity gradielt. the pellet conisisted almost exclusively of type c virus particles. most of the particles were dis- torted as a resultt of osmotic effects of the contcenztrated potassium tartrate. replicate in hamsters. it has been reported that inoculation of the moloney agent into hamsters produces a reticulum cell sarcoma ( ); however, to the best of our knowledge, the agent was never recovered from the tumor or from subsequent transplants. the lack of cf cross-reactivity with murine agents in the present study suggests that the virus is a new hamster virus. the fact that the hamster harbors a virus which resembles murine leukemia agents is important in light of the extensive role played by the hamster in cancer research. the possibility has not been ruled out that the agent is merely a passenger in these tumors, but if the virus should prove to be causally related it would provide a useful new laboratory model for study. in studies currently in progress, we are attempt- ing to produce specific antisera for further serological studies, to apply radioisotope tech- niques to confirm the nucleic acid type, and to disclose biological activity which might be em- ployed for viral assay. acknowledgments we thank judith marston, department of micro- biology, baylor university college of medicine, for her time and the use of her equipment in our ultra- violet analyses. we also thank judy adams, carolyn gist, bernie powell, karen rhew, irene s. cox, and h. koo for excellent technical assistance. this investigation was supported by american cancer society grant in -h-project , and by public health service research grants ca- , ca- , and k ca- , from the national cancer institute. literature cited . anonymous. . suggestions for the classifica- tion of oncogenic rna viruses. j. natl. cancer inst. : - . . bernhard, w., and p. tournier. . infection virale inapparente de cellules de hamsters d cel e par la microscopie electronique. ann. inst. pasteur : - . . hartley, j. w., w. p. rowe, w. i. capps, and r. j. huebner. . complement fixation and tissue culture assays for mouse leukemia viruses. proc. natl. acad. sci. u.s. : - . . martin, r. g., and b. n. ames. . a method for determining the sedimentation behavior of enzymes: application to protein mixtures. j. biol. chem. : - . . mayor, h. d. . biophysical studies on viruses e. r*. j. virol. "l.a: 'o, ll. ip.- .:,i %, a w. ., i..'. n."... . . k ..: o n a p ril , a t c a r n e g ie m e l l o n u n iv l ib r h ttp ://jvi.a sm .o rg / d o w n lo a d e d fro m http://jvi.asm.org/ vol. , virus associated with transplantable hamster tumors using the fluorochrome acridine orange. progr. med. virol. : - . . mccrea. j. f., r. s. epstein, and w. h. barry. . use of potassium tartrate for equilibrium density-gradient centrifugation of animal vi- ruses. nature : - . . moloney, j. b. . the murine leukemias. fed- eration proc. : - . . o'connor, t. e., f. j. rauscher, and r. f. zeigel. . density gradient centrifugation of a murine leukemia virus. science : - . . potter, v., and c. a. eivehjen. . a modified method for the study of tissue oxidations. j. biol. chem. : - . . smith, k. o., and j. l. melnick. . adeno- virus-like particles from cancers induced by adenovirus- but free of infectious virus. science : - . . stenback, w. a., and d. p. durand. . host influence on the density of newcastle disease virus (ndv). virology : - . . stenback, w. a., g. l. van hoosier, jr., and j. j. trentin. . virus particles in hamster tumors as revealed by e'ectron microscopy. proc. soc. exptl. biol. med. : - . . thomas, j. a., e. delain, and r. hollande. . morphogenese d'un virus du hamster associe a la souche bhk ou a des tumeurs. compt. rend. : - . o n a p ril , a t c a r n e g ie m e l l o n u n iv l ib r h ttp ://jvi.a sm .o rg / d o w n lo a d e d fro m http://jvi.asm.org/ pemetrexed plus cetuximab in patients with recurrent non-small cell lung cancer (nsclc): a phase i/ii study from the hoosier oncology group original article pemetrexed plus cetuximab in patients with recurrent non-small cell lung cancer (nsclc) a phase i/ii study from the hoosier oncology group shadia jalal, md,* david waterhouse, md,† martin j. edelman, md,‡ sreenivasa nattam, md,§ rafat ansari, md,� karuna koneru, md,¶ romnee clark, md,* arthur richards, md,† jingwei wu, ms,# menggang yu, phd,# brian bottema, ccdm, bsba,** angela white, ms,** and nasser hanna, md* purpose: pemetrexed is a standard treatment against recurrent non-small cell lung cancer (nsclc), and cetuximab has single- agent activity against nsclc. this study evaluates the safety and efficacy of the combination of these agents in patients with advanced nsclc. patients and methods: patients with recurrent nsclc and an eastern cooperative oncology group performance status of to were entered. patients on the phase i portion of the study received cetuximab mg/m intravenously (iv) on day � followed by weekly doses of cetuximab at mg/m iv with escalating doses of pemetrexed every weeks (dose levels: , , , mg/m ) in a standard � design. once the maximum tolerated dose (mtd) of the combination was determined, patients were enrolled on the phase ii portion. the primary end point was to determine the median time to disease progression (ttp) (null hypothesis weeks, alternative hypothesis weeks). results: thirty-six patients were enrolled (phase i: n � , phase ii: n � ). patient characteristics included . % men, median age years (range, – years), . % had performance status and . % had adenocarcinoma histology. the median number of pre- vious regimens was (range, – ). the maximum tolerated dose of pemetrexed in combination with cetuximab was determined to be mg/m . the median ttp was . weeks. the median survival time was weeks and -year survival was . %. conclusion: the combination of pemetrexed at mg/m every days with weekly cetuximab at mg/m was feasible; however, in this unselected patient population, the combination regimen does not seem to improve ttp compared with historical controls of either single agent. key words: recurrent non-small cell, pemetrexed, cetuximab. (j thorac oncol. ; : – ) the treatment of patients with metastatic non-small celllung cancer (nsclc) remains challenging. although a majority of patients with a good performance status (ps) achieve temporary disease control with initial therapy, cancer progression is universal. the success of achieving subsequent disease control is diminished possibly because chemothera- py-resistant clones become a more dominant cell population within the tumor. there are only three agents approved by the food and drug administration for treating patients beyond first-line therapy. pemetrexed, docetaxel, and erlotinib are each approved in the second-line setting, and erlotinib is also approved in the third-line setting. each has modest single- agent activity (response rate, � %) and results in an im- provement in median survival of only to months. the initial phase i and ii studies of pemetrexed were conducted without vitamin b and folate supplementation and identified the maximum tolerated dose (mtd) to be to mg/m . , the addition of these vitamins results in signifi- cantly reduced toxicity without apparent loss of efficacy. , subsequent trials have identified the mtd of pemetrexed when given with vitamin supplementations to be to mg/m . , at the time of this study, the efficacy of escalated doses of pemetrexed against nsclc was unknown. cetuximab (c , erbitux, bms) is a chimeric monoclo- nal antibody that targets the epidermal growth factor receptor (egfr), which is overexpressed in most patients with nsclc. through blocking ligand binding to egfr, cetux- imab leads to a decrease in receptor dimerization, autophosphor- ylation, and activation of signaling pathways. in a phase ii trial, evaluating cetuximab in patients with relapsed nsclc, the *indiana university simon cancer center, indianapolis, indiana; †oncology hematology care, inc., cincinnati, ohio; ‡university of maryland medical center, greenebaum cancer center, baltimore, maryland; §ft. wayne oncology & hematology, inc., ft. wayne, indiana; ||northern indiana cancer research consortium, south bend, indiana; ¶cancer care center of s. indiana, bloomington, indiana; #indiana university school of medicine, division of biostatistics, indianapo- lis, indiana; and **hoosier oncology group, indianapolis, indiana. disclosure: dr. edelman: research funds, honororia, consulting from eli lilly and bms; dr. clark: former lilly employee with lilly stock; dr. hanna: consultant in eli lilly and has received honoraria and research support from eli lilly. the other authors declare no conflicts of interest. address for correspondence: shadia jalal, md, iu simon cancer center, barnhill drive, rt , in . e-mail: sjalal@iupui.edu copyright © by the international association for the study of lung cancer issn: - / / - journal of thoracic oncology • volume , number , november response rate was . %, stable disease was seen in . % of patients, and median survival time (mst) was . months. therefore, we conducted this trial to evaluate the feasibil- ity of combining higher doses of pemetrexed with cetuximab in patients with recurrent nsclc. once a mtd was identified, additional patients were enrolled to estimate whether the com- bination may be superior to the activity of each agent individu- ally in this therapeutically challenging patient population. patients and methods eligibility criteria included age more than or equal to years, histologic or cytologic diagnosis of nsclc (any histology), previous therapy with at least one platinum con- taining regimen for either locally advanced or metastatic disease, eastern cooperative oncology group ps of to , and measurable disease by recist group response criteria. adequate hematologic (absolute neutrophil count � / mm , platelets � , /mm ), renal (creatinine clearance � ml/min), and liver function at baseline (bilirubin � upper limit of normal, aspartate transaminase � . � upper limit of normal, alkaline phosphatase � � upper limit of normal) were required. patients with previous brain metasta- ses were allowed if they had been adequately treated for their brain metastases and were asymptomatic and off steroids. patients must not have received radiation or chemotherapy within weeks of registration or any investigational drugs within days of registration. patients were excluded if they had serious concomitant systemic disorders or history of uncontrolled cardiac disease. they were also excluded if they had received previous therapy with pemetrexed or cetuximab, were pregnant, or breast-feeding. all patients provided writ- ten informed consent before enrolling on the study. pretreatment evaluation included history and physical examination, assessment of eastern cooperative oncology group ps, complete blood count with differential, and chem- istry panel obtained within days of registration, and a computed tomography scan of chest and upper abdomen obtained within days of registration. brain and bone imaging was obtained only if clinically indicated. women of childbearing age must have a negative pregnancy test within days of registration. laboratory tests were repeated weekly (days , , and ), and disease evaluation by computed tomography was repeated every other cycle. treatment plan all patients received a loading dose of cetuximab of mg/m intravenously (iv) over minutes on day � followed by weekly doses of mg/m iv over minutes. patients were premedicated with mg iv of diphenhydra- mine hydrochloride (or an equivalent antihistamine) to minutes before the first dose of cetuximab. all patients received folic acid, vitamin b , and dexamethasone per the label of pemetrexed. pemetrexed was given starting day of each day cycle following cetuximab. patients were treated with up to six cycles of pemetrexed and cetuximab and in the absence of disease progression were allowed to remain on weekly cetux- imab until the time of progressive disease or toxicity. up to four dose levels of pemetrexed were to be evaluated ( , , , and mg/m ) in a standard � design. three patients were initially enrolled at dose level . if no patients experienced dose limiting toxicities (dlts), three patients were enrolled at the next dose level. if one of three patients at any dose level experienced a dlt, three additional patients were enrolled at that dose level. if only one of six patients experienced a dlt, dose escalation was permitted. if two or more patients at any dose level experi- enced a dlt, then the previous dose was considered the mtd. the dlts for the phase i trial would include any toxicity that required dose modifications or delays based on � grade toxicities including hematologic toxicity, mu- cositis, diarrhea, and rash. once the mtd of pemetrexed in combination with cetuximab was determined, patients were subsequently en- tered onto the phase ii portion of the trial and treated at the recommended dose. statistical analysis the primary objective of the phase i portion was to determine the mtd of the combination of pemetrexed and cetuximab in this patient population. secondary objectives were to characterize the dlts. the primary end point of the phase ii portion was to estimate the time to disease progression (ttp) of the com- bination. the secondary end points included estimation of clinical benefit rate (complete response plus partial response plus stable disease lasting at least days), mst and to further characterize the toxicity profile of this regimen. all phase i patients that were treated at the mtd were rolled over to the phase ii portion when determining the efficacy analysis. the median ttp using historical controls is approximately weeks. with a total number of patients, % power, and � � %, this design could detect a doubling of the median ttp to weeks. anticipating a % lost-to- follow-up rate, the sample size for the phase ii portion would be approximately patients. results between may , and may , , patients were enrolled onto the study (phase i: n � , phase ii: n � ). baseline characteristics of phase i and phase ii patients are summarized in table . in the phase i portion of the study, the dlts identified included acneform rash, alanine aminotransferase elevation, fe- brile neutropenia, dizziness, and dyspnea. the mtd for pem- etrexed was determined to be mg/m iv every days when combined with cetuximab at mg/m administered iv on a weekly basis. treatment administered the median number of cycles administered on the phase i portion was (range, – ). nine patients tolerated treatment without dose modifications or delays of either pemetrexed or cetuximab. three patients required dose mod- ifications or delays of either pemetrexed or cetuximab (two of pemetrexed and one of cetuximab) because of toxicity. median number of cycles administered on the phase ii portion was (range, – ). ten and patients tolerated treatment with pemetrexed and cetuximab, respectively, with- journal of thoracic oncology • volume , number , november pemetrexed plus cetuximab in patients with recurrent nsclc copyright © by the international association for the study of lung cancer out dose modifications or delays. a total of nine patients required dose modifications (five of pemetrexed and four of cetuximab) because of toxicity. three patients required dose modifications of pemetrexed because of intercurrent illness. three patients received maintenance cetuximab for , , and cycles, respectively, after cycles of the combina- tion as allowed per study schema. efficacy efficacy data is based on patients including six treated at the mtd of the phase i portion of the trial, and patients treated on the phase ii portion (two additional pa- tients enrolled on the phase ii portion were not considered in this evaluation as they withdrew consent before receiving any treatment). the median ttp was . weeks and month ttp rate was . % (figure ). the mst was weeks, and -year survival rate was . % (figure ). one complete response was observed in addition to two partial responses for an overall response rate of . %. fifteen additional patients achieved stable disease ( . %). the clinical benefit rate (defined as complete responses, partial responses and stable disease lasting � days) was . %. safety/toxicity phase i and ii toxicities occurring in % or more of patients are outlined in table , respectively. no anaphylactic or infusion reactions were reported with either agent. of note, hematologic toxicities were minimal. the most common toxicities were acne-like rash, fatigue, mucositis, and nausea. discussion the current study demonstrated the feasibility of combin- ing pemetrexed up to mg/m iv every weeks with cetuximb mg/m iv weekly in patients with recurrent nsclc. nevertheless, although the combination was generally well tolerated, efficacy outcomes were disappointing and failed to demonstrate improved ttp over historical controls with each single agent. to our knowledge, only one other study has evaluated the combination of cetuximab with chemother- table . baseline characteristics for phase i/ii patients characteristics phase i patients (n � ) phase ii patients (n � ) age, yr median range – – sex male ( ) ( ) female ( ) ( ) ecog ps ( ) ( . ) ( ) ( . ) stage of disease iii ( ) ( . ) iv ( ) ( . ) tumor histology adenocarcinoma ( ) ( . ) squamous cell carcinoma ( ) ( . ) large cell carcinoma ( ) ( ) non-small cell carcinoma (nos) ( ) ( . ) smoking history current ( ) ( . ) former ( ) ( . ) never ( ) ( . ) previous treatments chemotherapy ( ) ( ) chemotherapy ( ) ( . ) � chemotherapy ( ) ( . ) previous targeted therapy ( ) ( . ) previous radiotherapy ( ) ( . ) median time since prior regimen, d median time since initial diagnosis to registration, d . values in parenthesis are in percent. ecog ps, eastern cooperative oncology group performance status. time to death (weeks) s ur vi va l p ro ba bl ity . . . . . . median survival = . weeks % ci: . - . weeks figure . kaplan-meier curve for median sur- vival in weeks (n � ). jalal et al. journal of thoracic oncology • volume , number , november copyright © by the international association for the study of lung cancer apy in patients with previously treated patients with nsclc. kim et al. combined cetuximab with docetaxel in pa- tients with advanced nsclc who progressed or recurred within months of first-line chemotherapy. median ttp was days and median overall survival was . months. the modest efficacy outcomes of these trials suggest that the addition of cetuximab in an unselected patient population in this setting is unlikely to result in significantly superior outcomes to single-agent therapy alone. because the completion of this trial additional informa- tion has emerged regarding the use of pemetrexed. pem- etrexed is now approved for use only in patients with nonsquamous histology based on both prospective and retro- spective analyses from separate trials demonstrating signifi- cant superiority of pemetrexed in patients with nonsquamous histology. – second, two randomized trials have compared giving a standard dose of pemetrexed at mg/m versus a higher dose ( or mg/m ) in the second-line set- ting. , neither of these trials demonstrated any therapeutic advantage to dose escalation of pemetrexed. similarly, additional information has been reported on cetuximab in the treatment of nsclc. a phase iii trial randomized untreated patients with egfr detectable nsclc to cisplatin and vinorelbine with or without cetuximab. response rates were modestly improved with the addition of cetuximab and median survival improved . months (hr, . ; p � . ). similar benefits were seen in patients with squamous cell and adenocarcinoma. a smaller phase iii trial compared carboplatin with a taxane (docetaxel or paclitaxel) with and without cetuximab. response rates were slightly improved with cetuximab, but there was no difference in overall survival. recently hirsch et al. reported that increased egfr gene copy number detected by fluorescence in situ hybrid- ization analysis may be predictive of improved disease con- trol rate ( % versus %, p � . ) and median survival (p � . ) in patients with nsclc treated with cetuximab and chemotherapy. future studies will be evaluating cetux- imab-based regimens in an enriched patient population eval- uating for differential effects on egfr fish-positive and -negative patient populations. in addition, patients with met- astatic colorectal cancer were more likely to benefit from cetuximab combined with chemotherapy if their tumors ex- pressed kirsten rat sarcoma viral oncogene wild type com- pared with those with mutant type. similar efforts are underway to evaluate kirsten rat sarcoma viral oncogene as a molecular marker of cetuximab in nsclc. time to progression (weeks) p ro gr es si on p ro ba bi lit y . . . . . . median ttp = . weeks % ci: . - . weeks figure . kaplan-meier curve for median time to disease progression (ttp) in weeks (n � ). table . phase i/ii toxicity toxicity phase i (n � ) any grade (%) phase i (n � ) grade / (%) phase ii (n � ) any grade (%) phase ii (n � ) grade / (%) acne-like rash . . . alt elevation . . hematologic anemia . thrombocytopenia . . neutropenia . . . febrile neutropenia . . dizziness . . . dyspnea . . cough . fatigue . . anorexia . . mucositis . . nausea . . diarrhea . . . constipation . . . fever (no neutropenia) . . headache . hypomagnesemia . vomiting . . . alt, alanine aminotransferse. journal of thoracic oncology • volume , number , november pemetrexed plus cetuximab in patients with recurrent nsclc copyright © by the international association for the study of lung cancer in conclusion, our study failed to demonstrate improved efficacy outcomes with the combination of pemetrexed and cetuximab in previously treated patients with advanced nsclc. future studies will involve enriched patient populations who are most likely to benefit from these agents. references . hanna n, shepherd fa, fossella fv, et al. randomized phase iii trial of pemetrexed versus docetaxel in patients with non-small cell lung cancer previously treated with chemotherapy. j clin oncol ; : – . . shepherd fa, dancey j, ramlau r, et al. prospective randomized trial of docetaxel versus best supportive care in patients with non-small cell lung cancer previously treated with platinum based chemotherapy. j clin oncol ; : – . . shepherd fa, et al. erlotinib in previously treated non-small cell lung cancer. n engl j med ; : – . . rinaldi da, rodrigues pereira j, et al; national cancer institute of canada clinical trials group. initial phase i evaluation of the novel thymidylate synthase inhibitor, ly , using the modified con- tinual reassessment method for dose escalation. j clin oncol ; : – . . rinaldi da, kuhn jg, burris ha, et al. a phase i evaluation of multitargeted antifolate (mta, ly ) administered every days, utilizing the modified continual reassessment method for dose escala- tion. cancer chemother pharmacol ; : – . . niyikiza c, baker sd, seitz de, et al. homocysteine and methylmalonic acid: markers to predict and avoid toxicity from pemetrexed therapy. mol cancer ther ; : – . . scagliotti gv, shin dm, kindler hl, et al. phase ii study of pemetrexed with and without folic acid and vitamin b as front line therapy in malignant pleural mesothelioma. j clin oncol ; : – . . nakagawa k, kudoh s, matsui k, et al. a phase i study of pemetrexed (ly ) supplemented with folate and vitamin b in japanese patients with solid tumors. br j cancer ; : – . . takimoto ch, hammond-thelin la, latz je, et al. phase i and pharmacokinetic study of pemetrexed with high-dose folic acid supple- mentation or multivitamin supplementation in patients with locally advanced or metastatic cancer. clin cancer res ; : – . . lilenbaum rc. the evolving role of cetuximab in non-small cell lung cancer. clin cancer res ; ( suppl): s– s. . herbst rs, shin dm. monoclonal antibodies target epidermal growth factor receptive-positive tumors. cancer ; : – . . steiner p, joynes c, bassi r, et al. tumor growth inhibition with cetuximab and chemotherapy in non-small cell lung cancer xenografts expressing wild-type and mutated epidermal growth factor receptor. clin cancer res ; : – . . hanna n, lilenbaum r, ansari r, et al. phase ii trial of cetuximab in patients with previously treated non-small cell lung cancer. j clin oncol ; : – . . kim es, mauer am, william wn jr, et al. a phase study of cetuximab in combination with docetaxel in chemotherapy-refractory/ resistant patients with advanced non-small cell lung cancer. cancer ; : – . . scagliotti g, park k, patil s, et al. favorable benefit to risk profile for pemetrexed plus cisplatin versus gemcitabine plus cisplatin in a large phase iii study of first-line therapy in advanced non-small cell lung cancer. eur j cancer suppl ; : . . peterson p, park k, fossella f, et al. is pemetrexed more effective in patients with non-squamous histology? a retrospective analysis of a phase iii trial of pemetrexed versus docetaxel in previously treated patients with advanced non-small cell lung cancer (nsclc). eur j cancer suppl ; : – . . ciuleanu te, brodowicz t, belani cp, et al. maintenance pemetrexed plus best supportive care (bsc) versus placebo plus bsc: a phase iii study. j clin oncol ; :abstract . . cullen mh, zatloukal p, sörenson s, et al. a randomized phase iii trial comparing standard and high-dose pemetrexed as second line treatment in patients with locally advanced or metastatic non-small cell lung cancer. ann oncol ; : – . . ohe y, ichinose y, nakagawa k, et al. efficacy and safety of two doses of pemetrexed supplemented with folic acid and vitamin b in previ- ously treated patients with non-small cell lung cancer. clin cancer res ; : – . . pirker r, szczesna a, von pawel j, et al. flex: a randomized, multicenter, phase iii study of cetuximab in combination with cisplatin/ vinorelbine (cv) versus cv alone in the first-line treatment of patients with advanced non-small cell lung cancer (nsclc). j clin oncol ; :abstract . . lynch tj, patel t, dreisbach l, et al. a randomized multicenter phase iii study of cetuximab (erbitux) in combination with taxane/carbopla- tin versus taxane/carboplatin alone as first-line treatment for patients with advanced/metastatic non-small cell lung cancer (nsclc): b – . j thorac oncol ; (suppl ):s –s . . hirsch fr, herbst rs, olsen c, et al. increased egfr gene copy number detected by fluorescent in situ hybridization predicts outcome in non-small cell lung cancer patients treated with cetuximab and chemo- therapy. j clin oncol ; : – . . van cutsem e, lang i, d’haens g, et al. kras status and efficacy in the first-line treatment of patients with metastatic colorectal cancer (mcrc) treated with folfiri with or without cetuximab: the crystal expe- rience. j clin oncol ; :abstract . jalal et al. journal of thoracic oncology • volume , number , november copyright © by the international association for the study of lung cancer 	pemetrexed plus cetuximab in patients with recurrent non-small cell lung cancer (nsclc) 	patients and methods 	treatment plan 	statistical analysis 	results 	treatment administered 	efficacy 	safety/toxicity 	discussion 	references fr; feb social science computer review schweik, green / spectral mixture analysis the use of spectral mixture analysis to study human incentives, actions, and environmental outcomes charles m. schweik indiana university university of massachusetts glen m. green indiana university the social science community has given increasing attention to the analysis of multitemporal satellite images in human dimensions of environmental change research. this article provides an overview of how image-processing techniques, such as radiometric calibration and spectral mixture analysis, can be applied. the application of these techniques is vital if the social science community wishes to develop a robust research program that allows the accurate comparison of distant but similar locations. furthermore, by mapping the institutional landscape, the article demonstrates how social scientists can begin to understand landscape change by carefully relating human incentives to actions and actions to outcomes as measured by a set of multitemporal satellite images. keywords: global change, satellite images, institutional analysis, spectral mixture analysis, image restoration, institutional landscape introduction environmental policy makers, public officials involved in land management, and researchers studying the human dimensions of environmental change all hope to answer a particular question, each from their particular perspectives: in what ways is the landscape changing, and how are humans influencing those changes? robust answers to this question are critical if we are to (a) understand the broad role humans play in the realm of global change as well as (b) guide officials at local and regional levels who are making real-life land use planning decisions and prioritizing the use of scarce public funds in land management. the national aeronautics and space administration’s landsat, or land monitoring sat- ellite system, has amassed an immense longitudinal archive of multispectral images for every location in the united states since its initiation in . these images provide a wealth of information characterizing the country’s forest resources as well as identifying where land cover is undergoing change. however, monitoring land cover using landsat images has only just recently begun to significantly penetrate disciplines beyond the physical sciences. most satellite-based case studies involve land-cover inventories, or classifications, at one point in authors’ note: support from the national science foundation (grant no. sbr- - ) is gratefully acknowledged. we are grateful for the exceptional editing skills provided by patty dalecki at the workshop in political theory and policy analysis, indiana university, bloomington. social science computer review, vol. no. , spring - © sage publications, inc. time where the satellite data associated with each picture element, or pixel, of the image is assigned to one land-cover class. programs such as the national wetlands inventory and state-level gap analysis provide examples of such inventory programs (igap, ; scott et al., ). these studies typically employ classification schemes based on variations in satellite-derived digital numbers, or dn values, which, unlike reflectance values, are not physical units and thus have limited utility for comparison outside the specific image. there- fore, a given classification from one image cannot be easily applied to another image at a dif- ferent location or time. recent technological advances now enable researchers, natural resource policy makers, and land managers to move beyond these traditional satellite image-based cross-sectional land inventories. gigabyte storage volumes, pentium-level or higher processors, cheaper random access memory, and more accessible geographic information system (gis) and remote-sensing software allow even users of desktop pc platforms to conduct satellite-based image change analysis. moreover, by first calibrating the image data so that picture element (pixel) values represent surface reflectance and then applying a relatively new remote- sensing analysis technique called spectral mixture analysis (sma), the policy analyst or social scientist can move beyond a statistically relative and somewhat abstract land classifi- cation to one based on the physical measures of surface reflectance. furthermore, sma pro- vides the ability to identify subpixel measures such as the percentage of land-cover compo- nents (e.g., soil, vegetation, water) that make up a pixel. these advances are important for two reasons: first, sma’s subpixel analysis allows the researcher to use more effectively the older landsat multispectral scanner data with coarse spatial resolution ( × square meters, roughly the size of a u.s. football field). pixels of this size will rarely be exclusively one land-cover type and are more likely to be composed of mixtures of different types of land cover. subpixel information is vital when working at this spatial resolution. second, by radiometrically calibrating images to surface reflectance val- ues and applying sma, the analyst can develop land-cover classification schemes that are more apt to be replicated and directly applied to other landsat images of comparable land- cover regions or ecosystems. such broad-based classifications can be used to develop land- scape change maps, which in turn can be linked with other geographic parameters known to drive human decision making and action. efforts such as these are critical as we continue to address issues of scaling in global change research as well as make more practical policy and land management decisions. this article provides an overview on how sma might be applied to understanding land- cover change in a forested region and how this change can be related to human activities. the first part provides a discussion on the utility of the landsat archive for this research agenda. part raises some open questions and discusses in more detail the problems related to stan- dard image classification procedures as they are currently used. part presents a more tech- nical discussion on traditional analytic techniques and then provides a description of how sma works. part presents a brief example of one recent application of sma in the context of public forest management in indiana. it shows how we can begin to develop longitudinal studies of land-cover change based on surface reflectance values that can be replicated by other researchers using other landsat images in comparable locations. it also presents two examples of how these analyses can be linked to human incentives and actions. the article concludes with a summary of the importance of this for the human dimensions of environ- mental change and global change research programs as well as for policy makers and land managers. schweik, green / spectral mixture analysis the utility of landsat mss and tm images for the study of the human dimensions of environmental change the landsat system, designed in the s and first launched in the early s, was specifically developed for the observation of earth’s terrestrial surface (campbell, , p. ). tables and provide a summary of the landsat system. since , six landsat satellites or platforms have been launched. landsat continues to be operational, working well beyond its intended life span. two primary sensor instruments exist on these platforms: the multispectral scanner (mss) and thematic mapper (tm). mss instruments were employed on landsats through . tm instruments were first included on landsat , remain on landsat , and an enhanced version will be launched on landsat . there are several justi- fications for the use of mss and tm imagery in studies concerning the human dimensions of environmental change, and many of these reasons are directly associated with scale issues discussed above. first, both the mss and tm data archives cover a broad spatial extent. each individual scene acquired by both landsat sensors covers a “footprint” on the earth approximately kilometers wide. furthermore, since landsat ’s launch in , all of the earth’s terrestrial surface between º n and º s latitude has been subject to image acquisition (campbell, , p. ). second, the landsat data archive covers a relatively long temporal duration (more than years). mss technology began in ; tm technology began years later in . although this cumulative period may be short in terms of the history of humanity, the time period covers a long enough temporal range to capture many of the human-induced changes that have occurred in forested landscapes. this is especially true in locations where land- cover change is rapidly occurring at present, such as in developing countries where tropical forests are being converted to agriculture or grazing land. third, the mss and landsat tm instruments provide a reasonably high degree of spectral resolution when compared to other remote-sensing platforms such as aerial photographs, france’s spot, or india’s irs satellites. the mss has four sensors, or bands, each sensitive to a different portion of the electromagnetic spectrum. as table shows, mss band responds to light at visible green wavelengths ( . - . micrometers, or µm), band responds to light in the visible red portion of the spectrum ( . - . µm), and both band and band to different portions of near-infrared wavelengths ( . - . µm and . - . µm, respectively). landsat tm data provide better spectral resolution, employing seven bands with sensors collecting additional data from the mid-infrared and thermal regions of the elec- tromagnetic spectrum. although other technologies such as the national oceanic and atmospheric administra- tion’s advanced high-resolution radiometer or hyperspectral instruments such as the air- borne visible/infrared imaging spectrometer provide even better spectral resolution than landsat data do (campbell, ; verbyla, ), the mss instrument probably provides adequate data for addressing many of the questions being asked in the human dimensions of global change research program. many of the activities humans undertake on the physical landscape are dramatic, involving vegetation-to-soil or soil-to-vegetation conversions. these types of changes provide a clear contrast in land-cover reflectance spectra (see figure ) that can be adequately captured using mss band passes and most certainly with tm spectral resolution. fourth, the temporal sampling of the landsat mss and tm systems are relatively high. location coverage repetition frequency was days for landsats , , and and is days social science computer review ta b l e l a n d s a t s a te ll it e in s tr u m e n t s p e c if ic a ti o n s s a te lli te p la tfo rm l a n d sa t : l a n d sa t : l a n d sa t : l a n d sa t : l a n d sa t : l a n d sa t : l a n d sa t : l a u n ch e d l a u n ch e d l a u n ch e d l a u n ch e d l a u n ch e d (f a ile d to (e xp e ct e d / / ; / / ; / / ; / / / / ; re a ch o rb it) la u n ch d a te , o u t o f s e rv ic e o u t o f s e rv ic e o u t o f s e rv ic e o p e ra tio n a l o p e ra tio n a l o u t o f s e rv ic e m id - ) in st ru m e n t / / / / / / (s ta n d b y m o d e ) m u lti sp e ct ra ls ca n n e r (m s s ) x x x x x — t h e m a tic m a p p e r (t m ) x x — x (e n h a n ce d t m + ) s o u r c e :a d a p te d fr o m c a m p b e ll ( ). ta b l e s u m m a ry o f in s tr u m e n t s e n s o rs in st ru m e n t b a n d s p e ct ra ls e n si tiv ity (m ic ro m e te rs ) r e so lu tio n (s q u a re m e te rs ) m u lti sp e ct ra ls ca n n e r (m s s )a . - . (g re e n ) × . - . (r e d ) × . - . (n e a r- in fr a re d ) × . - . (n e a r- in fr a re d ) × t h e m a tic m a p p e r (t m ) . - . (b lu e ) . × . . - . (g re e n ) . × . . - . (r e d ) . × . . - . (n e a r- in fr a re d ) . × . . - . (m id -i n fr a re d ) . × . . - . (t h e rm a l) . × . . - . (m id -i n fr a re d ) . × . e n h a n ce d t m + s a m e a s t m , e xc e p t: o p tic a lb a n d s . × . t h e rm a l × p a n ch ro m a tic × a .i n l a n d sa t sa te lli te p la tf o rm s , a n d , m s s b a n d s , , a n d w e re re fe rr e d to a s b a n d s , , , a n d re sp e ct iv e ly . for landsats and . this revisit frequency is probably adequate for addressing many of the most important human-induced processes related to landscape change. although cloud cover may restrict acquisitions for any given location within these latitudes, there is likely a large temporal archive of good quality still available for sampling. in our south-central indi- ana study region alone, we have identified approximately landsat mss and tm images with low cloud cover taken during the to time period. open questions related to the utility of landsat mss images for human dimensions of environmental change studies as the previous discussion indicates, there are ample reasons to consider the use of land- sat mss and tm for studies relating human decision making and action to land-cover change. tm images are certainly appropriate given their high spatial and radiometric resolu- tions. however, as noted above, tm technology was first introduced in , and about years of inventory exists. but if we are concerned about the temporal grain of the phenomena of interest (such as forest cutting) and are concerned that a -year sampling period may not capture a specific human activity adequately, we may wish to expand the temporal duration of the image data set we use. by employing landsat mss images, we extend the temporal range by another years back to . for many change analyses, this added decade may be vital. however, there is one significant limitation in applying mss data for these types of studies: the relatively poor spatial resolution provided. other satellite platforms described earlier, such as landsat tm and spot, provide rela- tively fine spatial resolution with . × . meter and × meter spatial resolutions, respectively (campbell, ). an mss pixel is coarser, capturing roughly a × meter area on the ground. this coarse spatial resolution limits the amount of information related to human activities that mss images can provide (wilkie & finn, ). we therefore face a tradeoff. it is clear that for these types of questions, images with a higher spatial resolution are favorable, such as ones created by landsat tm or spot, but mss adds an important years onto the time series. mss images are also much less expen- sive than tm images. from this perspective, the use of landsat mss is certainly helpful, social science computer review figure : typical continuous spectral reflectance curves for vegetation, soil, and water and a hypothetical mixture spectra source: adapted from richards ( ). economically efficient, and could also be absolutely critical to adequately capture the tempo- ral frequency of certain land-cover change phenomena. traditional analysis techniques and the promise of spectral mixture techniques the use of another technique, first developed by geologists and planetary scientists, called spectral mixture analysis, or sma (adams, smith, & gillespie, ; adams, smith, & johnson, ; smith, ustin, adams, & gillespie, a, b), has the potential to make landsat mss-based (and tm-based) analyses more applicable to the study of fine spa- tial scale human-induced land-cover disturbance. sma, also referred to as linear unmixing (envi, ), allows the analyst to move to a subpixel level of analysis and works with sur- face reflectance values, not image dns (see below). the following section provides a brief summary of the more traditional image-analysis techniques and then describes sma and how it might be used to improve the utility of landsat mss and tm image analysis for stud- ies related to human dimensions. traditional image-analysis techniques image-analysis techniques in this context refer to methods for displaying and interpreting band-to-band variations of multispectral satellite images. generally, traditional image analysis involves a suite of tools including single-band analysis, color composite generation, band-to-band ratioing and vegetation indices, principal component analysis (pca), and classification. these methods are widely used in analyzing landsat images (adams et al., ). for a more detailed description of each technique, see any standard remote-sensing textbook (e.g., campbell, ). only a brief synopsis will be included here. single-band analysis involves the review of individual bands of an image separately, using a gray-scale display. color compositing techniques superimpose three bands together, displaying each band of information using the three colors: blue, green, and red. for instance, for mss data, the visible green band is often displayed using blue, the visible red band is displayed as green, and the second infrared band is displayed as red. this simulates what would be generated if the image were taken using color infrared film (lillesand & kiefer, ). one limitation in using color composites is that only three channels of the information from a satellite image can be displayed at one time (adams et al., ). band-to-band ratioing is another analysis technique that uses multispectral satellite images for the study of land cover. vegetation indices are one group of methods that involve popular band-to-band ratios. vegetation chlorophyll is known to absorb visible light, whereas a leaf’s mesophyll tissue strongly reflects near-infrared light (campbell, ). as the amount of vegetation increases within the boundaries of a pixel, surface reflectance in the visible red decreases, whereas reflection in the near-infrared increases (hall, ). vegeta- tion indices, such as the normalized difference vegetation index (ndvi), take advantage of these properties. there are a number of indices that have been developed, but they all involve mathematical ratios or differences of the digital values of different spectral bands to produce a single value for each pixel. ndvi and other such ratios are useful in that they can provide a measure of the vegetation content within each pixel. by employing the characteristic spectral response of vegetation, these indices allow the analyst to develop a measure of one component of each pixel. there are a variety of studies that have used ratios as part of their analyses (e.g., cohen, ; sader & wynne, ; tucker, vanpraet, sharman, & van ittersum, ). one of the limitations of schweik, green / spectral mixture analysis ratios is that they generally do not take advantage of all the available information contained in multispectral images, usually using only two bands (adams et al., ). also, these indices may be influenced by many factors not associated with vegetation itself (e.g., soil background and sensor differences) (campbell, ). these ratios are also sensitive to the particular atmospheric conditions on the day of acquisition because the atmosphere can cause light in the visible portions of the spectrum to be preferentially scat- tered. many analyses may have employed vegetation ratios without removing these effects, which may lead to false conclusions (robinove, ). atmospheric correction is advised to remove the effects of atmospheric scattering (teillet & fedosejevs, ). others emphasize the importance of radiometric calibration, together with atmospheric correction, to convert the image from raw dn values to surface reflectance before calculating these ratios to mini- mize sensor drift problems and other factors (e.g., green, schweik, & hanson, ; price, ). unfortunately, the procedures required to remove much of the variance caused by nonvegetation effects are logistically difficult and not readily accessible to many researchers outside the remote sensing community. this topic will be discussed more fully later. pca is often used to remove interband correlation that typically exists within multispec- tral image data. pca identifies linear combinations of the original band data of an image to produce component images representing the axes of maximum variation (campbell, ). often the purpose of pca is to compress most of the information contained in a multiple- band image into a new image with fewer components. these product images are then used in place of the original band data (lillesand & kiefer, ). this procedure is sometimes help- ful in preprocessing the data to remove instrument noise such as image striping (idrissi, ). some scholars are investigating the use of the higher components (e.g., component and higher in tm images) to identify patterns in human activity (b. l. turner ii, personal communication, february , ). one benefit of pca is that it uses all the information provided in a multispectral image. classification methods are one of the most common analytic techniques applied to multispectral images (adams et al., ). this becomes quite apparent when one traverses the remote-sensing literature of the s, s, and s. most texts related to remote sensing devote significant effort (sometimes the majority of the discussion of analysis) to image classification procedures (see, e.g., campbell, ; lillesand & kiefer, ; ver- byla, ; wilkie & finn, ). using a classification algorithm, these methods assign individual pixels of a multispectral image to discrete categories (drury, ). the goal of this type of analysis is to greatly simplify continuous image data ( bands of -bit data for tm) using quantitative techniques for identification of spectrally similar land-cover classes within the image (lillesand & kiefer, ). two types of classification procedures are popular and work in conjunction with one another. unsupervised classification procedures group or cluster the multispectral values of the image into distinct classes (e.g., water, soil, and vegetation) based solely on the image statistics and produce a new raster (grid) map displaying the class designations within the image (hall, ). supervised classification involves the use of ground inventories to guide, or “train,” the computer to assign appropriate clusters of data to certain land-cover classes (hall, ). various algorithm strategies (e.g., maximum likelihood) have been employed to classify pixels not assigned to training areas (campbell, ). many important human dimensions of environmental change studies use image classification as their domi- nant image-analysis method (to name a few: brondizio, moran, mausel, & wu, ; lee & marsh, ; moran, brondizio, & mausel, ; sader, ; westman, strong, & wilcox, ). others have combined analytic strategies by employing both classification and band- to-band ratios (e.g., lee & marsh, ; wolter, mladenoff, host, & crow, ). social science computer review one great advantage of classification methods is that they use all the information (bands) provided by multispectral images. however, there are two great limitations of classification procedures. first, only one land-cover category or class can be assigned to each pixel or cell of an image matrix. yet, in many instances, the land represented by one × meter mss pixel will be a mixture of multiple land-cover types. second, most classifications employed are usually image dependent. typically they are based on raw dns and have not been con- verted to physical measures such as surface reflectance values (see the below discussion on image restoration). nonsurface sources of variability, such as sensor, illumination, and atmospheric differences, remain within the image values and may influence the way the data are classified. furthermore, because of scene dependence in classification, an expansion of geographic coverage within one individual scene may reveal a new land-cover type that may then alter classification groupings. for these reasons, classifications, based on satellite dns without image restoration, may not be transportable to other geographic locations. spectral mixture analysis efforts have been made to address these limitations in classification techniques (e.g., jackson, ). sma is a promising technique developed from the efforts of earth and plane- tary scientists (adams & adams, ; adams et al., ; bryan, finger, & chayes, ; kornder & carpenter, ; smith et al., a, b; ustin, smith, & adams, ). certain land-cover types (e.g., soil, water, and vegetation) have been shown to exhibit characteristic patterns of reflectance with wavelengths across the electromagnetic spectrum. figure shows typical reflectance spectra of general land cover derived from landsat image data, sometimes referred to as stick spectra. some landscapes may exhibit discrete and homogeneous land-cover types such as a lake. in instances such as these, pixel data may exhibit more “pure” forms of spectral responses (e.g., % water), and classification is an appropriate analytic technique. in reality, however, land-cover surfaces are often composed of a variety of mixes of mate- rials, and the distinction between them (especially from space) is not so clear. when a sensor observes a pixel composed of mixed types of surfaces, the reflectance spectra produced will not match any pure spectra like the examples shown in figure . rather, the pixel spectral response will exhibit a shape over the electromagnetic wavelength that is some mixture of spectra of those land-cover types present (campbell, ). for example, a pixel composed of both soil and vegetation will have a spectral response that is some combination of the gen- eral soil and vegetation spectra (see the hypothetical mixture example shown in figure ). the objective of sma is to identify primary spectral contributions within each pixel (adams et al., , p. ). it provides a means for determining relative abundances of land-cover materials present in any given pixel based on the spectral characteristic of the material (envi, ). sma involves two main steps. the first step is to define a set of pure spectra for selected land-cover material, often referred to as endmembers. endmembers can be identified using either (a) libraries of known spectra collected with a spectrometer in the field or in a labora- tory, (b) libraries of known spectra from previous sma studies, or (c) spectrally pure or “extreme” pixels identified within the images being analyzed. most applications of sma will use the third option because libraries of field-collected endmember spectra are rare, and field spectrometers are expensive and not readily available to researchers. the empirical por- tion of this article uses this third option. sma endmembers are selected by testing whether a given pixel’s spectra can be described as a mixture of pixel spectra previously analyzed or whether it is itself another end- member (adams et al., ). a simple hypothetical illustration is provided in figure . schweik, green / spectral mixture analysis figure a shows the spectra for vegetation, soil, and shade. the vertical gray bars repre- sent the wavelength response for bands and . figure b plots the reflectance values of various mixed pixels from an image for band and band . pixels composed of entirely one component—vegetation, soil, or shade—would plot at the corners of the triangle presented in figure b. pixel purity functions help to identify those pixels in an image that lie at the extremes of the pixel value distribution in n-dimensional space (envi, ). the number of endmembers chosen must be one less than the number of bands available in the multispec- tral image (envi, ). for mss images with four bands, this means that up to three end- members can be identified. for tm images, six endmembers can be identified if all seven tm bands are used; however, because landsat tm’s thermal band represents a fundamen- tally different wavelength, sma analyses typically use only the six optical tm bands and therefore five endmembers (see, e.g., adams et al., ). once endmembers are identified, their reflectance values associated with image bands can be documented, presented in a form similar to figure a, and stored in a spectral library for use in this and future studies. the second step in sma is to estimate, for each pixel, the abundance of each endmember contained within it by applying a linear mixing equation (adams et al., ; envi, ). the general form of this equation, in matrix form, is as follows: r [mxl] = em[mxn] * x [nxl] + e[mxl] where, r = original image values calibrated to surface reflectance, em = endmember spectra matrix, x = vector of unknown abundances or fractions, m = number of image bands used, n = number of endmembers used (must be m – or less), and e = root mean-squared (rms) error in the fit of the model. it represents the sum of the squared residuals over all m bands. this sma equation is used to convert the existing image spectra values for each pixel into endmember fraction matrices. one fraction image is produced for each endmember along with the rms error matrix. this procedure requires the fraction values produced in matrix x to be positive and sum to unity (adams et al., ; envi, ). these fraction images can be created for each time point in a multitemporal set of satellite images all using the same endmember spectra. social science computer review figure : graphic representing hypothetical image data for spectral mixture analysis source: adapted from adams, smith, and gillespie ( , p. ). to date, very few studies have applied sma toward questions related to the human dimensions of environmental change. but it has great potential over the more traditional image-analysis techniques for several reasons. first, like classification and pca, sma uses all the information (bands) of a multispectral image. second, unlike many traditional classification methods, sma is image independent. sma endmembers can be identified from image data, field, or lab inventories or from end- member fraction libraries (bateson & curtiss, ). for this reason, time series or multiple geographic location sma fraction coverages are more readily comparable than are products from classification products based on dns. third, endmember fraction coverages represent physical properties of the landscape based on physical units: surface reflectance values. classification maps often represent abstract human language artifacts—forest or pasture, for example—that may have different meanings to different people and are not directly tied to any measurable physical property. two areas may be called a pasture but may be composed of very different vegetation types that exhibit very different spectral qualities. analyses conducted using endmembers are based on specific spectral qualities of identifiable materials and not on human abstractions. they can therefore be more easily applied to other geographic areas. fourth, endmember fraction images are more appropriate for analysis of physical land- scapes exhibiting a high degree of continuously varying land cover, such as most forested areas. this is especially important for studies of more continuous landscapes where changes in light reflectance are gradational, such as heavily forested areas. although classification is quite useful in physical landscapes with discrete land forms, it is not as easily applied to land cover that is more continuous in nature and less easily distinguishable. fifth, and perhaps most important, sma provides the spectral data in terms of multiple endmember fraction coverages and not as a single pixel classification, hence allowing a more detailed analysis of pixel contents (adams et al., ). because of this, some refer to the mixing approach as a subpixel analysis (envi, ). from this we can get a better idea of what mixtures of land-cover features comprise any individual pixel. the same amount of data is used in sma and in classification techniques, but the employment of sma endmem- bers and the production of endmember fraction images allow for a more detailed analysis of individual pixel contents. this is important even for satellite images with relatively good spatial resolution, such as landsat tm, but becomes especially important when using coarse resolution instruments such as landsat mss. the importance of image restoration procedures given these justifications for sma and that sma has been around for more than a decade (e.g., adams & adams, ), why has it not been more widely used? there are two answers to this question. sma requires the identification of pure reflectance spectra from the image itself or by using spectra libraries (adams et al., ). the individual band dns provided in the raw images received from u.s. repositories, such as eros and space imaging/eosat, do not represent earth surface reflectance values that can be immediately used. more than years ago, robinove ( ) recognized and nicely summarized the problem that still is prevalent today: landsat digital images are commonly analyzed by using the digital numbers for each pixel recorded on computer-compatible magnetic tape. although this procedure may be satisfactory when only a single, internally consistent image is used, the procedure may produce incorrect schweik, green / spectral mixture analysis results if more than one image is used for analysis as in mosaics or temporal overlays. the digital numbers . . . vary depending on the calibration of the multispectral scanner in each satellite at a given time, the sun angle, the state of the atmosphere, the slope and aspect of the terrain, and sur- face cover. (p. ) to develop reflectance spectra of endmembers from the image itself, or use an endmember reference from a library, the image must have sensor, illumination, and atmospheric variation removed from the original image dns. in other words, the dns must be converted to surface reflectance. as robinove suggests, this conversion, sometimes referred to as image restora- tion (green et al., ; jensen, ), is vital for any sma temporal or spatio-temporal comparison. sma has not been used because it is logistically difficult to undertake full image restora- tion. when image comparisons are to be made, after preprocessing steps, three main restora- tion processes should be undertaken: radiometric calibration, atmospheric correction (chavez, ; teillet & fedosejevs, ), and radiometric rectification (hall, strebel, nickeson, & goetz, ). radiometric calibration involves converting image dns to radi- ance values and radiance to at-satellite reflectance values (hill, ; markham & barker, ; robinove, ). atmospheric correction removes the effects of the atmosphere (e.g., light scattering) thereby converting at-satellite reflectance values to surface reflec- tance (ballew, ; chavez, ; teillet & fedosejevs, ). radiometric rectification normalizes a set of time-series image data to one given image. this ensures that time-series images are directly comparable to one another. although the information on how to conduct radiometric calibration is available, the lit- erature is scattered and not readily available to the environmental change research commu- nity (green et al., ). the procedures required differ by satellite platform and by sensor type. a literature review of major remote-sensing journals revealed that over the past years only a small proportion of researchers actually converted dns to surface reflectance values (green et al., ). classification of one time point or multiple time points (by apply- ing the radiometric rectification procedures of hall et al., ) can be conducted as long as the researcher has no interest in applying the same classification to another geographic area (green et al., ). because of this ability to normalize images taken at different times, but of the same location, the technical difficulties in converting dns to surface reflectance val- ues may appear to many to be not worth the effort. this decision to leave landsat images in dn data space means that sma cannot be considered a viable analysis option on these prod- ucts. over the past years, we have worked to produce partially automated spreadsheets and accompanying documentation that allow the user to convert any landsat tm or mss image from dn representation to surface reflectance values (for more information on the availabil- ity of these procedures, contact either one of the authors). the second answer to why sma is not more widely used is more simple: sma computer programs are largely unavailable. sma itself is a fairly complicated process (see adams et al., ) and is largely prohibitive unless software is written to allow such an analysis to be undertaken. at the point of this writing, surprisingly, sma functionality is still not avail- able as part of the general analysis tools in many of the most prominent image-processing packages. we have been able to identify two software packages that have this functionality. the first is research software called impact, developed by milton smith, steve cothern, and bill gustafson at the department of geology’s remote-sensing laboratory at the university of washington. the second is envi®, a commercial remote-sensing software package with linear unmixing capability. with image processing and sma now outlined, let us provide an example of how sma can be applied to the study of human management of a public forest in south-central indiana. social science computer review spectral mixture analysis in a public forest of south-central indiana as an example of the use of sma for understanding the human dimensions of environ- mental change, we will focus the rest of our discussion on landscape change in one portion of the hoosier national forest, the pleasant run unit, which is located in south-central indiana. the brief analysis that follows seeks to show how sma can help identify differences in two forested landscapes subject to different management histories. for a more complete analysis of this area and other public forests in indiana, see schweik ( ). understanding the role humans play regarding landscape change in a particular region requires an investigation of what actions humans have taken over time and how these actions relate to the physical properties of land-cover reflectance. this involves examining the rela- tionship between light reflectance and the physical land cover and how human activities alter that land cover. however, merely linking human actions to physical outcomes is not enough. if we are to truly understand human-environment relationships then we must understand how the incentive structures humans face in their day-to-day activities alter their decision- making processes and the actions they ultimately take (ostrom, , in press). if social sci- entists wish to link human decision making to information captured by satellites, we must relate incentives to human actions, actions to outcomes, and outcomes to surface reflectance (schweik, ). to undertake such a change analysis using satellite images and sma requires a series of steps. first, we must carefully select the images we use in order to minimize weather and sea- sonal effects. second, it is important to undertake a theoretical consideration of how light reflectance is related to the land cover we wish to study—in this case, a forested landscape. this is useful toward the development of appropriate endmembers for fraction image genera- tion. third, theoretical considerations must also be given to the incentive structures humans face that relate to the choice of actions they undertake in the region. this allows us to develop expectations on how humans have altered the physical landscape. fourth, image processing is undertaken to convert the raw image dns to surface reflectance values. fifth, we must identify appropriate endmember locations and develop sma fraction images. finally, sixth, we can analyze sma results and create other output products such as fraction-based classifi- cations to assist in change analysis. we will briefly describe these steps below. image selection a crucial decision the analyst must make when undertaking change analysis using multi- spectral satellite images is image selection. over the -year landsat life span, there are a large number of images that might be selected for analysis for any particular location in the united states or elsewhere. careful image sampling is particularly important for it allows the analyst to minimize unwanted sources of variability. we followed several steps in choosing the images to analyze. first, we considered the temporal frequency of the human activities of interest. for this study, we are interested in public forest management. forest-cutting, development, and pres- ervation activities are most likely to be identified spectrally. development and preservation activities will begin at a particular time point and then will most likely remain throughout the landsat time series. forest cutting in this region, however, will result in an initial pattern of disturbance followed by regrowth and possible cutting again after or more years. given that the landsat system has been operational for more than years, maximizing the tempo- ral extent of the image time series can more appropriately sample the forest-cutting activity. schweik, green / spectral mixture analysis to capture the phenomenon of forest cutting throughout time, landsat images are selected early in the archives duration ( ) and late ( ) with at least one image in between. second, we considered the ramifications of seasonal effects on spectral response. for the study of forest resources, a late summer scene when tree canopies are full is appropriate. gaps where cutting exposes litter and soil are more apparent when a tree canopy is in full leaf cover than in fall or winter scenes. there is also a contrast between herbaceous growth and forests in late summer when dry conditions promote herbaceous senescence. for these rea- sons, we selected images acquired in september. the reader may also raise the concern of diurnal variability in satellite images. fortunately, landsat’s sun synchronous orbit ensures that all images are taken at nearly the same time each day. third, we considered the year-to-year climatic variability in the images. in the study done by schweik ( ), it became apparent that our selected image was dramatically differ- ent from two other time points. one hypothesis is that the difference is attributed to a higher- than-normal rainfall during that year. it is important, then, to select images that have had similar rainfall and temperature patterns during the months or even the year prior to image acquisition. finally, cost drives image selection. for this study, we used a georegistered time series of landsat mss images made available at no cost through the north american landscape characterization (nalc) program: / / , / / , and / / (lunetta, lyon, guindon, & elvidge, ; nalc, ). fortunately, the images made available through this program satisfied the first two selection criteria, but as noted above, the image exhibited signifi- cant spectral differences that are attributed to the higher rainfall that year. theoretical considerations related to the spectral properties of the landscape the analyst must next consider the spectrally important characteristics of the landscape under investigation. the human actions we wished to study were silviculture and develop- ment activities in a public forest. this meant considering the spectral characteristics of this broadleaf forest region. the major factors that affect reflectance from a midwestern oak-hickory forest canopy are leaf area and leaf optical properties that are a product of leaf morphology (green, ). it is well understood that leaf chlorophyll absorbs sunlight predominantly at red (. µm) and blue (. µm) wavelengths for use in photosynthesis. less green light (. µm) is absorbed by photosynthetic plants, thereby making them look green to the human eye. however, the reflectance spectra of vegetation spectra also reveals a significant peak at near-infrared wavelengths (. µm through . µm), an effect that the human eye cannot distinguish. this increase in reflectance is a response not to absorption by chlorophyll molecules but rather by scattering of light as it is reflected within the deeper internal structure of a leaf’s spongy mesophyll (campbell, ; knipling, ). in studies of similar oak-hickory forests in missouri (green, , ; green & arvid- son, ; green, arvidson, sultan, & guinness, ), distinctly high, near-infrared leaf- reflectance values were found to be associated with vegetation from xeric environments: areas with soils that exhibit low water-retention capacities. alternatively, these studies report forest vegetation in wetter, mesic environments to exhibit lower leaf-reflectance values in near-infrared. green ( ) attributes this phenomenon to different morphological qualities of the leaf internal structure as plants adapt to various environmental conditions: xeromor- phic leafs differ from mesomorphic ones by exhibiting a thicker outer cuticle layer (kramer social science computer review & kozolowski, ; spurr & barnes, ), a higher ratio of internal to external surface (kramer & kozolowski, ), and a thicker leaf with more palisade layers (green, ). it is thought that these differences cause additional scattering of infrared light in xeromorphic leaves. how do these phenomena relate to human actions on deciduous forests? xeromorphic leaf structures are often produced as a result of topographic conditions and water availability factors. they also appear to have a relationship with age of forest vegetation (schweik, ). water is gathered by the root system of trees (spurr & barnes, ), and younger trees will exhibit less extensive root networks, which result in less access to soil moisture. moreover, the canopy of younger trees, in an open stand that has recently been cut, will be exposed to more direct sunlight, thus promoting the development of more xeric leaf struc- tures, than will similar species growing in more shaded conditions (spurr & barnes, ). schweik ( ) analyzed well-documented cutting areas in a state forest in indiana and found that in general, environmental conditions and age-class appear to have a relationship with the amount of reflected near-infrared light. xeric topographic environments and young vegetated stands reflect higher amounts of near-infrared light than do older forest vegetation or vegetation residing in more mesic environments. the choice of mss imagery to maximize this study’s temporal duration limits us to the selection of three endmembers (four bands minus one). for this study, we chose endmembers as follows: one endmember represents xeric (brighter near-infrared) vegetation, and one represents nonliving matter (exposed soil and developed regions). because of the three-endmember limitation when using mss images, the third endmember represents darker near-infrared vegetation and also shadowed vegetation (shade) (adams et al., ). this endmember may also pick up pine stands as well because pine tends also to exhibit dark vegetation spectra. if we were using solely land- sat tm data, we would be able to employ five endmembers: xeric vegetation, mesic vegeta- tion, soil, shade, and pine. see adams et al. ( ) for one sma example using tm imagery. theoretical considerations about human behavior in the region similar theoretical considerations must be addressed concerning human behavior and action in the region. given that the forests studied are under public forest management, we focused our investigation on the national forest property manager. taking a “new- institutionalist” approach to the study of these managers (e.g., see eggertsson, ) and fol- lowing ostrom’s ( , in press) work, we assume that property managers face a suite of alternative actions that they and their employees might consider undertaking. they choose between alternative activities based on the incentive structures—the costs and benefits—that are established to encourage or discourage different types of activities. broadly speaking, the suite of activity choices in indiana public forests can be categorized as three types: commod- ity- (e.g., silviculture practices), recreation- (e.g., trail management), and preservation- related activities (koontz, ). the costs and benefits associated with any particular activity are influenced by attributes of the physical world, attributes of the human communities interested in the forest and the institutional landscape governing the area. the physical attributes of the region sometimes make the costs of undertaking an action so high that certain activities will not be considered, such as building a road on an extremely steep slope. other activities may have extremely high costs because of the reaction or anticipated reaction of users of the public forest. anticipated schweik, green / spectral mixture analysis protests from environmental groups and landowners adjacent to the property may also increase the expected costs of particular activities such as clear-cutting of forest stands. we define the institutional landscape as the set of rules that are established to determine the spatial distribution of where activities can and cannot be conducted along with the moni- toring and sanctioning mechanisms used to enforce these rules (schweik, ). a property manager’s selection of certain activities to undertake is also encouraged or discouraged by the configuration of this institutional landscape. in the case of the hoosier national forest, legislation, revenue rules and allocations, standard operating procedures, and planning documents place significant constraints on where particular activities can be undertaken (koontz, ; schweik, ), and many of these restrictions involve a geographic component. a full institutional history of the hoosier pleasant run unit is beyond the scope of this article (see schweik, ). but as an example, we will focus on two different regions of pleasant run that have had different institutional histories over roughly the past years. these two units have different types of activities that are permitted, required, or prohibited within their boundaries as specified in the hoosier forest management plans. the two areas of interest for this article are area . , designated wilderness in , and area . , desig- nated an intensive recreation area; their boundaries are identified in figures and . from knowledge of the institutional histories and assumptions that the behavior of the property managers will largely comply with the incentive structures created by these institu- tional configurations, one can develop a set of expected and testable hypotheses related to forest land-cover change. hypothesis : the area designated as wilderness (area . ) will exhibit little evidence of human disturbance and signs of natural vegetation regrowth. as the deam wilderness area ages, image pixels are expected to exhibit declining percentages of the xeric endmember and increasing per- centages of the mesic endmember over time. hypothesis : the area designated as intensive recreation (area . ) will exhibit higher levels of development activities and permanent soil exposure than will the wilderness area. consistently higher percentages of the soil endmember will be found in this management area over time. image processing image-processing steps involved a series of procedures to minimize nonsources of radio- metric variability, such as that produced by the satellite and atmosphere, and to ensure geo- metric comparability. we first verified that the images had little or no variability produced by differences in sensor sensitivity (striping) and sensor power surges (speckle). georeferenc- ing has been performed on all nalc program scenes. this is a crucial step in change analy- sis and ensures comparability across the time series of images. the danger is that misregis- tration between images may be mistaken for land-cover change and could cause significant errors in change analysis results. the nalc program data sets have specifications of geo- referencing error below one pixel (nalc, ). we performed additional georeferencing verification by overlaying a very accurate road network gis coverage over each image to visually confirm the georeferencing of each image in the area of study. we then converted the image dns to at-satellite reflectance and minimized atmospheric effects using the spreadsheet-based procedure that we developed (green et al., ). we then normalized the / / and the / / image to the base / / image using the radiometric rectification procedure developed by hall et al. ( ). this processing converted the dns of each nalc image to surface reflectance. these postprocessed images can now be readily compared. social science computer review endmember identification and sma fraction image development the next step is to identify candidate xeric vegetation, mesic vegetation, and soil end- member pixels within the image region that exhibit extreme spectral qualities for their land- cover type (recall figure b). the / / mss image was searched for locations that appear to be pure representations of these land-cover types. xeric forests were identified in mss images by their high near-infrared reflectances. in indiana, they tend to be found on hill sum- mit areas and on southwest-facing slopes. these areas have higher near-infrared reflectances and appear brighter in image composites. mesic forest vegetation regions, on the other hand, tend to reside in valley bottoms (where water availability is high) and exhibit relatively darker near-infrared reflectance. searching for extreme pure endmember locations is an iterative process. first, we removed water areas from consideration by identifying those pixels that exhibited low values across all mss bands. these locations were removed from consideration so that sma did not have to model them. next, we used three-dimensional scatter plots of band data to iden- tify extremely bright soil and xeric vegetation locations. three extremely bright soil loca- tions, four extremely bright xeric vegetation locations, and three dark mesic vegetation loca- tions were identified using this technique. using differential global positioning systems technology with about five meters’ accuracy, we verified these locations in the field to make sure they were indeed good examples of these endmember types. the spectra for all other pixels within the hoosier forest region, theoretically, should be explained by linear combina- tions of these three endmember spectra. several sma models were run combining various soil, xeric, and mesic/shade candi- dates. sma output for each mss time point is four new images: three representing the frac- tional percentage of each endmember for each pixel and the fourth representing the rms error or goodness of fit of each model. areas not modeled well are identified by high values in the rms error image. the final model has to fit two selection criteria: (a) the spectral vari- ability present is largely accounted for, and (b) the model produces endmember percentages that fall between % and % (envi, ; ustin et al., , p. ). time-series analysis requires a third criteria: the models must satisfy the first two criteria for all image dates. this modeling process is complicated by limitations in mss degrees of freedom and because the mesic/shade endmember does not model both categories particularly well. the model that satisfied these conditions the best used endmembers of a cleared agricultural field, a xeric vegetation location residing on a summit and partially on a southwest-facing slope, and a mesic vegetation location that resides in a valley and is somewhat shaded. figure displays the stick spectra graphs for the final soil, xeric vegetation, and mesic vegetation endmembers chosen. the resultant fraction and rms images produced from this sma model for one region of the / / time point are shown in figure . similar products were produced for the other dates. the landscape in figure highlights an example that will help the reader interpret sma- derived images. circled in the image is a known area where a forest property manager office exists. note that this area is very bright in the soil fraction image, meaning that these pixels have a high percentage of exposed soil. this area is also dark in the xeric endmember fraction image and, therefore, exhibits a low percentage of this endmember. finally, it is moderately dark in the mesic/shade endmember, so some mesic vegetation and shadow also exist within this area. this provides an example of how sma provides the analyst with the capability to conduct subpixel analysis—a capability not possible using traditional image classification schweik, green / spectral mixture analysis analysis techniques. by employing sma, we have effectively moved our analysis from one depicting changes in reflectance to one depicting changes in surface composition—a mea- surement system much more accessible to the social science community. we now also have a mechanism to evaluate how well this system is defined for each pixel (the rms image). change map creation and analysis with the sma fraction images developed, several types of analysis techniques can be applied. schweik ( ), for example, analyzed specific locations where known forest man- agement activities have occurred and found that these sma fraction images are sensitive to various types of cutting and development activities. however, the hypotheses stated above require an analysis at a broader spatial scale: the landscape. hypothesis predicts that the deam wilderness will exhibit, overall, a general decline in the percentage of the xeric endmember with time. because the institutional landscape for this area prohibits timber-cutting activities and the incentive structure strongly encourages prop- erty managers to adhere to this mandate, it is expected that vegetation in this area designated as wilderness will exhibit signs of aging across the time series. as trees get older, it is expected that the cell structure of their leaves will become less xeromorphic as their root sys- tems become more extensive, accessing more moisture, and as a greater proportion of their leaves become protected from direct exposure to sunlight. social science computer review figure : stick spectra of chosen endmembers: soil, bright (xeric) vegetation, and dark (mesic) vegetation and shade (landsat multispectral scanner path , row , / / ) a color composite image can be used qualitatively to test this hypothesis by assigning each of the three xeric fraction images of each time-point image to blue, green, and red. in the resulting image, different colors are associated with different temporal trajectories. the dif- ficulty with this method is that the continuous nature of the fraction image data (and satellite image data in general) makes interpretation of the subtle color changes difficult. this is exactly why humans find classification maps so useful: they aggregate continuous data into a few discrete categories, greatly simplifying analysis. recall the earlier argument about traditional classification approaches being image dependent and not readily transferable to analyses of other images. using sma fraction images, the analyst can now develop classifications that no longer are based on statistical measures of a single image but are instead developed from endmember fractions, grounded in physical measures of reflectance, and acquired at specific locations on the earth’s surface. for example, the analyst can now define what he or she considers a xeric forest based on these percentages of certain materials. in our study, we define a xeric forest as one possessing % or higher of the xeric endmember and less than % of either the soil or mesic/shade endmembers. with these definitions established, the analyst can easily convert the continuous data of the three endmember fractions for any image into a classification map based on endmember values using software modeling functionality. moreover, the same classification definitions can be used for each image time point, something that cannot easily be accomplished using traditional image classification procedures. side-by-side image classification maps based on sma fractions for the wilderness area are shown in figure , and figure presents similar maps for the intensive recreation area. to demonstrate the utility of these products, let us now briefly address the hypotheses stated earlier and interpret these classification maps. hypothesis predicted that given that area . is treated as wilderness, the vegetation here will exhibit signs of continued forest aging. in terms of sma fraction images, it is expected that the vegetation in the wilderness area will, in general, be more xeric in and schweik, green / spectral mixture analysis figure : spectral mixture analysis for a portion of the landsat multispectral scanner / / image note: bright white = high percentage; dark = low percentage; rms = root mean squared. move more toward mesic conditions in after years of root growth and increasing moisture availability. reviewing the side-by-side classification images in figure , several pieces of evidence tend to confirm this hypothesis. first, the northern peninsula of the deam wilderness, circled and labeled a in both maps of figure , reveals a significant amount of exposed soil (light areas or gold in color) in the map, whereas in the map, it is replaced by a combination of the mesic vegeta- tion/pine/shade class. a review of a september tm image reveals a significant pine stand on this peninsula. because pine is not indigenous to this area, this suggests that a major planting initiative occurred in this area between and , probably to protect soil ero- sion into the adjacent lake monroe. second, the conversion of soil to vegetation is apparent in the area identified by b in fig- ure , which exhibits a pattern we would expect in an area left alone to regenerate. scattered soil-dominated pixels (gold) in have been replaced by more prevalent xeric spectra (red), mesic/pine/shade spectra (dark green), or a combination of xeric/mesic vegetation (bright green). thus, as the deam ages after its designation in , forest vegetation appears to be returning. third, the large circles labeled c on the figure maps reveal a significant area that exhib- its what appears to be following patterns of natural regrowth. many of the stands that exhib- ited xeric (red) characteristics in remain xeric in . we used a digital elevation model for the region to identify locations that are naturally more xeric in nature (e.g., sum- mits and southwest-facing slopes). these areas would be expected to remain xeric over time because of the natural effects of their physical environment. but many other xeric-dominated social science computer review figure : change in the hoosier national forest pleasant run unit area . (deam wilderness) note: for a color version, visit the web site at http://www.indiana.edu/~cipec/publications/ publications.html. mss = multispectral scanner; sma = spectral mixture analysis. areas in have been replaced in with higher percentages of mesic vegetation (bright green or dark green areas in the color versions of the maps). these areas take on the expected high-to-low xeric to lower xeric change trajectory that schweik ( ) identifies as a spec- tral pattern of natural regrowth in these forests. this evidence also strongly supports hypothesis . hypothesis predicts that the high recreation area will exhibit patterns of development (permanent soil exposure and nonvegetative land cover) in areas where the property manag- ers build or maintain recreation infrastructures. reviewing areas identified by a in figure reveals a higher level of soil exposure in this area in than that found in the deam wilder- ness area. note, however, that in much of this soil has been replaced with mesic or mesic/xeric land cover. the history of the area helps to explain this trajectory. the lake adja- cent to this area was created in the mid- s. the soil exposure in is a result of signifi- cant recreation infrastructure development that occurred shortly thereafter. however, as time has passed, property managers have worked to allow the remaining forest surrounding these recreation areas to continue to age because larger trees are more visually appealing to visi- tors. therefore, the change trajectory of the area identified by a depicts a management strat- egy of development coupled with visual enhancement (natural regeneration) efforts. as trees have aged, their canopies have grown and covered many of the developed areas (e.g., roads) that were more visible in the early development in . areas identified in figure by b in these maps exhibit a similar natural regeneration spectral trajectory as managers work to keep visual enhancement areas pleasing to visitors. schweik, green / spectral mixture analysis figure : change in the hoosier national forest pleasant run unit area designated . (inten- sive recreation area) note: for a color version, visit the web site at http://www.indiana.edu/~cipec/publications/ publications.html. mss = multispectral scanner; sma = spectral mixture analysis. discussion and conclusion: the promise of sma for studying the human dimensions of environmental change the tremendous advances in desktop computing, even over the past years, has produced great interest within the social science community on the use of multitemporal, multispectral satellite images for the study of human-environment relationships. the technology and know-how that used to only reside in the hands and minds of scholars with graduate degrees in remote sensing can now be more commonly applied by scholars in various disciplines such as geography, anthropology, public administration, and political science: all fields with an interest in understanding the human role in global change. let us summarize our main thesis. multitemporal, multispectral satellite images, such as landsat, are very useful in the study of the human dimensions of environmental change stud- ies because the archive covers a tremendous spatial extent of the earth’s surface and more than a quarter century in duration with very frequent temporal sampling. their radiometric and spatial resolution continues to improve as new satellites are launched. however, to link the information provided in these images to human incentives, actions, and outcomes such that comparisons can be made between studies, we must carefully process the images to remove nonsurface sources of variability that cannot be attributed to variations in land cover and land-cover change. to remove this variability, images should be processed from dns to physical units: surface reflectance values, through radiometric calibration and atmospheric correction procedures. in doing so, analyses can be duplicated and replicated by researchers studying other locations, and accurate comparisons can be made across space and time. fur- thermore, by reporting endmember characteristics and locations and applying sma to gen- erate endmember fraction images, the researcher can define classifications grounded on a physical measure of the earth’s surface: reflectance and based on how much of what material is present. by using the xeric/mesic distinction, we have classified our forests based on quan- titative, ecologically significant properties tied to measurable forest properties such as leaf morphology. unlike traditional classification, these categories can be readily applied and replicated by others working in distant but comparable environmental regions. to understand the role humans have played in changing land cover, one must give careful attention to both natural as well as human parameters. the analyst must use effective sam- pling strategies and gis products (e.g., digital elevation models or other topographic prod- ucts) to account for natural influences such as rainfall, elevation, and slope aspect. to under- stand the human dimensions of change and the motivation behind them, the analyst should identify (a) the suite of alternative actions undertaken by humans in a particular con- text, (b) how these actions physically disturb the landscape and affect light reflectance, and importantly, (c) what incentive structures (e.g., rules, social norms, and legislation) have driven human decision-making and activity choice (schweik, ). by linking incentives to actions and actions to outcomes measured with physically based yardsticks such as sur- face reflectance values, more straightforward human-environment relationships can be understood. image processing and sma have the potential to produce other benefits as well. if physi- cal relationships can be identified between vegetation reflectance and sma percentages with other important global change parameters, such as those associated with forest health under climate change, researchers may be able to use spectral measurements as proxies for these vitally important but costly-to-measure parameters and develop image-based inven- tory estimates appropriate for particular forested landscapes. we conclude with guidelines on how to link the study of human incentives and actions with sma-based landscape measurements. the researcher/analyst should social science computer review . investigate and document the incentive structures that drive human actions in the context of the study location; . inventory the suite of potential activity choices that humans in this area undertake; . consider the temporal frequency of land-cover disturbances that these activities will produce and select sufficient images to critically sample the phenomenon of interest; . control for season and weather effects through careful selection of satellite images; . identify a subset of locations where these actions have occurred and investigate the physical response produced in the image time series; . control for topographic effects through the use of gis products such as a digital elevation mod- els; and . convert the multispectral images from dns to surface reflectance values such that nonsurface sources of variation can be removed and not attributed to human actions. in doing so, findings can be compared with other studies of similar forest types. a much more detailed description and example of this process is shown in schweik ( ). for more information, contact the authors or visit our world wide web site at http://www.indiana.edu/~cipec. notes . image preprocessing involves the removal of pixel-to-pixel variability not associated with atmospheric or sur- face phenomena. it involves the conversion of digital images from cd or tape to the appropriate image-processing software format, removing any artifactual noise from the image, such as line dropouts and line striping, and image georeferencing. . at-satellite reflectance describes the ratio of light received by the sensor from the entire earth-atmosphere system over the solar irradiance at earth-sun distances. these values still contain the effects of the atmosphere, such as scattering and absorption. . in our case, we use model maker, which is part of imagine® software, to produce this classification. . these classifications cannot easily be interpreted in black and white. therefore, the color classifications for this article are available at http://www.indiana.edu/~cipec/publications/publications.html. references adams, j. b., & adams, j. d. ( ). geologic mapping using landsat mss and tm images: removing vegetation by modeling spectral mixtures (third thematic conference, remote sensing for exploration geology). erim, , - . adams, j. b., sabol, d., kapos, v., filho, r. a., roberts, d. a., smith, m. o., & gillespie, a. r. ( ). classifica- tion of multispectral images based on fractions of endmembers: application to land-cover change in the brazil- ian amazon. remote sensing of environment, , - . adams, j. b., smith, m. o., & gillespie, a. r. ( ). imaging spectroscopy: interpretation based on spectral mix- ture analysis. in c. m. pieters & p. englert (eds.), remote geochemical analysis: elemental and mineralogical composition (pp. - ). new york: cambridge university press. adams, j. b., smith, m. o., & johnson, p. e. ( ). spectral mixture modeling: a new analysis of rock and soil types at the viking lander site. journal of geophysical research, , b , - . ballew, g. i. ( ). a method for converting landsat mss data to reflectance by means of ground calibration sites (stanford remote sensing lab technical report, srsl - ). stanford, ca: stanford university. bateson, a., & curtiss, b. ( ). a method for manual endmember selection and spectral unmixing. remote sens- ing and environment, , - . brondizio, e. s., moran, e. f., mausel, p., & wu, y., ( ). land use change in the amazon estuary: patterns of caboclo settlement and landscape management. human ecology, ( ), - . schweik, green / spectral mixture analysis bryan, w. b., finger, l. w., & chayes, f. ( ). estimating proportions in petrographic mixture equations by least-squares approximation. science, , - . campbell, j. b. ( ). introduction to remote sensing ( nd ed.). new york: guilford. chavez, p. s. ( ). radiometric calibration of landsat thematic mapper multispectral images. photogrammetric engineering and remote sensing, , - . cohen, w. b. ( ). response of vegetation indices to changes in three measures of leaf water stress. photogram- metric engineering and remote sensing, , - . drury, s. a. ( ). a guide to remote sensing: interpreting images of the earth. oxford, uk: oxford university press. eggertsson, t. ( ). economic behavior and institutions. cambridge, uk: cambridge university press. envi. ( ). envi tutorials, envi version . . boulder, co: better solutions consulting limited liability co. envi. ( , february ). envi tutorial: multispectral processing using envi’s hyperspectral tools. available: http://www.envi-sw.com/e_msihsi.htm. green, g. ( ). physical basis for remotely sensed spectral variation in a semi-arid shrubland and an oak- hickory forest: implications for mapping soil types in vegetated terrains. unpublished doctoral dissertation, washington university, department of earth and planetary sciences, st. louis, mo. green, g. ( ). forest structure and leaf spectral control on canopy reflectance: implications for geologic map- ping using remote sensing in forested terrains (working paper). bloomington: indiana university, anthropo- logical center for training and research on global environmental change. green, g., & arvidson, r. e. ( ). soil types and forest canopy structures in southern missouri: a first look with ais data. in g. vane & a. goetz (eds.), proceedings of the second airborne imaging spectrometer data analy- sis workshop (pp. - ). pasadena, ca: nasa jet propulsion laboratory. green, g., arvidson, r., sultan, m., & guinness, e. ( , april). geobotanical information contained in landsat thematic mapper images covering southern missouri. paper presented at the fourth thematic conference, remote sensing for exploratory geology, san francisco. green, g., schweik, c. m., & hanson, m. ( ). radiometric calibration of landsat multispectral scanner and thematic mapper images: guidelines for the global change community (working paper). bloomington, in: center for the study of institutions, population, and environmental change. hall, f. g. ( ). adaptation of nasa remote sensing technology for regional-level analysis of forest ecosystems. in v. a. sample (ed.), remote sensing and gis in ecosystem management. washington, dc: island press. hall, f. g., strebel, d. e., nickeson, j. e., & goetz, s. j. ( ). radiometric rectification: toward a common radio- metric response among multidate, multisensor images. remote sensing of environment, , - . hill, j. ( ). a quantitative approach to remote sensing: sensor calibration and comparison. in a. s. belward & c. r. valenzuela (eds.), remote sensing and geographical information systems for resource management in developing countries (pp. - ). brussels, belgium: ecsc, eec, eaec. idrissi. ( ). idrissi users guide version . . worcester, ma: clark university. igap. ( , march ). [indiana gap homepage]. available: http:// . . . /h /bertha/gap. jackson, r. d. ( ). spectral indices in n-space. remote sensing of the environment, , - . jensen, j. r. ( ). introductory image processing: a remote sensing perspective. englewood cliffs, nj: prentice hall. knipling, e. ( ). physical and physiological basis for the reflectance of visible and near-infrared radiation from vegetation. remote sensing of the environment, , - . koontz, t. ( ). federalism and national resource policy: comparing state and national management of public forests. unpublished doctoral dissertation, indiana university, bloomington. kornder, s. c., & carpenter, j. r. ( ). application of a linear mixing algorithm to the normal alkane patterns from recent salt marsh sediments. organic geochemistry, ( ), - . kramer, p., & kozolowski, t. ( ). physiology of trees. new york: mcgraw-hill. lee, c. t., & marsh, s. e. ( ). the use of archival landsat mss and ancillary data in a gis environment to map historical change in an urban riparian habitat. photogrammetric engineering and remote sensing, ( ), - . lillesand, t. m., & kiefer, r. w. ( ). remote sensing and image interpretation. new york: john wiley. lunetta, r. s., lyon, j. g., guindon, b., & elvidge, c. d. ( ). north american landscape characterization data- set development and data fusion issues. photogrammetric engineering & remote sensing, ( ), - . markham, b. l., & barker, j. l. ( , august). landsat mss and tm post-calibration dynamic ranges, exoatmos- pheric reflectances and at-satellite temperatures. eosat landsat technical notes, , - . moran, e., brondizio, e., & mausel, p. ( ). secondary succession. national geographic research and explora- tion, ( ), - . social science computer review north american landscape characterization (nalc). ( , april ). north american landscape characteriza- tion. available: http://edcwww.cr.usgs.gov/glis/hyper/guide/nalc. ostrom, e. ( ). governing the commons: the evolution of institutions for collective action. cambridge, uk: cambridge university press. ostrom, e. (in press). self-governance of common-pool resources. in p. newman (ed.), the new palgrave diction- ary of economics and the law. london: macmillan. price, j. c. ( ). calibration of satellite radiometers and comparison of vegetation indices. remote sensing of environment, , - . richards, j. a. ( ). remote sensing digital image analysis: an introduction. new york: springer-verlag. robinove, c. j. ( ). computation with physical values from landsat digital data. photogrammetric engineering and remote sensing, ( ), - . sader, s. a. ( ). spatial characteristics of forest clearing and vegetation regrowth as detected by landsat the- matic mapper imagery. photogrammetric engineering and remote sensing, ( ), - . sader, s. a., & wynne, j. c. ( ). rgb-ndvi color composites for visualizing forest change dynamics. interna- tional journal of remote sensing, , - . schweik, c. m. ( ). spatial analysis of natural resources in east chitwan, nepal: conceptual issues and a multiscale research program. in g. shivakoti, g. varughese, e. ostrom, a. shukla, & g. thapa (eds.), people and participation in sustainable development: understanding the dynamics of natural resource systems (pro- ceedings of an international conference, - march , institute of agriculture and animal science, tribuvan university, rampur, chitwan, nepal). bloomington: indiana university, workshop in political the- ory and policy analysis. schweik, c. m. ( ). the spatial and temporal analysis of forest resources and institutions. unpublished doc- toral dissertation, indiana university, bloomington. scott, j. m., davis, f., csuti, b., noss, r., butterfield, b., groves, c., anderson, h., caicco, s., d’erchia, f., edwards, t. c., ulliman, j., & wright, r. g. ( ). gap analysis: a geographic approach to the protection of biological diversity. wildlife monographs. available: http://www.gap.uidaho.edu/gap/aboutgap/ wildlifemonographs/c.htm. smith, m. o., ustin, s. l., adams, j. b., & gillespie, a. r. ( a). vegetation in deserts: i. a regional measure of abundance from multispectral images. remote sensing of environment, , - . smith, m. o., ustin, s. l., adams, j. b., & gillespie, a. r. ( b). vegetation in deserts: ii. environmental influ- ences on regional abundance. remote sensing of environment, , - . spurr, s. h., & barnes, b. v. ( ). forest ecology. malabar, fl: krieger. teillet, p. m., & fedosejevs, g. ( ). on the dark target approach to atmospheric correction of remotely sensed data. canadian journal of remote sensing, ( ), - . tucker, c. j., vanpraet, c. l., sharman, m. j., & van ittersum, g. ( ). satellite remote sensing of total herba- ceous biomass production in the senegalese sahel, - . remote sensing of the environment, , - . ustin, s. l., smith, m. o., & adams, j. b. ( ). remote sensing of ecological processes: a strategy for develop- ing and testing ecological models using spectral mixture analysis. in j. ehleringer & c. field (eds.), scaling physiological processes: leaf to globe (pp. - ). san diego, ca: academic press. verbyla, d. l. ( ). satellite remote sensing of natural resources. boca raton, fl: lewis. westman, w. e., strong, l. l., & wilcox, b. a. ( ). tropical deforestation and species endangerment: the role of remote sensing. landscape ecology, ( ), - . wilkie, d. s., & finn, j. t. ( ). remote sensing imagery for natural resources monitoring: a guide for first-time users. new york: columbia university press. wolter, p. t., mladenoff, d. j., host, g. e., & crow, t. r. ( ). improved forest classification in the northern lake states using multi-temporal landsat imagery. photogrammetric engineering and remote sensing, ( ), - . charles m. schweik has a ph.d. in public policy from indiana university and is a research associate with indiana university’s center for the study of institutions, population, and environmental change (cipec). he currently is a visiting scholar at the center for public policy and administration, university of massa- chusetts, amherst. his e-mail address is cschweik@pubpol.umass.edu. glen m. green is a remote-sensing postdoctoral scholar at cipec, indiana university. he received his ph.d. from washington university (department of earth and planetary sciences) in st. louis, missouri. his e-mail address is glgreen@indiana.edu. schweik, green / spectral mixture analysis asept/oct —vol. , no. journal of soil and water conservation jim gulliford reflections on the swcs annual international conference doi: . /jswc. . . a a nd they gathered in reno. the th annual international conference of the soil and water conserva- tion society met in the high desert valley truckee meadows at the foot of the sierra nevada mountains. this fertile valley feels the pressure of urbanization and competi- tion for limited water resources. it reminds us of one of many challenges we face: competition for the soils and water that will produce the food necessary to feed our future generations. resilient landscapes, our conference theme, reflects a second compelling chal- lenge for the future. throughout the world, the agricultural landscape shows the scars of the tremendous pressure to produce in today’s climate and economic conditions. flood, fire, drought, and storm have shown us why conservation management deci- sions are as important to the farmer and rancher as seed, fertility, and pest manage- ment. we know that there is resilience in our soils, but we also know that we are pressing the limit of that resilience in far too many cases. dr. gene kelly set the stage for the conference in the pritchard lecture. he examined the vulnerability of soils to changing climate and land use condi- tions from across the great plains to the south african savannah. if you missed the conference, visit the swcs web site to hear his challenging and entertaining thoughts on soil formation, utilization, and resilience. it is an essential message for our members. our day two plenary examined the sci- ence and practice of soil health. opening with the lyrical “what’s soil health got to do with it?” wayne honeycutt described the usda natural resources conservation service soil health initia- tive and moderated a usda agricultural research service science panel of jorge delgado, jerry hatfield, and jean steiner. the panel reflected on the importance of soil structure, chemistry, and biology to soil health. they described conservation management and practices as tools to both protect soils and restore them. while there are limits to soil resilience, there are con- servation practices that can help manage and improve soil characteristics. the swcs house of delegates meet- ing was an important organizational activity at this year’s conference. the swcs board of directors president dan towery and the executive director jim gulliford presented a proposal the board of directors has developed to restructure the swcs board to position the society to better serve its members and chapters and to make stronger contributions to the science, practice, and policy of resource conservation internationally. the response and comments of the delegates were help- ful to the board to both understand and address concerns of the membership. visit the members only page of the swcs web site for a complete explanation of the reorganization proposal. i can never say enough about the depth and quality of the workshops, symposia, general sessions, and posters that are the foundation for learning and networking each year. the berg and society fellows forum provided a balanced look at the science and environmental consequences of hydraulic fracturing. conference par- ticipants had an opportunity for in-depth study of nitrogen management tools, leadership practices, and the use of social indicators in watershed management proj- ects. the usda conservation effects assessment project rangeland assessment and national institute of food and agriculture water resources research, education, and outreach were featured symposia sessions. each year, we celebrate the accom- plishments of our members and partners. jean steiner and robert bird were named fellows of the society for their career contributions to the field of soil and water conservation, and the hoosier chapter received this year’s outstanding chapter award. we celebrated the th anniversary of the certified professional in erosion and sediment control program with con- ference sponsor envirocert international inc. and saw an example of how that pro- gram is growing internationally. three outstanding tours were offered and filled this year as attendees took advantage of the opportunity to learn more about conservation programs unique to the reno area. as i watched the busses load wednesday morning, i heard an indi- vidual relate that for years she had heard about the importance of the snow survey program and how pleased she was to be able to visit a monitoring station and learn more about this water management tool. you can’t get that experience sitting at your desk. while each annual conference is unique, one common theme each year is the essential contribution that our plan- ning committee, host chapter, and sponsors make to the operational success of the conference. tommy bass, the program committee chair, and his committee did a great job preparing the program. the california-nevada chapter was a great host. several members of this chapter— tibor horvath, wendy rash, rob roy, tom esgate—and all their volunteers pro- vided the personal touch that makes each conference unique to the host location. dupont pioneer, envirocert international jim gulliford is executive director of the soil and water conservation society, ankeny, iowa. jim gulliford swcs conference echoes c o p yrig h t © s o il a n d w a ter c o n serva tio n s o ciety. a ll rig h ts reserved . w w w .sw cs.org ( ): a - a jo u rn a l o f s o il a n d w a ter c o n serva tio n http://www.swcs.org a journal of soil and water conservationsept/oct —vol. , no. inc., monsanto, and all of our sponsors and exhibitors provided financial support and gave our participants the opportunity to learn about their products and services that support conservation on agricultural lands. our staff and attendees can’t thank you enough for those essential contributions. hold the date! we’ll do it again next year in suburban chicago at the weston lombard on july – , . i prom- ise you an opportunity to learn from and interact with leading conservation researchers, practitioners, and policy mak- ers in our field. plan now to participate and present your work next year. c o p yrig h t © s o il a n d w a ter c o n serva tio n s o ciety. a ll rig h ts reserved . w w w .sw cs.org ( ): a - a jo u rn a l o f s o il a n d w a ter c o n serva tio n http://www.swcs.org impact of genomic changes on the treatment of lung cancer: mtp - journal of thoracic oncology • volume , number , supplement , august th world conference on lung cancer copyright © by the international association for the study of lung cancers mtp - egfr-targeted therapy, tue, sept , : - : impact of genomic changes on the treatment of lung cancer johnson, bruce e. dana-farber cancer institute, brigham and women’s hospital, harvard medical school, boston, ma, usa three years have passed since mutations of the tyrosine kinase domain of the epidermal growth factor receptor (egfr) were discovered in patients with lung cancer who had dramatic clinical responses to treat- ment with gefitinib. additional laboratory and clinical studies have provided further insights into the biological impact of egfr mutations in cell culture experiments and in patients with lung cancer. in vitro characterizations of lung cancer cell lines and host cell lines transfected with these mutant and wild type egfrs show most cell lines with mu- tated egfrs are growth inhibited by - fold lower concentrations of gefitinib and erlotinib ( - nm) compared to wild type egfr (approximately - um). lung cancer cell lines with mutations of the egfr treated with concentrations of gefitinib and erlotinib that are achievable in the plasma undergo apoptosis rather than growth arrest. the other common genomic changes that arise in adenocarcinomas of the lung that have an impact on tyrosine kinase inhibitors and other targeted agents include kras, braf, and lkb . prospective studies of lung cancer patients with egfr mutations treated with gefitinib and erlotinib have reported a close association between egfr mutations, increased chance of clinical response and longer survival. patients with egfr mutations treated with gefitinib or erlotinib have a response rate of approximately %, a median time to progression in excess of approximately one year, and a median survival in excess of two years. this has led to the development of commercial tests to determine if the dna from tumors retrieved from patients with adenocarcinoma have a mutation of the egfr. the genomic change associated with resistance to treatment with gefinib and erlotinib is a dna mutation which changes the threonine to methionine at the th amino acid of egfr known as the (t m) mutation as well as amplification of the met oncogene. irreversible in- hibitors including hki- and pf- can cause growth inhibition in nsclc cell lines with both the resistance and sensitizing mutations, while gefitinib and erlotinib do not. hki- and pf- entered directed phase i and phase ii trials in patients previously treated with gefitinib and erlotinib and we await the results of the trials. mtp - pulmonary toxicity of therapy, tue, sept , : - : pulmonary toxicity of therapy jett, james r. mayo clinic, rochester, mn, usa the toxicities of dyspnea, hypoxia, pneumonitis, and/or fibrosis related to thoracic radiotherapy have been observed for many years. toxicities are generally graded by using the nci common toxicity criteria or the who toxicity criteria (http://www.fda.gov/cder/cancer/toxicityframe. htm). dose-volume histograms have been developed to evaluate the volume of lung receiving various doses of radiation. the v is the per- centage of lung volume that receives a radiation dose of gy or more. in general, the larger the v volume, the higher the risk of radiation pneumonitis, and v volume of % or more are associated with high risk of acute radiation pneumonitis and late scarring ( ). acute radia- tion pneumonitis occurs within - months of radiotherapy, and may develop in - % of patients. direct injury to endothelial and epithelial cells may result in radiographic findings of pulmonary infiltrates in the field of radiation, but also “out of field” responses have been described and biopsies demonstrate a bronchiolitis obliterans organizing pneumo- nitis (boop) like reaction (also called cryptogenic organizing pneumo- nia – cop). the rtog evaluated acute and late ( days from start of rt) pulmonary toxicities associated with completed trials of thoracic radiotherapy with sequential or concurrent chemotherapy ( ). acute pulmonary toxicity of grade or greater occurred in - % of partici- pants. late pulmonary toxicities occurred in - % of patients. severe late lung toxicity was greater in trials with concurrent chemotherapy and hyperfractionated rt, as well as induction chemotherapy followed by concurrent chemoradiotherapy (one daily). sequential chemotherapy and radiotherapy had lower rates of late pulmonary toxicity ( %). ste- reotactic body radiation therapy (sbrt) is being utilized with increas- ing frequency, especially for medically inoperable lung cancer patients. with this technology, large dose fractions are given to very localized fields. in a phase ii trial of sbrt, investigators administered - gy in fractions over - weeks ( ). grade - toxicity occurred in of patients enrolled. toxicity was greater with centrally located tumors as compared to the peripheral cancers. there were treatment related deaths. four of these deaths were associated with pneumonia. this increased toxicity with centrally located tumors is due to bronchial stenosis and scarring due to the high dose of radiation. gemcitabine gemcitabine pulmonary toxicity has been reported in - % of patients treated with this agent and may be fatal. the ct features include ground-glass opacities, thickened septal lines and reticular opaci- ties. the distribution is diffuse and bilateral. gemcitabine pulmonary toxicity may be difficult to differentiate from viral pneumonitis. it is important to discontinue treatment as soon as it is suspected because an additional treatment may prove to be fatal. in severe cases, corticoste- roids are usually given although their efficacy is uncertain. gemcitabine is a potent radiation sensitizer with recall phenomena. the maximum dose of gemcitabine given with concurrent thoracic radiotherapy is substantially attenuated, from the usual doses when given alone, because of the risk of pneumonitis. in a phase i trial of the calgb, the maximum tolerated dose of twice weekly gemcitabine was mg/m ( mg/m /week) given with concurrent thoracic radio- therapy ( ). in a phase ii trial with gemcitabine ( mg/m on d , ) and cisplatin given concurrent with thoracic radiotherapy, grade - dyspnea was observed in of ( %) of participants ( ). taxanes the taxanes, paclitaxel and docetaxel, are widely used in treatment of lung cancer. acute type i hypersensitivity reaction with paclitaxel are well known. pulmonary infiltrates following treatment have been reported in low frequency with paclitaxel and somewhat more cases following docetaxel treatment. docetaxel has also been associated with the development of pleural effusions. the combination of docetaxel and gemcitabine has been tested in a phase ii trial by japan lung cancer cooperative group. they observed an unexpectedly high rate of grade interstitial lung disease of % ( of patients) with this combina- tion ( ). others have also reported high pulmonary toxicity with this combination. sekine and associates observed pneumonitis in % ( of ) of japanese patients treated with consolidative docetaxel fol- lowing concurrent chemoradiotherapy with other agents ( ). similarly, the hoosier oncology group reported pulmonary toxicity in % of patients receiving consolidative docetaxel ( ). phase ii studies a phase ii study of combined vegf inhibitor (bevacizumab+sorafenib) in patients with metastatic breast cancer: hoosier oncology group study bre - lida a. mina & menggang yu & cynthia johnson & cyndi burkhardt & kathy d. miller & robin zon received: april /accepted: may /published online: june # the author(s) . this article is published with open access at springerlink.com summary purpose: angiogenesis plays an essential role in tumor development, invasion and metastasis. we evaluated the efficacy and safety of dual angiogenesis blockade with bevacizumab and sorafenib in patients with metastatic breast cancer. patients and methods: patients who had received no more than prior chemotherapy regimens in any setting were treated with sorafenib mg as a single oral dose daily plus bevacizumab intravenously mg/kg every other week. response was assessed by response evaluation criteria in solid tumors (recist). the primary endpoint was progression free survival (pfs). results: eighteen pa- tients were enrolled. median age was yo, all had good performance status kps of or , and patients had received or prior chemotherapy regimens. median pfs was . months. there were no complete or partial re- sponses; patients had stable disease for > months. tox- icity was substantial with ( %) patients reporting grade toxicity. seven ( %) patients discontinued therapy due to adverse events including hypertension (n= ), gi toxicity (n= ), sensory neuropathy (n= ), rash (n= ), pain (n= ) and wound complication (n= ). given the lack of clear efficacy and increased toxicity, accrual was terminated. conclusion: the combination of sorafenib and bevacizumab has substantial toxicity and minimal efficacy in patients with previously treated metastatic breast cancer. further study of this combination is not recommended. keywords bevacizumab . sorafenib . metastatic breast cancer metastatic breast cancer is generally incurable with only a few patients achieving long-term survival after standard chemo- therapy [ ]. though an expanding arsenal of active agents is available for the treatment of metastatic disease, overall sur- vival has changed little during the last half century. extensive laboratory data suggests that angiogenesis plays an essential role in breast cancer development, invasion and metastasis. of the identified angiogenic factors, vascular en- dothelial growth factor (vegf; also known as vascular per- meability factor) is the most potent and specific and is a crucial regulator of both normal and pathologic angiogenesis [ ]. bevacizumab, a monoclonal antibody directed against vegf-a, has modest activity as a single agent but improves response rate and pfs but not overall survival when adminis- tered in combination with chemotherapy. there is also strong evidence supporting a role for platelet-derived growth factor receptor (pdgfr) signaling in breast cancer autocrine and paracrine signaling. pdgfr has been detected in the stromal compartment of many human breast cancers by immunostaining [ ]. high levels supported in part by a breast cancer research foundation grant to kdm l. a. mina : k. d. miller (*) indiana university melvin and bren simon cancer center, indiana cancer pavilion, rt- , barnhill dr, indianapolis, in , usa e-mail: kathmill@iupui.edu m. yu : c. johnson department of biostatistics, indiana university school of medicine, indianapolis, in, usa c. burkhardt hoosier oncology group, indianapolis, in, usa r. zon northern indiana cancer research consortium, south bend, in, usa invest new drugs ( ) : – doi . /s - - - of immunostaining for pdgf receptor ligands, pdgf-a and pdgf-b, have been commonly detected directly on breast cancer cells in archival human tumors [ ]. sorafenib is a multikinase inhibitor targeting several serine/threonine and receptor tyrosine kinase receptors in- cluding vegf, pdgf and the mitogen activated protein kinases (mapk) pathway [ ]. sorafenib monotherapy ( mg twice daily) had little activity in patients with previously treated metastatic breast cancer in two previously reported trials [ ]. the addition of sorafenib to chemother- apy drugs is currently being investigated in the ties (trials to investigate the effects of sorafenib in breast cancer) program. two of these studies have already been completed and presented—the nu b and the solti- trials. the solti trial reported by baselga et al. showed a signif- icant clinical benefit with the combination of capecitabine and sorafenib when compared to capecitabine monotherapy, pfs of median, . vs. . mo; hazard ratio . ; -sided p= . . this was achieved at the expense of more toxicity. the nu b trial reported by gradishar et al., compared the safety and efficacy of sorafenib in combination with paclitaxel vs paclitaxel with placebo. results showed no improvement in pfs which was the primary endpoint [ – ]. considering that pro-angiogenic peptides are produced in increasing numbers as tumors progress, we hypothesized that combined inhibition of multiple angiogenesis pathways, whether serial or parallel, may increased activity. results of a phase i trial of a combination of bevacizumab and sorafenib in patients with a variety of tumour types (azad et al., a, b) has already been reported including ovarian cancer. their results showed an unexpectedly high partial response rate (pr ⁄ %) in patients with relapsed eoc compared with – % response rate reported with bevacizumab alone [ ]. our trial evaluates combined vegf inhibition with sorafenib and bevacizumab in pre- treated patients with advanced breast cancer. patients and methods patient eligibility eligible patients, ≥ years old, had histologically con- firmed adenocarcinoma of the breast with evidence of met- astatic disease, measurable according to recist . criteria. patients were also required to have an ecog per- formance status of or as well as adequate hematologic, renal and hepatic function. patients with her positive ( + by immunohistochemistry or gene amplification by fluores- cence in situ hybridization) disease must have received prior trastuzumab therapy. patients could not have received more than prior chemotherapy regimens, considering all adju- vant and neoadjuvant therapy as one regimen. prior hormonal therapy was allowed. prior radiation therapy was allowed as long as the irradiated area is not the only source of evaluable disease. the study was performed in compliance of good clinical practice, the helsinki declaration, and federal and institu- tional guidelines. ethical review boards at each institution approved the trial protocol prior to patient enrollment. all patients provided written informed consent. treatment plan sorafenib was administered orally at mg daily; sorafenib dose could be increased to mg twice daily in patients without significant toxicity in the first two cycles. bevacizumab was given intravenously at mg/kg every other week. one cycle was considered to be days. treat- ment continued until disease progression or unacceptable toxicity intervened. toxicities were evaluated using the common terminology criteria for adverse events (ctcae) version . patients were evaluated clinically every week for the first cycle than every other week for the subsequent cycle. hypertension was mon- itored weekly during the first cycle, then prior to each bevacizumab infusion. urine protein creatinine ratio or dip- stick urinalysis was performed every weeks. complete blood count and serum chemistries were obtained before each cycle. no bevacizumab dose reduction was allowed and no dose modification was specified for grades / toxicities. bevacizumab was discontinued for any arterial thromboem- bolic event, all grade non hematologic toxicities, and grades and allergic reactions. bevacizumab therapy was held for other clinically relevant grade toxicities, then resumed upon resolution or adequate control. sorafenib was held for all grade / toxicities and, depending on clinical severity, either permanently discontinued or restarted at a lower dose upon resolution. disease status was assessed according to recist criteria at the end of every third cycle. statistical considerations the primary endpoint was pfs, defined as the time from first treatment to disease progression or death from any cause. the study was designed to detect an improvement from median pfs of months based on the previous expe- rience with bevacizumab monotherapy to months. assum- ing a type i error of % and % power, patients were required. to account for potential dropouts and unevaluable patients, a total sample size of patients was initially planned. based on the toxicity experience in the first patients, an unplanned analysis of pfs was requested. when this failed to suggest significant benefit for the combined regimen, the study was terminated. invest new drugs ( ) : – secondary endpoints included overall response rate (orr), clinical benefit response defined as the percentage of evaluable patients with confirmed complete response (cr), partial response (pr), or stable disease (sd) for at least months (cr+pr+sd), and safety. the median and range of continuous variables are presented. frequencies and per- cents of categorical variables are provided. exact binomial % confidence intervals were calculated for response. the method of kaplan-meier was used to calculate median dura- tion of follow-up, survival, and progression free survival [ ]. results patient characteristics from october to march , patients were enrolled. patient characteristics are shown in table . nearly half of the patients had triple negative (er-, pr-, her -) disease, and most had received or prior chemotherapy regimens. efficacy median pfs was . months, % ci ( . – . ). there were no complete or partial responses. six patients had stable disease but only three ( %) were stable for months or more. three patients died within days of last therapy. toxicity in general, this toxicity was substantial. although no grade toxicity was noted, ( % of patients reported grade toxicity (table ). hypertension was common with ( %) patients experiencing grade or toxicity. gi toxicity, sensory neuropathy, rash, and pain were also relatively com- mon. seven patients discontinued therapy due to toxicity; reasons for treatment discontinuation were uncontrolled hy- pertension (n= ), nausea (n= ), neuropathy (n= ), pain (n= ), rash (n= ) and wound complications (n= ). one patient had a grade hypersensitivity reaction; however patient was able to continue with treatment with premedication. no he- matologic toxicities were reported, however elevation in the liver transaminases were noted in two patients. discussion this report summarizes the results of the study of the com- bination of sorafenib with bevacizumab in previously treat- ed patients with metastatic breast cancer. our results show that the combination of the two angiogenic inhibitors in this patient population is poorly tolerated. all patients experienced some kind of toxicity and in half of them the toxicity was reported to be grade . the trial was stopped early because of the poor safety profile as well as the lack of significant efficacy of this combination over bevacizumab alone in the short follow up period where it was evaluated. the significant toxicity of this combination questions the safety of dual vegf blockade or vegf/pdgf inhibition. this problem had been encountered in the earlier phase i studies of sorafenib and bevacizumab where a recurrent theme was the inability to dose escalate those drugs and where the maximal tolerated dose ended up having to be the table patient characteristics (n= ) median range age years – years median mean number of cycles . n % gender female . male . race caucasian . african american . ecog performance status . . breast cancer subtypes er-/pr-/her − . er-/pr-/her + . er+/pr-/her − . er+/pr+/her − . er+/pr+/her unknown . prior chemotherapy regimens . or prior hormonal therapy . prior radiation therapy . table treatment-related toxic effects toxicity category grade grade grade total n % n % n % n % gi . . . derm . . . . cv . . . neuro . . . lab . . . . other . . . . invest new drugs ( ) : – initial dose which is usually much lower than the efficacious dose [ , ]. when it comes to double angiogenesis block- ade, our experience also mirrors some of the other combi- nations like with the combination of bevacizumab with sunitinib. several studies with this combination in solid tumors had to be halted secondary to significant side effects including anemia, myelosuppression, fatigue and gastroin- testinal complications. in a phase i study by feldman et al., sunitinib and bevacizumab was tried in patients with ad- vanced renal cell carcinoma, again significant toxicity was encountered, with similar toxicity profile with the one we encountered with hypertension, fatigue and gi problems being very common [ ]. other studies reflect the same reality. in a recent phase trial the sabre-b study evalu- ated the triple combination of paclitaxel, bevacizumab and sunitinib in metastatic breast cancer, that study accrual was also stopped because poor tolerability and inability to dose escalate sunitinib as planned secondary to the toxicity pro- file [ ]. many other trials reproduced similar results [ , ]. the broad spectrum of signaling inhibition of the tki’s brought increased toxicity with dose reductions and limited drug exposure that ultimately, we believe, hindered activity. overall results with vegf tkis in mbc have been disappointing. the broader spectrum of activity had provid- ed hope for greater efficacy. however, with that broader profile comes increased toxicity, need for dose reductions and the general inability to combine with chemotherapy. in conclusion drug development is a challenging task; the combination of biologics is very unpredictable and could be misleading. a deeper understanding of the biology of breast cancer and its underlying molecular identity is warranted prior to embarking into further targeted therapy trials. conflict of interest lida a mina, md.: no conflict of interest cynthia s johnson: no conflict of interest cynthia burkhardt: no conflict of interest menggang yu: no conflict of interest robin zon, md: no conflict of interest kathy d miller, md: honoraria to disclose: yes genentech research funding to disclose: genentech and trial supported by bayer open access this article is distributed under the terms of the creative commons attribution license which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited. references . greenberg pa et al ( ) long-term follow-up of patients with complete remission following combination chemotherapy for met- astatic breast cancer. j clin oncol : – . ferrara n, davis-smyth t ( ) the biology of vascular endo- thelial growth factor. endocr rev : – . bhardwaj b et al ( ) localization of platelet-derived growth factor beta receptor expression in the periepithelial stroma of human breast carcinoma. clin cancer res : – . coltrera md, wang j, porter pl, gown am ( ) expression of platelet-derived growth factor b-chain and the platelet-derived growth factor receptor beta subunit in human breast tissue and breast carcinoma. cancer res : – . huang j, li x, hilf r, bambara ra, muyan m ( ) molecular basis of therapeutic strategies for breast cancer. curr drug targets immune endocr metabol disord : – . moreno-aspita a, hillman d, wisenfeld tj, et al ( ) bay – as single oral agent in patients with metastatic breast cancer previously exposed to anthracycline and/or taxane. asco abstract . baselga j, gianni l, gradishar wj et al ( ) phase iib double- blind, randomized, placebo-controlled trials for the efficacy and safety of sorafenib in patients (pts) with metastatic or locally advanced breast cancer (bc): review of the trials to investigate the effects of sorafenib in bc (ties) program. presented at: asco annual meeting. chicago, il, usa, – june . gradishar wj, kaklamani v, prasad sahoo t, et al ( ) a double- blind, randomized, placebo-controlled, phase b study evaluating the efficacy and safety of sorafenib (sor) in combina- tion with paclitaxel (pac) as a first-line therapy in patients (pts) with locally recurrent or metastatic breast cancer (bc). presented at: nd annual san antonio breast cancer symposium. decem- ber – , . abstract . baselga j, grupo español de estudio tratamiento y otras estrategias experimentales en tumores sólidos, roché h et al ( ) solti- : a multinational double-blind, randomized phase iib study evaluating the efficacy and safety of sorafenib compared to placebo when administered in combination with capecitabine in patients with locally advanced or metastatic breast cancer (bc). presented at: ecco and th esmo multidisciplinary congress. berlin, germany, – september ( ) (abstract lba). updated at: nd annual san antonio breast cancer symposium. san antonio, tx, usa, – decem- ber (abstract ). demonstrates significant activity of sorafenib in combination with capecitabine . azad ns et al ( ) combination targeted therapy with sorafenib and bevacizumab results in enhanced toxicity and antitumor activ- ity. j clin oncol : – . kaplan el, meier p ( ) nonparametric estimation from incom- plete observations. j am stat assoc : – . sosman j, flaherty k, atkins m et al ( ) updated results of phase i trial of sorafenib (s) and bevacizumab (b) in patients with metastatic renal cell cancer (mrcc). j clin oncol ( s):a . feldman dr et al ( ) phase i trial of bevacizumab plus esca- lated doses of sunitinib in patients with metastatic renal cell carci- noma. j clin oncol : – . mayer el et al ( ) sabre-b: an evaluation of paclitaxel and bevacizumab with or without sunitinib as first-line treatment of metastatic breast cancer. ann oncol : – . rini bi et al ( ) a phase i study of sunitinib plus bevacizumab in advanced solid tumors. clin cancer res : – . socinski ma, scappaticci fa, samant m, kolb mm, kozloff mf ( ) safety and efficacy of combining sunitinib with bevacizumab+paclitaxel/carboplatin in non-small cell lung cancer. j thorac oncol : – invest new drugs ( ) : – 	a... 	abstract 	patients and methods 	patient eligibility 	treatment plan 	statistical considerations 	results 	patient characteristics 	efficacy 	toxicity 	discussion 	references setting the stage: developmental biology in pre-college classrooms sandra borland*, , karen crawford, and victoria brand, indiana university axolotl colony, bloomington, indiana, st mary’s college, st. mary’s city, md, usa and osborn middle school, phoenix, arizona, usa. abstract exercises that employ dynamic living material have proved highly successful at generating interest in science among young students. developing embryos and larvae are espe- cially well suited for such endeavors, for they can be handled without expensive or elaborate equipment, and their changing nature engages students. using amphibian embryos, which are relatively large and exhibit profound, easily observed morphological changes, and amphibian larvae, which are easily kept and observed, captures the attention of children. by designing inquiry- based exercises and focused discussion sessions, a high intellectual content can be integrated into these endeavors. the long-term implications for generating an informed citizenry, improving the participation of women in science, and empowering elementary school teachers are profound. professional developmental biology researchers should feel encouraged to participate in these types of activities. key words: lesson, education, inquiry, elementary school, classroom, axolotl, regeneration int. j. dev. biol. : - ( ) - / /$ . © ubc press printed in spain www.ijdb.ehu.es *address correspondence to: dr. sandra borland. indiana university axolotl colony, bloomington, in , usa. fax: + - - - (c/o axolotl colony). e-mail: sjborlan@indiana.edu background information scholarly interests of the authors sandi borland is the outreach coordinator for the iu axolotl colony. in her tenure there, she has developed an extensive outreach program to the national pre-college educational commu- nity. she has developed materials for students and teachers who want to use the mexican axolotl (ambystoma mexicanum) in their science programs. she has worked in the schools and at the axolotl colony with thousands of children, in addition to giving workshops for indiana teachers to learn how to use axolotls in science and cross-curricular activities, especially using the inquiry method to allow students to get the most out of science. karen crawford is a professor of biology, with research interests in amphibian appendage regeneration. she has transferred some of her interests in science to the children in her own family and to young school children in a local school. her goal is to generate intellectually stimulating experiences that require minimal expen- ditures of time, energy, and monetary expense. she has devised limb regeneration exercises that have proven to be highly effective for both the students and their teachers. an example of her research interests can be found in the following: crawford, k., ( ). ooplasm segregation in the squid embryo, loligo pealeii. biol. bull. ( ): - . crawford, k., ( ). culture method for in vitro fertilization to hatching of the squid, loligo pealeii. biol. bull. ( ): - . victoria brand is a collaborative peer teacher in science for - to -year-old students. initial experience with outreach efforts produced enough enthusiasm and ideas among the teachers and students at her school to inspire follow-up projects in subsequent years, and she and her fellow teachers have developed a whole curriculum unit from the initial project. changes in shape, size and color capture the imagina- tion of young students introducing young children to the life sciences through study of developmental biology (though not often called developmental biol- ogy in pre-college situations) is a relatively common strategy. in the united states, students study biological life cycles as part of the science standards set by national- and state-level school administra- tive committees. cocoons are brought into the classroom for stu- dents to watch the emergence of a butterfly or moth. mice, hamsters, and gerbils are housed as pets, and rodent pups are reared (even if unintentionally) as class projects. young children watch such events and are fascinated as they observe changes. then they ask questions. the use of living material thereby serves as a jumping- s. borland et al. off point for scientific inquiry. a natural interest in changing shapes, sizes, and colors leads young students to ask questions based on their observations and to offer predictions based on their prior experiences. this process in turn leads to a search for answers, which are obtained by asking a teacher or expert, doing an internet search, or reading available literature. this scenario usually occurs on its own, regardless of whether it is part of a formal lesson. studying developmental biology is integral in the natural study of living things. it is a study of beginnings, of change, and of structure, process, and purpose. little wonder that it captivates the intellectual energy of students. use of living material in the public school classroom is often the most cost-effective strategy for generating interest in the biological sciences the u.s. national research council (an advisory board to the u.s. government) has emphasized that: biological science that is presented to elementary school children should have natural history as a major focus, be integrated with other subjects wherever possible, and emphasize observation, interpretation, and hands- on involvement, rather than memorization of facts… students... should observe and care for organisms in the classroom (national research council, committee on high school biology education. [ ]. fulfilling the promise: biology education in the nation’s schools. national academy press, washington, d.c.). live material with accompanying supplies can be obtained from various biological supply companies (e.g., nasco scientific, ward’s biological supply, carolina biological). many materials can be obtained locally or even collected from nature (at no cost). supporting materials and activities for classrooms can usually be found on the internet. the teacher doesn’t have to take a course to teach with these materials. they can choose to learn with their students, demonstrating by their own curiosity and questioning the love of inquiry and discovery. in the united states, the usual teacher-training programs for elementary school teachers empha- size language and math teaching skills, at the expense of science skills. teachers who elect to use living material in the classroom find themselves empowered, and easily able to compensate for lack of formal science training by engaging and joining students in relatively straightforward yet intellectually stimulating exercises. in a study of ten-year-old students, it was observed that using small, live animals increased interest in and motivation for study- ing biology (tamir and schurr, ). using living material in the early classroom has a subtle, yet predicted influence on career choice among young girls (fig. ). girls are thought to make important career decisions in the late elementary school years (ages - ) (orenstein, ). elemen- tary school teachers in the united states are overwhelmingly female and provide role models for young girls. these teachers, therefore, need to be enthusiastic and involved in their science teaching. when teachers are engaged, students tend to follow suit. girls are thought to develop more confidence in their science abilities when they are engaged in hands-on exercises that spark their curiosity and give them an embodied experience (rosser, ). in this manner, a long-term beneficial effect on closing the gender gap in science is likely to result. information-age opportunities are increasingly avail- able to young students in today’s world, the internet has greatly enhanced the access children have to resources at all levels of complexity. a student in elementary school can learn about pigment mutations in amphibia by using a search engine, or, log onto one of the many attractive science sites designed especially for young students. students aged - can research a genetic disease and become ac- quainted with ideas and jargon concerning its causative basis. in addition, as a result of increased outreach efforts of u.s. colleges and universities, materials and personal expertise are now available to some of our pre-college classrooms. in many instances, teachers and students can learn science directly from practicing scientists. during such collaborations, students learn about authentic research projects while developing an under- standing and respect for the power of the scientific method. research itself can become part of the lesson and is then viewed as a practical, accessible, and necessary part of their world. three examples of projects that introduce young chil- dren to developmental biology three successful outreach programs are described below that illustrate the profound effect college/university outreach programs can have on the learning experiences of young children. these descriptions are designed to encourage pro- fessional developmental biology researchers to get involved with public school science classes. the motivating effect re- searchers can have on both young students and their teachers is very powerful. the first two descriptions explain on-going outreach programs for pre-school through elementary school age children, while the third description explains an on-going middle-school project. fig. . a five year-old child, observing the living axolotl embryo at the iu axolotl colony. photo courtesy of sue hicks and jitka horne, hoosier courts cooperative nursery school. developmental biology in the pre-college classroom what am i going to do with all these axolotls? by sandra borland for more than years, the indiana university axolotl colony has developed collaborations with many schools and teachers, from preschool through pre-college levels. the positive effects of this program have been noted and even documented over the years by many teachers and administrators and their students. for example, ginny coppedge, a local teacher of - year olds, has explained the manner in which students who were indifferent to school and learning became enthusiastic not only about sci- ence but also about writing and drawing as well (coppedge, ). similar stories have been heard from public school science teachers using axolotls in their classrooms at all levels. those teachers were eager for something different to stimulate interest in students, who often complain of being bored in the classroom. often these teachers are also trying to find an interest in and method for teaching science themselves. in the united states, pre-school and early grades are often ignored with regard to learning science. our culture emphasizes math and language development at early ages, often at the exclusion of science. even now, with new education guidelines that require science be taught at the early grade levels, science is punched and prodded into curricula that will still place greatest stress on math and language. unfortunately, this does not culti- vate the natural interest of young students in the world around them. their world is real, not abstract, like math studies and reading sometimes appear to be. the ideal is to use science to bring math and reading into the real world. let the natural interests of the students drive their desire to know how to read and mathematically analyze. this is what many teachers in early grades have done with our axolotls. in the course of the i.u. axolotl colony’s normal breeding program, which is entirely self-sustaining and supplies materials to labs all over the world, we accumulate surplus animals and embryos. although the availability of surplus material did not stimulate the outreach, it did make it easy to participate in outreach activities. we had animals to spare and teachers who wanted something different for their classrooms, for their stu- dents, and for themselves. consequently, axolotl aquaria have now been set up in classrooms in many schools around the united states. initially, we brought children into the colony for tours (fig. ). we set up axolotl life-cycle demonstrations and collected repre- sentatives of other amphibian and reptilian species. we prepared booklets with collections of axolotl art and information for students to take home and share with their families. we discussed the axolotl’s status as an endangered species and issues of conser- vation and the environment. we reviewed anatomical structures and purpose: what make axolotls special? how do they grow and change? what makes axolotls different from other amphibians and from reptiles, mammals, birds and fish? and we answered questions, endless questions! anyone who has worked with young children knows that keeping them engaged and interested in the same topic for minutes is extraordinarily difficult. never- theless, axolotls succeeded in holding the attention of even the youngest students. i found that if i let the students’ questions drive the discussion, we could hold their attention for longer than minutes. students are so interested that they would even listen to each other’s questions! teachers very often ended up thrilling their students by adopting an axolotl or two for their classrooms. a year later we sometimes received messages and questions concerning the now adult-size adoptees, now loved by the next year’s crop of students. very often we received panicked yet excited calls and e-mails from teachers whose now grown axolotls had just produced several hundred eggs. they were amazed, their students were thrilled, and the event often created excite- ment throughout the school and the community. many teachers fig. . (left) pre-school teacher, jitka horne, learns about developing axolotl embryos with her students. photo courtesy of sue hicks and jitka horne, hoosier courts cooperative nursery school. fig. . (right) using living material as a focus is early education classrooms has wide appeal for more than basic science. children discuss, draw, collaborate, observe, predict and discover as they get experience in basic biology. photo courtesy of sue hicks and jitka horne, hoosier courts cooperative nursery school. s. borland et al. study habitats (natural and artificial) experiment with environment preferences collect data on growth rates compare axolotls with different pigments observe locomotion compare with other species and animal groups discuss metamorphosis and “growing up” discuss ability to grow new parts and interest of scientists in this process map natural location of axolotl on world map observe pollution in local streams and relate to axolotl habitat read about axolotls in culture and history create myths/stories about axolotl creation today, we have axolotl developmental biology kits available for teachers to purchase at a nominal cost. the idea behind the kits was to make it easy for teachers to have as large or as small an axolotl project as they can handle, with the supplies and informa- tion at hand to get them started. kits can be ordered with embryos or larvae and come with basic supplies (i.e., water treatment chemicals, food) and information on development and care and safety for students and animals. the inquiry model is ideal for exploring the developmental biology of axolotls. thus it is important not to serve up too much information but instead to allow students to think and ask and discover for themselves. their observations are often astounding, and they sometimes notice significant details that teachers often miss. for instance, when some students try to portray axolotls by drawing, anatomical structures are sometimes included that adults fig. . graphic representations by preschoolers tell us much about their abilities to observe and relate to their world. here, not only do we see surprisingly accurate drawings of developing axolotl embryos, but you see that the child is involved with that world as part of her own. sketch courtesy of sue hicks and jitka horne, hoosier courts cooperative nursery school. fig. . researcher karen crawford discusses axolotl regeneration with elementary students. and their classrooms have been reported in their local newspa- pers. other teachers from the schools or the region would contact those teachers or the colony so that they could do the same thing in their own classrooms. such is the effect of natural science, and, in this case, of axolotls, on stimulating interest in students, their families, and teachers. eventually, we began to travel with our axolotl material to class- rooms and school community science fairs. time after time, i found that if i let the students guide the discussion through questions and comments, the seriousness and involvement of each individual student increased. they observed and asked questions. i answered their questions and listened to many of their stories. more and more teachers have discovered the joy of teaching with axolotls. below is a list of some of the activities that can be done with young students (figs. , ): early axolotl embryos count embryos manipulate and divide embryos into sets sort embryos by stage of development prepare staging series (which comes first, second, third, etc.) observe effects of gravity experiment with effects of temperature on developmental rate; collect data from test groups and represent differences graphi- cally use simple dissecting scope or magnifying glass to observe embryo structures and changes observe beating heart and blood moving through the cham- bers observe circulation in gills of near-hatchlings draw pictures of observations or changes in a lab notebook axolotl larvae observe development of limbs experiment with the effects of temperature, type and quantity of food, and other parameters on growth rate developmental biology in the pre-college classroom allow the students to tell their stories about receiving anesthetic. place the axolotl (~ cm in length) in . % benzocaine in % holtfreter’s solution. while discussing regeneration, amputation, and animal care, pass the container with the animal in it around for the students to observe. . introduce the idea of regeneration and metamorphosis. this stimulates stories of lizards losing tails, crabs without claws, starfish without arms, even uncles growing fingertips back. en- courage students to share their stories with the entire class. . describe the amputation procedure, showing students the scis- sors that will be used. answer questions about blood, and assure everyone that there will be very little shed. . once the animal is “asleep” (~ minutes), amputate one hand. show each student the animal’s amputated hand. . return the axolotl to % holtfreter’s solution, answer ques- tions, and wait for the students to notice the animal has awakened. . review animal care, feeding, cleaning, and how to make obser- vations. follow-up and (~ minutes each) . after weeks, i returned to the classroom to check on how things were going, answer questions and show the class some other axolotls (representing different pigmentation mutants). the amputated axolotl, now named “nippy,” was doing well and had a nice late bud blastema. . after a month, nippy had developed four digits and was growing well. the students decided to keep nippy until everyone had a chance to feed, clean, and record regeneration for the class. conclusions . regeneration to the digit stage was complete in weeks. . the students learned a lot about regeneration, scientific obser- vations, and animal care. fig. . handout used with elementary students comparing regeneration in different amphibians at different stages in the life cycle. never notice. also, artwork reveals concepts of relationships, parenting, nature, structure and purpose (fig. ). drawing axolotls has helped teachers learn about their students, not just about axolotls! when teaching science, it is important not to underestimate the information young students are ready to process and understand. correlations and comparisons are how children learn from the very beginning of life. this should continue throughout development and should be the guiding principal helping children understand the interrelatedness of the species and the environment, a complex yet logical process for a child. elementary students study regeneration with a re- searcher by karen crawford over a period of one month, students in a montessori (private elementary school) combined classroom for – year olds ob- served normal hand regeneration on an adult axolotl. montessori schools are privately operated around the world and emphasize, among other principles, the notion that children learn best from engagement, and that cooperation (vs. competition) between students fosters student learning. thus, this experimental exercise was well received by students, teachers, and parents. the first day, while the animal was being anesthetized, we discussed using anesthetic, regeneration in amphibians and other animals (fig. ), general laboratory safety, feeding and animal care, and making regular observations (fig. ). since they had previously drawn the phases of the moon over a month’s time in the fall, i distributed a calendar to each student for them to draw the phases of regeneration. once the animal was anesthetized, it was passed around in its container for the students to observe. next, it was placed on a plate and its right front hand was amputated using small scissors. upon returning the animal to its water, the students observed it as it “woke up.” the axolotl re- mained in the classroom for one month. stu- dents took turns caring for it, drew pictures of the regeneration process, and compared their pictures with illustrations i provided. at weeks, i returned to the classroom to check on how things were going and to answer questions. at that time, the limb blastema was at the late bud stage. by one month, four digits had formed. the students kept the animal until each of them had had a chance to care for it. at weeks, i met with the class again to review their data and the outcome of our experiment. benefits? their observations and questions were amazing! the students and parents now seek me out to tell me extraordinary things. science came alive in this classroom. rather than dealing with an abstract phenomenon, this exercise had a profound and personal meaning for each and every student. the procedures we employed were as follows: day (~ . hours duration) . introduce the animal, describe where it lives and why it is unique. answer questions. . discuss anesthetic and why we use it. s. borland et al. . a favorable memory was achieved. i am frequently approached by students throughout the school, who talk to me about their thoughts and observations about science, especially the girls! . when possible, follow-up visits enhance learning and future connections with students. . first, second, and third graders are an enthusiastic and willing bunch of scientists! the juvenile salamander in the middle - school class- room by victoria brand what do you get when you cross juvenile salamanders (a. tigrinum) with excited inner city - year olds? well, if you add to the mixture two dedicated teachers, ten-gallon tanks with assorted humming water pumps, shipments of worms, and a definite “what is that smell?” you end up with the osborn middle school “water dog unit.” over the past years, students have learned about “real science” by doing a variety of exercises with our axolotls. getting our program started - an unlikely beginning four years ago, a senior biology undergraduate at arizona state university (tempe) wrote a grant proposal that was intended to support the salamander brain research being conducted at the university by creating a source for salamanders. the idea of the proposal was to find a middle or secondary school willing to raise juvenile salamanders to maturity and then send them along to the university’s research laboratories. when our science collaborative peer teacher heard of the project, she recognized it as a golden opportunity for our middle-school students to conduct scientific research. additional grants were written and eventually funded by the u.s. national science foundation and petsmart (a retail store which specializes in pet supplies). the university grant paid for the first batch of youngsters. the seventh grade science teachers developed an instructional unit which would support the state of arizona standards for teaching science and which would also accomplish the university’s goal of being supplied with large numbers of metamorphosed salamanders. problem solving as a student endeavor how to get large numbers of juveniles to metamorphose? that became the scientific question for the class. it was up to the students to develop hypotheses and conduct appropriate re- search. what wonderful ideas and year olds can devise! the first year our animals had a massive fungal infection. students used their computers to e-mail the university, called on a local exotic animal veterinarian, and surfed the internet to research information to help them combat that plague. we now have a gallon of rid ick (antifungal medicine) available at all times. another teacher developed a wonderful lesson on fungus to go with the unit on salamander growth. over the years, both teachers have infused technology throughout the lessons. one teacher developed a lesson to measure the amount of oxygen elodia infuses into water. while we have never become the salamander factory intended (the students have not quite discovered the key to the puzzle), the relationship between osborn middle school and the university of arizona has grown, and the project lives on, impacting the lives of hundreds of seventh graders every year. a typical day in aquarium central a classroom day usually begins with several students arriving before classes begin to check on their “dogs” (as they are fondly referred to). students can be observed feeding, cleaning, observ- ing their subjects and discussing any changes they have discov- ered (fig. ). when the bell rings to begin class, students are given instructions at the beginning of the -minute period. lessons ranged from computer research of the characteristics of the sala- mander and its habitat, to a math lesson where students calculate the cost of the entire operation. the remaining part of each day involves the students carefully removing their animals from their tanks and taking measurements in both grams and centimeters that are carefully entered into tables the students have prepared in their lab books. visual observations are recorded to see if there are notable changes in tail length, gill length, legs, or coloring. students examine their animals to determine whether the experimental variable they have applied has caused the process of metamor- phosis to commence. while students are working on their projects, teachers walk around the classroom asking questions and listening to excited students revelations. occasionally there is an animal fatality, and the specimen is carefully placed in a plastic bag and frozen for use by our local wildlife rescue organization. project outcomes by the end of the unit, some animals will have metamorphosed while others remain as juveniles. most are adopted and find homes with their seventh grade scientists. we have some still living after years, as reported by students who return to our classroom to share their memories. the rest are returned to the local vendor. the groups create powerpoint presentations that illustrate the hypothesis they were testing as well as the data and the results of the experiment. they share this with each other in a mock “conven- tion of scientists.” they also create an excel (computer) document that graphs their measurements over time to insert in their presen- tation. forty of the students are selected—by application—to attend an all-day field trip to arizona state university to share their findings with college students and professors and to have a chance to view labs and see the university’s animal facilities. fig. . middle-school students (aged - ) weigh and measure juvenile axolotls as part of a science lessons studying metamorphosis in amphibians. developmental biology in the pre-college classroom the students leave our class with an increased knowledge of the scientific method, an understanding of metamorphosis, and new questions. as in real life science, sometimes the variable chosen has no effect, and that is an acceptable result. after years, we have not found any consistent variables to report, but we are starting to see possible patterns. interestingly, adding food color- ing has had a rather consistent effect in a number of tanks. the students thought perhaps the threat of water contamination caused the animals to accelerate changes? some of the most valuable aspects of this project include the students’ enhanced ability to collaborate, the sparkle that comes on in a child’s eyes when the idea of a college education becomes a real goal, and how they leave our class respecting living beings in a world where life is often an expendable commodity. references coppedge, v. ( ). axolotls in the second grade. int. j. dev. biol. : - . orenstein, p. ( ). schoolgirls: young women, self-esteem, and the confidence gap. doubleday, new york. rosser, s.v. ( ). re-engineering female friendly science. new york teacher’s college press. tamir, p. and schurr, y. ( ). back to living animals: an extracurricular course for fifth-grade pupils. j. biol. educ. : - . wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ this is a peer-reviewed, post-print (final draft post-refereeing) version of the following published document, this is the peer reviewed version of the following article: wynn, neil a ( ) letters from the greatest generation: writing home in wwii. edited by howard h. peckham and shirley a. snyder, forward by james h. madison. indiana university press. . xi + pp. $ . . history, ( ). pp. - . issn , which has been published in final form at https://doi.org/ . / - x. . this article may be used for non-commercial purposes in accordance with wiley terms and conditions for self- archiving. and is licensed under all rights reserved license: wynn, neil a ( ) letters from the greatest generation: writing home in wwii, edited by howard h. peckham and shirley a. snyder, forward by james h. madison, indiana university press, . history, ( ). pp. - . doi: . / - x. official url: https://doi.org/ . / - x. doi: http://dx.doi.org/ . / - x. eprint uri: http://eprints.glos.ac.uk/id/eprint/ disclaimer the university of gloucestershire has obtained warranties from all depositors as to their title in the material deposited and as to their right to deposit such material. the university of gloucestershire makes no representation or warranties of commercial utility, title, or fitness for a particular purpose or any other warranty, express or implied in respect of any material deposited. the university of gloucestershire makes no representation that the use of the materials will not infringe any patent, copyright, trademark or other property or proprietary rights. the university of gloucestershire accepts no liability for any infringement of intellectual property rights in any material deposited but will remove such material from public view pending investigation in the event of an allegation of any such infringement. please scroll down for text. peckham, howard h., and snyder, shirley a. letters from the greatest generation: writing home in wwii, (bloomington & indiana: indiana university press, ). $ . xi, . . collected by the indiana war history commission and first published in as letters from fighting hoosiers, this collection of letters selected from , sent by men and women from indiana who served overseas during world war ii provides insights into the impact of war on ordinary people. mail to, and from, home was important for the morale of both service personnel and those at home, even though it was censored by military and by the writers themselves. the letter writers often spoke of the importance of mail from home, but we just have their side of the correspondence and we can only imagine what they meant to their recipients. the collection is organized by location covering various different theatres in europe, africa, and asia. military service enabled americans to see the world and its sights almost as tourists: as one wrote, “it isn’t everyone who gets to visit paris sometime during his lifetime.” ( ) paris was just one of many exciting places visited – the pyramids, the sphinx, the nile, rome, naples, london, stratford, all enabled the g.i.s to experience another world and to compare situations with those back home. for many this led to the realisation that it was “wonderful to be an american.” ( ) sometimes almost light-hearted such as the opening descriptions of the attack on pearl harbor simply as an “air raid” ( ) or a surprise visit from “our small yellow friends”( ), almost all the letters are extremely well-written (one was a poem), and they were presumably selected partially for this reason. the letters vary in length from several detailed pages to single paragraph missives – in one short letter an airman tells his parents he is alive in a prison hospital, but has had a leg amputated. the more detailed correspondence describes the minutiae of daily life and routine in the military, through from training to combat itself. accounts of battle, being under fire, or witnessing a french firing squad shoot a collaborationist, are often given in surprisingly matter-of-a-fact language, possibly reflecting “the callousness a man develops toward death” ( ); a marine presents a powerful description of the fear one feels under shelling, but follows it immediately with “i wish i could be with you this christmas.” ( ) rarely did the war seem glamorous as many letters refer to being cold, dirty, but comment positively on the food supply – a parachutist recalled the t-bone steaks, potatoes, green beans, coffee and pudding eaten before dropping into france and a female red cross worker in france outlined a monthly liquor ration of champagne, scotch, gin, and cognac. better still was the seven-day furlough in the united states riviera recreation area where one soldier reported he and his colleagues enjoyed fine food and entertainment including glenn miller’s band. on the other hand, there are several accounts here of german prisoner of war camps in which hundreds of dead soldiers, polish and russian, were found starved to death or murdered, and details of the concentration camps that writers felt necessary to share with those back home so that, in one writer’s words, “if enough people understand what totalitarianism is, they will guard against it at home.” ( ) letters from the pacific theatre tended to have a different tone from those from europe – the terrain and the combat often seemed grimmer and there was an obvious sense of relief when it came to an end. the process of selection and editing obviously means that these letters may not be representative of the ordinary serviceman and woman’s views and experiences; they are almost too well-written. nonetheless, they still provide the personal and emotional responses to the war often lost in standard historical accounts. the reality of war is here in the letters and the brief biographies of the writers also often make tragic reading as several of the authors did not survive the war. included is a moving letter written to his family by an airman to be read in the event of his death – he was killed in action a year later. the only illustration in the book is a photograph of a twenty-three year old soldier who died in the battle of the bulge the day after his photograph was taken. neil a. wynn university of gloucestershire reprinted from . •. forestecologyand ; _management forest ecology and management ( ) - expanding the scale of forest management: allocating timber harvests in time and space eric j. gustafson usda forest service, north central forest experiment station, forestry sciences laboratory, highway k, rhinelander, w! , usa accepted april , . e elsevier forest ecology and management aims and scope. forest ecology and management publishesscientificarticlesconcernedwith forest management and conser- vation,and in particularthe applicationof biological,ecological and social knowledgeto the managementof man-made and nat- ural forests.the scope of the journal includesall forest ecosystemsof the world.a refereeingprocess ensuresthe qualityand internationalinterestof the manuscriptsacceptedfor publication.the journalaims to encouragecommunicationbetweenscien- tists in disparate fields who share a common interest in ecology and natural-resourcemanagement, and to bridge the gap between researchworkersand forest managersin the fieldto the benefit of both.the journal shouldbe of interestto research workers, managersand policymakersin forestry,naturalresources,ecologicalconservationand relatedfields. founding editor laurencel. roche, murroe,ireland editors-in-chief -forthe americas,australia,new zealand and the pacific: for the restof the world: r.f. fisher g.m.j. mohren departmentof forest science forestproductionecologygroup departmentof vegetationecology texas a&m university dlo-institutefor forestryand natureresearch college station,tx - , usa p.o. box aa wageningen,the netherlands book review editor d. binkley,forestand wood science,colorado state university,fort collins,co , usa editorial advisory board g. abrahamsen,agriculturaluniversityof norway,_s, j.a. maghembe,sadcc/icraf agroforestryproject, norway zomba, malawi m.r. ahuja, bfh, instituteof forestgenetics, f. makeschin,institutfqrbodenkundeund standortslehre, , grosshansdorf,germany tharandt,germany r. alfaro, canadian forestry service,victoria,b.c., d.c. malcolm, universityof edinburgh,edinburgh,uk canada e. miilk nen, finnishforestresearch institute,vantaa, f. andersson,swedishuniversityof agriculturalsciences, finland uppsala,sweden m.a.r. nahuz, institutode pesquisastecnologicas, p. attiwill,universityof melbourne,parkville,vic., australia s_iopaulo,sp, brazil j.c. calvo, instituteof technology,cartago, costa rica r. p_iivinen,europeanforestryinstitute,joensuu,finland j.d. deans, instituteof terrestrialecology,penicuik, r.f. powers,pacificsouthwestresearchstation,redding, midlothian,uk ca, usa s. diamandis,forestresearch institute,thessaloniki,greece t. pukkala,universityof joensuu,joensuu,finland d.p. dykstra,cifor, jakarta, indonesia l. rasmussen,danishforestand landscaperesearch e.p. farrell,universitycollegedublin,dublin, ireland institute,hoersholm,denmark p.h. freer-smith,forestry commission,farnham,uk d.d. reed, michigantechnologicaluniversity,houghton, o. garcia, engref, nancy, france mi, usa r.a. goyer, louisianastate university,batonrouge, la, g. ruark, usda forestservice, washington,dc, usa usa r. sands, universityof canterbury,christchurch,nz j.b. hall, universitycollegeof north wales, bangor,uk j.a. stanturf,stoneville,ms, usa f. houllier,frenchinstituteof pondicherry,pondicherry,india o. sziklai, universityof britishcolumbia,vancouver, b.c., b.m. kumar,keralaagriculturaluniversity,kerala, india canada j.p. lassoie,cornelluniversity,ithaca, ny, usa k. von weissenberg,universityof helsinki,helsinki,finland j.n. long,utah state university,logan, ut, usa a.e. lugo, internationalinstituteof tropicalforestry, rio piedras, pr, usa publication information: forest ecology and management (issn - ). for volumes - are scheduledfor pub- lication.subscriptionpricesare available uponrequest from the publisher.subscriptionsare acceptedon a prepaid basis only and are entered on a calendaryear basis.issuesare sentby surface mail exceptto the followingcountrieswhere air deliveryvia sa/ mail is ensured: argentina,australia,brazil, canada, hong kong, india, israel, japan, malaysia, mexico, new zealand, pakistan, pr china, singapore, south africa, south korea, taiwan, thailand, usa. for all other countries airmail rates are available on request. claims for missing issues should be made within six months of our publication(mailing) date. please address all your requestsregarding orders and subscriptionqueries to: elsevier science b.v., order fulfilment department, p.o. box , ae amsterdam, the netherlands, tel. (+ - ) , fax (+ - ) . in the usa and canada: for further informationon this and other elsevierjournals please contact: elsevier science inc., journal information center, avenue of the americas, new york, ny , usa. tel. ( ) ; fax ( ) ; telex - aep ui. • back volumes: please contact the publisher. forest ecology • and management elsevier forest ecology and management ( ) - expanding the scale of forest management: allocating timber harvests in time and space eric j. gustafson usda forest service, north central forest experiment station, forestry sciences laboratory, highway k, rhinelander, w , usa accepted april abstract this study examined the effect of clustering timber harvest zones and of changing the land use categories of zones (dynamic zoning) over varying temporal and spatial scales. focusing on the hoosier national forest (hnf) in indiana, usa as a study area, i used a timber harvest allocation model to simulate four management alternatives. in the static zoning alternative, harvests were dispersed throughout the timber harvest land base ( % of hnf) for decades. the three dynamic zoning alternatives varied in the degree to which harvests were clustered in time and space. two levels of harvest intensity were simulated, and at each level of harvest intensity, the area harvested was held constant among all four zoning alternatives. the dynamic zoning strategies resulted in substantial increases in the amount of forest interior and reductions in the amount of forest edge across the landscape, as well as an increase in the average age of stands when harvested. the greatest reduction in fragmentation was produced by the alternative that most tightly clustered harvests in time and space (i.e. intensive harvesting of small blocks in a relatively short time). when harvest intensity was high, this alternative produced amounts of forest interior and edge comparable to those of the dispersed alternative with half the rate of harvest. the results suggest that the injection of dynamics in specifying disturbance regimes, and the clustering of disturbance in time and space_ can be used to sustain larger blocks of mature forest than can static zoning. dynamic zoning encourages explicit specification of the disturbance regimes that will be imposed across the land base over long periods of time. keywords- forest managementplanning;fragmentation;disturbance;forestinterior;forest edge; multiple use; temporal scale; clustering timberharvests;simulation modeling . introduction is mandatedby law to provide for multiple uses and values through the national forest management act forest management has become controversial, of . industrial forest owners have also made a stemming from fundamental differences in how for- commitment to provide multiple forest values in the est resources are viewed by different segments of management of their forest lands (wallinger, ). society. the management of federally owned forests forest management plans typically allocate the land base among several land use categories, and projec- • tions are made of the impacts of the plans on a suite t tel.: - - ; fax: - - ; e-mail: of forest values, including biological diversity, recre- ericgus@newnorth.net, ational opportunities, and commodity production. - / )$ . copyright © elsevier science b.v. all fights reserved. ph s - ( ) - " e.i. gustafson / forest ecology and management ( ) - . because of this goal to provide for multiple uses, ditions of managed forests (crow and gustafson, planners often find themselves attempting to provide ). for mutually exclusive uses of land, such as timber a poorly understood consequence of static zoning production and old-growth forest. a typical model is is that forest age class distributions become skewed to designate several land use categories, and to allo- over long time periods (gustafson and crow, ). cate land to these categories, thus grouping suites of stands in timber production areas are kept in rela- compatible land uses into spatially defined zones, tively early seral stages; other management areas examples of objectives associated with various land experience little disturbance, and the forest in those use categories might be 'a physical setting to provide areas will eventually be dominated by late-seral types. opportunity for solitude and a feeling of closeness to intermediate seral stages should become rare as a nature'; or 'provide for recreation facilities'; or consequence of the deterministic disturbance regime 'maintain habitat diversity, provide a sustained yield imposed by static management strategies over long of timber, and provide dispersed recreation opportu- periods of time, and community composition may nities.' usually several blocks of land (management change markedly. deterministic disturbance regimes areas) are allocated to each land use category, and may reduce the natural variability of landscapes, these blocks are dispersed throughout the forest, resulting in undesired ecological conditions ostensibly to provide the values associated with each (mladenoff and pastor, ; swanson et al., ; category across the landscape. this approach pro- mccarthy and burgman, ). vides for multiple uses at the landscape scale, but a recent trend in us national forest management may not adequately integrate multiple uses within has been a reduction of more than % in timber each management area (behan, ). for example, production since , to the lowest levels since timber production has increasingly been viewed as about (haynes et al., ). this trend has being incompatible with many non-commodity uses primarily been in response to pressure to provide for of the forest, and is often segregated from those uses. more non-commodity values from national forests, forest plans typically consider a -year planning such as wildlife habitat and a natural setting in which horizon. in many cases, it is not possible to provide to experience nature. as an example of this, the all potential uses within a management area over a hoosier national forest in southern indiana amended .. -year period. however, interesting possibilities its forest plan, which emphasized clearcutting arise whenconsidering longer temporal scales, on % of the land base (usda forest service, should the designation of the land use(s) within a ), changing the management emphasis to un- management area be static for long periods of time even-age management.-this amendment reduced the (static zoning), or should it be dynamic, with several expected timber output by % and set aside % of land uses rotating among several management areas the land base for non-commodity purposes (usda (dynamic zoning) at a scale of centuries? for exam- forest service, ). ple, timber production might be allowed periodically it remains to be seen if a policy of sharply in a non-timber production area to prevent native curtailed commodity production will be socially ac- oak-hickory forests from succeeding to beech-ma- ceptable in the long term. virtually every member of ple. on the other hand, timber production areas society uses wood-based products, and the demand could be allowed to lay fallow to provide non-com- for wood is projected to rise more than % by the modity values for some period of time. the current year (haynes et al., ). forest products are management paradigm appears to allow for spatial renewable, unlike many substitutes. reduced timber and temporal management dynamics within a man- production on federal lands increases demand for • agement area, but little thought has yet been given to private and foreign timber. industrial forest owners .dynamics in the designation of management areas also experience pressure to provide non-commodity over long time periods (over years). the interac- values from their forests. forest planners have the tion of the spatial and temporal domains of manage- unenviable task of attempting to balance the conflict- "ment activity has been inadequately explored, but ing demands by society for commodity and non- has significant consequences for the ecological con- commodity values from forested lands. the chal- .,. x ej. gustafson / forest ecology and management ( ) - lenge will be to develop new management paradigms gated, the disturbance occurs in a relatively small that allow commodity production while maintaining area over a short time period, and a relatively long _.,_,_.,_,,_.... •, non-commodity values, period free from harvest disturbance follows. thus, one of the potential ecological consequences of dynamic zoning produces a clustering of harvest timber harvest is a reduction in the amount of habitat disturbance in both space and time. dynamic zoning for forest interior species, many of which are thought is a potential tool to produce dynamic landscape to be declining in abundance (robbins et al., ; heterogeneity (mladenoff and pastor, ) by im- hill and hagen, ). most harvest methods create plementing harvesting cycles, and encouraging ex- openings that perforate blocks of contiguous forest plicit specification of disturbance regimes over large i and introduce edge habitat within the forest. many spatial and temporal scales. interior species are thought to be sensitive to the size in this study, i used a timber harvest allocation • of forested blocks (blake and karr, ; freemark model to compare four cutting strategies that differedt and collins, ), and internal edges may provide in the spatial and temporal dispersion of harvest improved habitat for generalist predators and brood allocations. my objective was to quantify the changes parasites (gates and gysel, ; brittingham and in forest interior habitat and forest edge produced by " ' temple, ; small and hunter, ; robinson et different harvest dispersion strategies, providing in- al., ' ). : sight into the utility of dynamic zoning strategies for the practice of dispersing cutting units has been forest management. recent studies have demon- " " implicated as a major contributor to the reduction in strated the value of clustering harvests spatially forest interior habitat and the increase in linear edge through time (li et al., ; gustafson and crow, (franklin and forman, ; gustafson and crow, ; wallin et at., ), but here i also examine ' ! ; wallin et al., ). progressive cutting across the effect of dynamically changing the locations of the landscape has been proposed as an alternative to timber harvest zones. the traditional approach of dispersing cutting units across the landscape (li et al., ). under this , strategy, timber harvesting would proceed somewhat . methods systematicall _, across the landscape. openings pro- duced by harvest would be clustered in both time . . study area and space, allowing more interior habitat to be sus- tained on the landscape as a whole. the practical the study was conducted on a rectangular study " application of this approach is complicated by dis- area ( ha) that included the entire hoosier continuous ownership of the landscape and the vari- national forest (hnf) purchase area, located in ability in the suitability of stands for harvest at any southern indiana, usa (fig. ). the hnf was used j given point in time. a variant of this approach might to provide realistic ownership and management area be to progressively designate timber harvest manage- (ma) patterns for assessing alternative cutting strate- •ment areas across the landscape over successive gies. the hnf is typical of national forests in the planning periods (dynamic zoning). this would also eastern usa in that the ownership pattern is highly . have the effect of clustering harvest openings within fragmented by privately-owned inholdings, and the , the larger landscape, but would allow more flexibil- hnf owns only about % of the land within the ity in the placement of individual harvest treatments purchase area. in published forest plans, ma , within the management area. flexibility at the water- boundaries have been drawn that specify the man- shed scale is essential to mitigate the effects of agement direction for the federally owned land within cutting on stream flow and sediment production each ma. i defined the land base on which timber _ (hombeck and swank, ) and to protect special harvest was to be simulated using the ma bound- res urce'features and habitats (naiman et al., ). aries of the hnf amended plan (usda forest spatial clustering of harvests by progressive cutting service, ). ownership boundaries were digitized also has, implications for disturbance (by harvest) from : scale paper maps produced by the us return intervals. when harvests are highly aggre- geological survey (usgs) for the forest service • .o • , • . .. ej. gustafson / forest ecology and management ( ) - united states tized. a forest cover map of the entire purchase area t was generated from usgs-land use data acquisi- tion (luda) data, and all layers were gridded to a common cell size of by meters. it was not feasible to idigitize stand age maps of the entire hnf and stand age data were not available for private i assumed that the distribution of pastland, so .... .. harvest activity (and therefore stand ages) is spatially-. •.' random on the hnf. i tested the assumption that stands reaching rotation age and past harvest alloca- i study arean tions are randomly distributed using nearest-neighbor • / analysis (davis, ) on ten subsets of hnf stand maps (mean (+ sd) size of subsets + ha). ' the observed mean nearest-neighbor distance be-o tween stands of similar age was compared with the o fig. . location of study area. distance expected if stands were randomly dis- - ' tributed, and a z-statistic was computed. the null (fig. (a)). ma bounclaries were manually trans- hypothesis that stands are randomly distributed could • ferred from _e forest plan maps (approximate scale, not be rejected at the % confidence level for eight " . ) to " scale usgs maps and digi- of the ten subsets (see gustafson and crow, ). __" _" "; .. .. r _' i ' " i _] " i j" " _.v_,. , "" '%r : '_'e" -_" ,._., : c "'*:" ' ' " ¢ ';..-, ' % -_ ]l_ai,t;" 'i ,, scale '_.,,_-_ b'" 'l i kilometers _ fig. . map of (a) distribution of land owned by the hnf within the study area and (b) timber land base on which harvests were simulated. the solid-line rectangles represent the subsetsused for the -year and -year hiatus alternatives, and the dashedlines represent the subsetsused for the -year hiatus alternative. upper case letters indicate the order in which timber harvest was allowed on the subsetsfor the - and -year hiatus alternatives, and the numbers indicate the order in which timber harvest was allowed on the subsets for the -year hiatus alternative. • • • ej. gustafson / forest ecology and management ( ) - . . timber harvest allocation model existed in timber production areas during the initial . decades of simulation to meet target harvest levels. harvest is a timber harvest allocation model a consequence of this procedure was that the distri- that was constructed to allow the input of specific bution of stands less than years of age (openings) rules to allocate forest stands for even-age harvest in the first two decades of simulation was not explic- (clearcuts and shelterwood) and group selection, us- itly modeled, so the initial forest condition appeared ing parameters commonly found in national forest less fragmented than is probably the case. plan standards and guidelines. the model produces landscape patterns that have spatial attributes result- . . experimental design ing from the initial landscape conditions and the propose d management activities. the model is sim- the land base harvested over a period of plistic in that it does not attempt to optimize timber decades was determined by the management area ' production or quality, nor does it predict the specific boundaries specified in the hnf amended locations of future harvest activity,, as it ignores plan (usda forest service, ). in the many considerations such as visual objectives and amended plan, timber harvest was allowed on road access. instead, the model stochastically mimics ha, but for the alternatives simulated in this study i , ° the ,allocation of stands, for harvest by forest plan- also allowed timber harvest on an additional " ners, using onlythe constraints of the standards and ha, to allow for higher timber outputs than projected . • guidelines and ma boundaries. modeling this pro- under the amended plan. timber production cess allowsexperimentation to link variation in man- was allowed on approx!mately % of the hnf land agement strategies with the resulting pattern of forest base, and only on land owned by the hnf (fig. , openings. (b)). harvest was constructed to be used in con- the experimental treatments consisted of altema- junction with a grid-cell geographic information tive designations of timber harvesting areas on the system (gis), with routines for direct input and hnf that varied as to where timber harvest was output oferdas v. . gis files, but supporting allowed during each decade and for how many ' files exported in text format from other raster gis decades it was allowed there. the total land base that systems. timber harvest allocations were made by was harvested (timber harvest land base) over a the model using a digital stand map, where grid-cell period of decades was identical among all altema- values reflect the age (in years) of the forest in that tives. for the static zoning alternative, harvest was ceil. harvest takes a gis stand age map as input, allowed throughout the timber harvest land base and produces a new stand age map incorporating during all decades. three dynamic zoning alterna- harvest allocations. harvest allows control of the tives were simulated. for the ' -year hiatus' alter- ," size distribution of harvests, the total area of forest to native, the timber land base was divided into three be harvested, and the rotation length (by specifying subsets; timber harvest was allowed on only two of the minimum age on the input stand map where these subsets at a time (beginning with subsets a and harvests may be allocated). harvest selects har- b, fig. (b)), and the third was temporarily set aside vest locations randomly within currently active tim- from timber harvest for years. the treatments ' her production mas, checking first to ensure that the were rotated every decades, so that each subset forest is old enough to meet rotation length require- was harvested for successive decades and then set , ments. this is .consistent with the random distribu- aside from timber harvest for decades. for the tion of past harvest activity, as discussed above. ' -year hiatus' alternative, the same three subsets since the initial forest ages were unknown, but were used, but timber harvest was allowed on n!y • assumed to be spatially random, i allowed the model one of these subsets at a time (beginningwith subset to choose harvest locations randomly from all cells a, fig. (b)), and the other two were temporarily set within timber production areas by assigning all forest aside from timber harvest. again, the treatments an initial age of years. this assumed that suffi- were rotated every decades, and each subset was cient area of forest old enough to be harvested harvested for successive decades and then set aside • ej. gustafson/ forest ecology and management ( ) - r .............. : --.-.-- f- for decades. finally, for the ' -year hiatus' pattern (static, -year hiatus, -year hiatus, and alternative, the timber harvest land base was divided -year hiatus), and two levels of harvest intensity into'five subsets, and each subset was harvested for ( ha per decade and ha per decade). three decades (beginning with subset , fig. (b)) and then replicates of each factorial combination were pro- was set aside for decades. total area harvested, duced. : size of harvest openings, and rotation interval (minimum age for cells to be harvested) were held constant across all treatments and decades, so that . . analysis timber production was the same for all four scenar- ios. at each decade, i used a gis to determine the timber harvest parameters were chosen to fall amount of forest interior habitat (forest over m within the parameter space of the plan and from an opening or edge). a simple fortran amended plan alternatives simulated elsewhere routine was written to calculate linear forest edge. " (gustafson and crow, ) and are detailed in clearcut stands in the hnf generally achieve canopy table . only harvest methods that produce forest closure in - years (t. thake, personal commu- o openings (clearcut, shelterwood and seedtree) were nication, ), so cells harvested were assumed to simulated,, and harvest placement was not con- create openings in the forest for years and were strained by adjacency prohibitions. the intensity of then assumed to return to a closed canopy condition. • harvest (total area harvested) is an important deter- a different definition of canopy closure would change minant of forest interior and edge (gustafson and the absolute amount of interior and edge, but the crow, ), so i conducted simulations at two relative differences among alternatives would be levels of harvest intensity, one having twice as much similar to those reported here. the total amount of area harvested each decade as the other. this al- edge and interior was plotted over simulated time for lowed me to assess the relative contribution of both each of the alternative cutting patterns. for compari- dispersio n of harvests and intensity of harvest to the son, the results of simulating the original forest amount of forest interior and edge. higher levels of plan and the amended plan (gustafson and sustained harvest were not possible without increas- crow, ) were also plotted. ing the timber land b_e. thus, a complete factorial to evaluate the relative effects of harvest disper- design was implemented, with four levels of cutting sion and harvest intensity on forest interior and edge, table harvest .intensities used in the simulation of timber harvest alternatives on the hoosier national forest. the high and low intensity parameters were used for the simulation of the static and dynamic zoning cutting alternatives. analysis included three replicates of simulations conducted for decades model parameter low intensity high intensity plan plan • " mean clearcut opening size (ha) . . . - . . mean group pening size (ha) na na . . i • m_imum opening size (ha) . . . . total harvested per decade a (ha) . . . . harvest iate per decade t, (%) . . . . rotation length (years) - " timber land base (ha) c a represents harvest activity across the entire forest. total hnf ownership is approximately ha. b represents the percentage of forest within the timber harvest land base that is harvested each decade. c total area where harvest is allowed during at least part of the -decade simulations. • ej. gustafson / forest ecology and management ( ) - an anova. was used to test for treatment effects time periods were included in the analysis to account reflecting harvest dispersion (dispersion), harvest for the potential correlation of measures of forest " intensity (intensity), and time (decade). the interior and edge in successive decades. [......"'"_ ........"- ' static -yr a hiatus b | •. , ,_i. __. _ _:_ _' _ lz \ • i_,'_,' . ,_' ,,.-. ,,,, .. . • _ non-forest/harvested scale forestedgehabitat kilometers forestinterior o o .. -yr -yr.. •. hiatus hiatus c. d f ,'._-. :_:,. fig. . forest interior in the study area at the end of decades of simulated harvest at the 'high intensity' harvest rate ( ha per decade) under the four zoning alternatives. the solid lines represent the approximate location of the hnf purchase boundary, and simulated harvests occurred only on hnf land within those boundaries. • o ej. gustafson / forest ecology and management ( ) - . results _ static zoning - -v- -year hiatus - - -a - -year hiatus - -o - plan j ; _ •::_-._._+_,-__,+_+._..,..... r -year hiatus + amended plan the total amount of forest interior varied markedly o among the simulated altertlatives (fig. ), with the _ ^..} highest amount produced ,by the pattern that most o •• _ • y • tightly clustered harvests in time (i.e. longest hiatus '_ - _period) and space ( -year hiatus, fig. (d)). under o i_ the static alternative, none of the timber land base • z was set aside at any time; large amounts of forest - • t_ edge habitat can be seen scattered throughout the _ oo hnf purchase area, and few blocks of forest inte- o , , ," , , , , , , , , ,,,riorremain (fig. (a)).+under the dynamic zoning is°° "_*"**'*"*'_°"'*_- i alternatives, increased amounts of forest interior can lsoo be seen in the.areas that had just completed their fallow period; for+example, examine subset b (fig. decade . (b), fig. (b)) and subsets a and b (fig. (b), fig. fig. . the amount of forestinteriorproducedby timberharvest (c)), representing the -ye_ and -year hiatus alternatives over simulated time on the hoosier national forest alternatives, respectively. forest interior reaches its (hnf).the 'static' alternative is the least clustered in space and • time, while the ' -year hiatus' alternative is the most clustered • highest levels on _e landscape as a whole under the in space and time. the results plotted are for the 'high intensity' -year hiatus alternative (fig. (b), fig. (d)). harvest rate ( ha per decade)- the plan and timber production is evident in subset under this amended plan are simulations of published plans for the hnf +alternative, where the density of harvest openings is (gustafson and crow, ), and are shown for comparison. quite high, due to the high level of clustering. the replications of the simulations produced little variability in forest interior and edge. the variability openings existed on two zones, perforating contigu- was too low to show clearly with error bars on line ous forest habitat across a broader portion of the graphs, so error bars are not shown. the standard landscape. one might expect that forest edge would deviation from the mean of interior area produced by not show such a pattern, since edge is introduced three replicates never exceeded . % in any combi- around an opening, regardless of the spatial disper- nation of treatments, and the standard deviation from sion of the openings. however, when harvest inten- the mean linear edge never exceeded . %. sity was high, openings (cells less than years old) the dynamic zoning strategies resulted in more within the timber production zones begin to coalesce, forest interior across the landscape than the static reducing edge. the oscillation in the amount of edge / zoning alternative (fig. ), with the highest amount seen in fig. reflects this periodic coalescence of produced by the pattern that most tightly clustered openings near the end of production in a zone, and harvests in time and space ( -year hiatus, fig. generation of relatively higher amounts of edge when (d), fig. ). the dynamic zoning strategies also new cutting zones were opened. this oscillation is resulted inless forest edge across the landscape than not evident at low harvest intensity (not shown). .. the static zoning alternative, with the least amount levels of fragmentation under a dynamic zoning againproduced by the pattern that most tightly clus- alternative with a high intensity of harvest were tered harvests in time and space ( -year hiatus, comparable with those produced by the static altema- " fig. ). tive with a low level of harvest (fig. ). note in fig. • the periodic rise and fall in the amount of forest that even the -year hiatus alternative (high inten- interior and forest edge evident in the dynamic zon- sity is plotted) produced approximately the same ing alternatives' was caused by the initiation of cut- amount of forest interior as the amended plan, ting on a new cutting zone. openings were produced even though the total area harvested under the -year in the new zone before all the openings closed on the hiatus alternative was twice that of the amended previous zone, so that for a -decade period harvest plan (table ). with a cutting intensity similar to [] --- / l ' " + • , " . e.i. gustafson / forest ecology and management ( ) - ,.,' static zoning -.-v - -year hiatus . - -m- -year hiatus - - plan _ -year hiatus, high intensity •_ _ --d - static, low intensity -year hiatus ---b-- amended plan o_ -e _ _ e..e.e.t.e.o-_._ _ - - _'. ® " _- _, i_ - - tu z_ i-- ° |_ - -w w m tu u.. n,, . b-h...._ _ ..=-.h....m- a it. n," ' ' i ' ' i , , i , , i , ' z ..i decade o fig. . amount of forest interior produced over simulated time by . the ' -year hiatus' alternative at high-intensity harvest ( ha decade per decade) and the 'static' alternative at low-intensity harvest fig. . amount of forest edge produced by timber harvest altema- ( ha per decade). the 'static' alternative is the least clustered . fives over simulated time on the hoosier national forest (hnf). in space and time, while the ' -year hiatus' alternative is the the 'static' altemat.ive is the least clustered in space and time, most clustered in space and time. while the ' -year hiatus' alternative is the most clustered in space and time. the results plotted are for the 'high intensity' forest. the plan, with an intensity of harvest harvest rate ( .ha per decade). the plan and approximately half thafof the high-intensity dynamic amended plan are simulations of published plans for the hnf zoning altemative_, produced higher amounts of edge (gustafson and crow, ), and are shown for comparison, due to the use of group selection. • differences in the spatial dispersion of harvests that of the plan (low intensity), all the altema- (zoning) appear to have a greater effect on the tires, including the static one, produced more forest amount of forest interior than do differences in har- . interior than the plan (plot not shown). this vest intensity. all the main effects are highly signifi- wasdue to the use of smaller openings in the cant in the anova models; however, examination plan, including extensive use of group selection, of the sums of squares shows that the spatial disper- which resulted in more openings that perforated the sion of harvests (dispersion) explains . % of l table analysis of variance comparing the effects of harvest intensity (intensity), the spatial dispersion of harvest activity (dispersion), and the time period simulated (decade) on the area of forest interior and linear forest edge maintained on the landscape. analysis included three replicates of simulations conducted for decades source d.f. forest interior (kin ) forest edge (kin) ss f prob > f r ss f prob > f r intensity . . . . dispersion = . . . . , 's v _ . . . . • " v b - . . . . , v c . . . . • ' s& v & a . . " . . decade . . . . error total . . a orthogonal contrast of the static (s) zoning alternative with the -year ( ) hiatus alternative. b orthog nal contrast of the -year ( ) hiatus alternative with the -year ( ) hiatus alternative. c orthogonal contrast of the -year ( ) hiatus alternative with -year ( ) hiatus alternative. a orth gona! contrast of the static (s) and -year ( ) hiatus alternatives with the -year ( ) and -year ( ) hiatus alternatives. • = _,_ii_i_!__¸_¸ e.i. gustafson/ forestecologyandmanagement ( ) - the total variance f forest interior, while harvest produced, but from the temporal and spatial configu- intensity (intensity) explains . % of the vari- ration of its extraction. specifically, as harvests be ..... _,,_ ....... , ance and decade explains . % (table ). or- came more aggregated in time (longer hiatus inter- thogonal contrasts partitioning the variation caused val) and space, the level of fragmentation decreased, by dispersion (table ) show that the greatest the average age of forests in timber management variance is explained by differences between the zones inereased, and the disturbance interval neces- -year and the -year hiatus alternatives ( . %), sary to achieve a given level of harvest was length- and that the variance explained by differences be- ened. relative to static zoning, dynamic zoning in- tween the two ieast aggregated alternatives (static creases opportunities to reduce the amount of edge and -year hiatus) and the two most aggregated and increase both the amount of interior habitat and ' alternatives ( -year and -year hiatus) is . %. timber production by clustering harvest activity and intensity is more important in explaining the lengthening disturbance intervals. although i simu- length of forest edge, explaining . % of the total lated specific hiatus intervals, the important point is ' variance, while dispersion explains only . % not the length of these intervals, but the temporal and of the variance a_nddecade explains . % (table spatial dynamics of the clustering that coincidentally ). with a given harvest size, each opening produces produced these intervals. these results were obtained a fixed amount of edge, and the number of openings by simulating dynamic zoning on a national forest, produced is proportional to harvest intensity. the but the principle of clustering harvests in both time - • • spatial dispersion of openings has some impact on and space can be applied to the management of any edge, in that more aggregated harvests tend to pro- large land base. industrial forests are managed to duce a coalescence of openings that reduces the maximize mean annual increment of timber volume • _ relative amount of edge produced. orthogonal con- and to favor the regeneration of certain tree species. trasts partitioning, the variation in edge caused by clustering disturbance by dynamic zoning with a dispersion show trends.similar to those of forest rotation interval < years would produce less interior, but at lower levels of variance explained fragmentation than dispersing disturbance with a (table ). similar interval. dynamic zoning can also serve to clustered harvests with longer hiatus periods re- cluster operational activities such as road improve- suited in an increase in the average age of stands in ment, access control, and site preparation, lowering timber production zones on subsequent re-entries, costs of production. the average age of forest cells at the end of the simulations reported here did not include the decades under the static alternative was . years, effects of any disturbance other than timber harvest- the average age of cells in a zone after its hiatus ing. such effects may be significant on some land- period was . years under the -year hiatus scapes, but are probably minimal on the hnf. on ' alternative, . years under the -year hiatus the hnf, prescribed fire has been used to maintain alternative, and . years under the -year hia- barrens and oak-hickory communities, but wildfire tus alternative. the dynamic zoning alternatives had is rare and localized in this moist central hardwood the effect :of aggregating older forest stands by clus- region. windthrow is more common, but its effects tefing disturbance. _ are also generally local. natural disturbance in this • region would produce some fine-grained, local patch-. . iness, but its overall impact on landscape pattern . discussion _ would likely be minimal on a landscape of this size, even over a period of decades. •these results demonstrate the potential benefits of consideration of-the temporal and spatial scale of enlarging the spatial and temporal scale of forest disturbance is critical for the understanding of eco- ' management planning and of incorporating long-term logical processes (urban et al., ; wiens, ; temporal dynamics (dynamic zoning) into manage- reice, ). the designation of mas on managed ment plans. i found differences in forest fragmenta- forests essentially specifies the disturbance regime tion that resulted not from the amount of timber for each area of the forest. timber harvest imposes e.j. gustafson / forest ecology and management ( ) - periodic disturbance that changes the community at a shorter rotation, helping to enhance biodiversity and spatial "scale of several hectares, and the intensity of to maintain soil productivity (swanson and franklin, harvest is a major determinant of the resulting corn- ). a greater diversity of seral stages would exist munity structure and composition. in other mas, across the landscape, although the interspersion of disturbance may be suppressed, with very little man- the types would be less. in addition, stands would be agement that might directly change community corn- older when they are re-entered, introducing position or landscape pattern. thus, static zoning economies of scale in the harvesting and processing i causes specific disturbance regimes to exist in perpe- of larger trees. however, longer rotations may pro- tuity in specific locations. it is not clear what the duce changes in community composition that in some impacts of static zoning on biotic diversity might be cases may be undesirable. a dynamic zoning strat- over long time periods. it has long been accepted egy allows for flexibility in rotation length (or distur- that disturbance produces the spatial heterogeneity bance interval), while still retaining the benefits of ! that is necessary to maintain diversity. however, the clustereddisturbance in time and space. long-term role of disturbance is sometimes mini- clearly, many other factors besides forest frag- mized in the management of 'natural ecosystems' mentation impact forest management plans. for ex- (attiwiu, - ). non-equilibrium theories of corn- ample, increasing the area of old-growth forest would . munity structure suggest that the diversity of species be difficult on a dynamically zoned land base and and the coexistence of similar species that is seen in would probably require integration of dynamic zon- " most communities are due to some level of distur- ing with old-growth islands (sensu the 'long-rotation bance and the resulting opportunityfor recruitment island' concept of harris( )). some flexibility in of new species to the community (connell, ; rotation intervals may be required to meet vegetation huston, ; lewin, ; reice, ). how management goals on other parts of the landscape. ecosystems will respond to novel disturbance regimes public acceptance of periodic changes in the location is not often understood (swanson et al., ). for of natural appearing recreation areas is difficult to example, in the central hardwood region, there ap- predict, and may be problematic for the implementa- pears to be a trend toward the conversion of native tion of a dynamic zoning strategy. these issues ' oak-hickory communities to beech-maple, thought certainly need to be investigated. nevertheless, the t be the result of fire suppression (lorimer, ). results of these simulations suggest that it may be it is far from clear how the rest of the flora and technically possible to extract timber from a large fauna might respond to the development of a forest land base while maintainingmost of that land base in " type to which they are not adapted and that may not a relatively undisturbed state for long periods of have existed in many areas since pleistocene glacia- time. tion. the dynamic zoning alternativessimulatedhere most eastern forests still bear the legacy of "z" do not adequately mimic 'natural' disturbance widespread disturbance and abuse, and most are regimes, and were in fact deterministic with a long relatively young. it is prudent to protect parts of the temporal period. furthermore, i simulated a very forest from timber harvest to develop a diversity of limited set of silvicultural and management options, forest conditions across the landscape. however, this . however, the dynamic management of disturbance may be wise only in the short term, and explicit . overlong time periods allows managers greater flex- thought must be given to the nature of disturbance •ibility, and coupled with the clustering of disturbance regimes that will be imposed across large forested . in time and space can be used to sustain larger areas over the long term. blocks of mature forest than can a static alternative. as timber production zones are more tightly clus- ' teredin time and space, the effective rotation interval . conclusion becomes 'longer, providing large blocks of mature forest habitat on zones nearing the end of rotation. it is perhaps inevitable that conflicting demands . these older forests would be expected to have greater on our forests will increase. the value of forested structuralcomplexity than forests managed on a ecosystems for recreation and as repositories of bio- , e.i. gustafson / forest ecology and management ( ) - logical diversity will increasingly be recognized, references while the deriaand for wood products will also in- crease. pressure to provide both commodity and attiwill, p.m., . the disturbance of forest ecosystems: the non-commodity valuesfrom our forests will require ecological basis for conservative management. for. ecol. new and creative ways to manage these valuable manage., : - . resources. behan, r.w., . multiresource forest management: a paradig- a criticism of timber harvest is that it reduces the maticchallengeto professionalforestry.j. for., ( ): - . bblake, j.g. and karr, j.r., . breeding birds of isolated habitat values of certain species of concern and woodlots:area and habitatrelationships.ecology, : - reduces the aesthetic enjoyment of the forest. a . criticism of setting aside lands from timber harvest is brittingham, m.c. and temple, s.a., . have cowbirds caused ' that wood fiber is locked up and wasted, putting forestsongbirds to decline? bioscience, : - . connell, j.h., . diversity in tropical rainforests and reefs. more pressure on other woodlands and foreign coun- science, : - . tries to produce the fiber to meet the demands of crow, t.r. and gustafson, e.j., . ecosystem management: society. a dynamic zoning paradigm begins to sat- managing natural resources in time and space. in: k.a. kohm isfy both these criticisms through better and more and j.f. franklin (editors),creatinga forestry for the twenty- judicious integration of multiple uses. when harvests first century. island press, washington, dc, pp. - . are clustered in both space and time, more forest davis, j.c., .statisticsand data analysisin geology.john * interior is preserved, and more areas are distant from wiley, new york, pp.franklin, j.f. and forman, r.t:t., . creating landscape any signs of recent harvest activity. the specific patternsby forestcutting:ecologicalconsequencesand princi- location of these areas would shift across the land- pies. landscape ecol., " - . scape on a time scale of several decades, but the freemark, k. and collins, b., j . landscape ecology of birds amount of land in these conditions would remain breeding in temperate forest fragments. in: j.m. hagen and d.w. johnston (editors), ecology and conservation of constant, if timber harvest activity is moved progres- neotropical landbirds. manomet bird observatory, woods sively across the landscape, less of the forest is hole,ma, pp. - . permanently set aside, and harvested stands will be gates, j.e. and gysel, l.w., . avian nest dispersionand older on average, introducing economies of scale in fledgling success in field and forest ecotones. ecology, : their harvest. even within timber harvest areas, har- - . gustafson, e.j. and crow, t.r., . modeling the effects of vestingwould be spread out over - years, so forest harvesting on landscape structure and the spatial distri- that when stands are cut at the end of the period, the butionof cowbirdbrood parasitism.landscapeecol., : - stands cut first will have regenerated to a closed . canopy condition. although a significant portion of a gustafson, e.j. and crow, t.r., . simulating the effects of managed forest might still be set aside from timber alternativeforest managementstrategieson landscapestruc- ture.j. environ. manage., : - . _,- harvest; a dynamic zoning paradigm could support harris, l.d., . the fragmented forest. university of chicago higher levels of timber extraction with less forest press,chicago, il, pp. fragmentatio n than the static zoning alternative. dy- haynes, r.w., adams, d.m. and mills, j.r., . the namic zoning also encourages explicit specification rpa timber assessment update. gen. tech. rep. rm- , of the disturbance regimes that will be imposed usdaforestservice. across the land base over long time scales. hill, n.p. and hagen, j.m., iii, . population trends of some. northeastern north american landbirds: a half-century of data.,. • wilson bull., : - . hombeck, j.w. and swank, w.t., . watershed ecosystem • analysis as a basis for multiple-use management of eastern acknowledgements forests. ecol. appl., : - . huston, m., . a general hypothesis of species diversity. am. nat., : - . i thank t. crow, j. johnson, s. tang, c. morgan, lewin,r., .supply-sideecology.science, : - . ' li, h., franklin, j.f., swanson, fj. and spies, t.a., . s. shirley, d. mladenoff and two anonymous re- developing alternative forest cutting patterns: a simulation viewers for critical reviews of earlier drafts of the approach.landscapeecol., : - . manuscript. i thank l. burde for editorial assistance. lorimer, c.g., . the role of fire in the perpetuation of oak i[_l e.i. gustafson / forest ecology and management ( ) - ' ]- _i'l;iiii¢....j forests. in: j.e.. johnson (editor), proc. of challenges in oak swanson, f.j. and franklin, j.f., . new forestry principles management and utilization. wex cooperative extension from ecosystem analysis of pacific northwest forests. ecol. - service, university of wisconsin-extension, madison, wi, pp. appl., : - . - . swanson, f.j., jones, j.a., wallin, d.o. and cissel, j.h., . mccarthy, m.a. and burgman, m.a., . coping with uncer- natural variabilitymimplications for ecosystem management. tainty in forest wildlife planning. for. ecol. manage., : lli: m.e. jensen and p.s. bourgeron (editors), vol. ii: ecosys- - . tern management: principles and applications. pnw-gtr- mladenoff, d.j. and pastor, j., . sustainable forest ecosys- , usda forest service, pp. - . / terns in the northern hardwood and conifer forest region: urban, d.l., o'neill, r.v. and shugart, h.h., . landscape / conceptsand management. in: g.h. aplet, n. johnson, j.t. ecology. bioscience, : - . olson and v.a. sample (editors), defining sustainable usda forest service, . land and resource management ' forestry. island press, washington dc, pp. - . plan, hoosier national forest. usda forest service, eastern , naiman, rj., decamps, h. and pollock, m., . the role of region, hoosier national forest, bedford, in, usa. • riparian corridors in maintaining regional biodiversity. ecol. usda forest service, . plan amendment, land and re- appl., : - . source management plan, hoosier national forest. usda _ reice, s.r., . nonequilibrium determinants of biological forest service, eastern region, hoosier national forest, bed-community structure. am. sci., : - . ford, in, usa. robbins, c.s., sauer, j.r., greenberg, r.s. and droege, s., . wallin, d.o., swanson, f.j. and marks, b., . landscape population declines in north american birds that migrate to pattern response to changes in pattern generation rules: land-use . the neotropics. proc. natl. 'acad. sci., : - . legacies in forestry. ecol. appl., : - . " robinson, s.k., thompson, f.r., donovan, t.m., whitehead, wallinger, s., . a commitment to the future: af&pa's . . d.r..and faaborg, j., . regional forest fragmentation and sustainable forestry initiative. j. for., ( ): - . the nesting success of migratory birds. science, : - wiens, j.a., . spatial scaling in ecology. funct. ecol., : . • - . small, m.f. and hunter, m.l., . forest fragmentation and - ' .. avian nest prex!ation in forested landscapes. oecologia, : - . , r forest ecology and management submission of manuscripts. manuscriptsshouldbe submittedin triplicate.authors fromthe americas,australia,new zealand and the pacificare requested t°university,sendtheircollegemanuscriptsstation,t° f°resttx - ,ec°l°gyandusa;management'allother authorsdr" richardarerequestedf"fisher, todepartmentsendtheirofmanuscriptsforestscience,tothetexaseditoriala&m b office, forest ecology and management, p.o. box , ad amsterdam, the netherlands. authors in japan please note: upon request, elsevier science japan will provide authors with a list of people who can check and improve the englishof their paper (before submission). please contact our tokyo office: elsevier science japan, -- , - - higashi-azabu,minato-ku,tokyo , japan; tel. ( )- - ; fax ( ) - . electronic manuscripts: electronicmanuscripts have the advantage that there is no need for the rekeying of text, thereby avoidingthe possibilityof introducingerrorsand resultingin reliableand fast deliveryof proofs. for the initial submissionof manuscriptsfor consideration,hardcopiesare sufficient.for the processingof accepted papers, electronicversionsare preferred.after final acceptance, yourdisk plustwo, finaland exactly matchingprintedversionsshould . be submittedtogether. double density (dd) or highdensity (hd) diskettes( . or . inch) are acceptable.it is importantthat the file saved is in the native format of the wordprocessorprogram used. label the disk with the name of the computerand wordprocessingpackage used, your name, and the name of the file on the disk. further informationmay be obtainedfromthe •publisher. all questions,arising after acceptance of the manuscript,especiallythose relating to proofs, should be directed to elsevier editorialservices, mayfieldhouse, banbury road, oxford ox dh, uk, tel. ( . ) , fax ( ) . advertising information: advertising orders and enquiries may be sent to: international= elsevier science, advertising department, the boulevard, langford lane, kidlington, oxford, ox gb, uk; tel. (+ )( ) ; fax (+ )( ) . usa and canada: weston media associates, dan lipner, p.o. box , greens farms, ct - , usa; tel. (+ ) ( ) ; fax (+ ) ( ) . japan: elsevier science japan, marketing services, - - higashi-azabu, minato-ku, , tokyo , japan; tel. (+ )- - - ; fax (+ )- - - . us mailing info: forest ecology and management ( - ) is published monthly by elsevier science b.v. (molenwerf , postbus , ae amsterdam). annual subscription price in the usa is us$ . (valid in north, central and south america), includingairspeed delivery. second classpostagerate ispaid at jamaica, ny . _ usa postmasters: send address changes to forest ecology and management publications expediting, inc., meacham avenue, elmont, ny . airfreight and mailing in the usa by publications expediting inc., meacham avenue, elmont, ny . forest ecology and management has no page charges for a full and complete guide for authors, please refer to forest ecology and management, vol. , nos. - , pp. - y. copyright © , elsevier science b.v. all rights reserved. - / /$ . (_ the paper usedin this publicationmeets the requirementsof ansi/niso z . - (permanence of paper). printedin the netherlands "landscape and urban planning = an international journal of landscape ecology,• • landscape planning and landscape design , o .' editor-in-chief: j.e. rodiek, college of architecture, texas a & m university, college station, "ix - , usa . . aims and scope the premise that research r.¢. szaro, washington,oc,usa, a journal concerned with linked to practice will j.w.thomas,lagrande,or,usa..... " conceptual, scientific and ultimately improve the human p.j.trowbridge,ithaca,ny,usa,t.h.d.turner,london,uk, • design approaches to land made landscape. m.j.vroom,wageningen, -use. by emphasizing thenetherlands, _" ecological understanding and w.v.wendler,collegestation,tx, a multi-disciplinary approach editorial advisory board" usa, to analysis and planning and i.d.bishop,parkville,vic.,australia, b.-e.yang,seoul,korea,• e.g.bolen,wilmington,nc,usa, e.h.zube,tucson,az,usa design, it attempts to draw i.d.bruns,schomdorf,germany, attention to the interrelated j.b.byrom,edinburgh,uk, • nature of problems posed by t.c.daniel,tucson,az,usa, abstracted/indexed in •nature and human use of r.m.degraaf,amherst,ma,usa, applied ecologyabstracts, land. in addition, papers j.g.fabos,amherst,ma,usa, biologicalabstracts, " dealing with ecological s.gonzalezalonso,madrid,spain, current contents b & s, processes and interactions m.hough,etobicoke,on,canada, environmental periodicals within urban areas and p.jacobs,montreal,pq,canada, bibliography, geobase, ' d.s.jones,melbourne,vic.,australia, geographicalabstracts, between these areas and the h.lavery,milton,qid.,australia, landsearch. surrounding natural systems w.m.marsh,flint,mi,usa, which supportthem, will be d.l.mitchell,dallas,tx,usa, considered. papers in which d.g.morrison,athens,ga,usa, specific problems are j.l.nassauer,st.paulmn,usa, subscriptiondata examined are welcome. m.helischer,guelph,on,canada, d.d.paterson,vancouver,bc,canada, volumes - (in issues) topics might include but are notlimited to landscape a.ramos,madrid,spain, subscriptionprice: p.shepard,c/aremont,ca,usa, dfl. . (us$ . ) ecology; landscape planning o.r.skage,a/narp;sweden, incl.postage and landscape design. r.c.smardon,syracuse,ny,usa, • " landscape ecology examines g.sorte,alnarp,sweden, issn - how heterogeneous f. stearns,rhinelander,wi,usa; p. ' combinations of ecosystems , are structu red, how they elseviersciencepublishers, function and how they - p.o.box , aeamsterdam, • change. landscape planning _"_". thenetherlands examines the various ways ___ fax:( ) - humans structure their land __l customersintheusaandcanada:use changes landscape elseviersciencepublishers "• p.o.box ,madisonsquarestation, design involves the physical strategies and forms by which g lsevi g r newyork,ny - ,usafax:( ) - land use.change is actually _,_out_,guilderpricequotedappliesworldwide,exceptin_e _/americas (nor_, cen_al and south america}. the us dollar directed landscape and science publishers pricequotedappliesin_eamericasonly.nonvatregistered • customers in the european community should add the urban planning is based on appropriate vat rate applicable in their ¢ount_ tothewice. . the middle (review) the middle (review) anthony harkins middle west review, volume , number , spring , pp. - (review) published by university of nebraska press doi: for additional information about this article [ this content has been declared free to read by the pubisher during the covid- pandemic. ] https://doi.org/ . /mwr. . https://muse.jhu.edu/article/ https://doi.org/ . /mwr. . https://muse.jhu.edu/article/ media reviews feminist scholar uma narayan observes how western stereotypes of those who live in the so- called third world enable injustice and oppression to be “explained away” as merely symptomatic of a “national culture” that is somehow “timeless” and “outside history.” worrisome in some con- temporary discussions about lgbtq rights across the globe is a similar ex- planation of antigay bigotry, not as a transnationally circulated discourse partly imported from the u.s., but as a cultural norm. without the juxta- position of images in god loves uganda, ugandan religious practices risk being explained away as “cultural,” deemed symptomatic of a “dark” and “backward” third world. god loves uganda indirectly critiques this line of reasoning because it shows the transnational nature of oppressive discourse and its infl uence by capitalist logic and conservative values. williams helps viewers understand that these images of religious and political hatred are a result of cultures in contact. the midwest is no more isolated than uganda as the two col- laborate on policy and profi t at the expense of the poor and dispossessed. michael t. macdonald university of michigan–dearborn dearborn, michigan note . uma narayan, dislocating cultures: identities, traditions, and third world feminism (new york: routledge, ), . the middle, abc, – . during a panel discussion at the paley center for media in new york with the stars and creators of the middle— the hit abc sitcom about a decid- edly lower middle class family of fi ve living in the middle of the country— lead actress patricia heaton tried to explain the secret to the show’s suc- cess across such a broad socioeconomic range of viewers: “we are very specifi cally midwestern, and very specifi cally economically lower middle class and struggling . . . and i think specifi city makes everything much more universal.” certainly the show’s regular opening sequence of an empty road sur- rounded by corn fi elds with crows cawing in the background seems to es- tablish an explicit and specifi c midwestern location and identity, as does middle west review • vol. no. heaton’s accompanying voiceover with matching images: “orson, indi- ana: heart of the heartland, proud home of little betty snack cakes, the demolition derby for the homeless, and the world’s largest polyurethane cow.” this montage shrewdly blends a detached commentary on the town’s— and by extension, the region’s— déclassé culture with genuine hometown pride. what is most striking about this supposedly “specifi cally midwestern” show, however, is how rarely it represents, in anything more than a passing way, an actual midwestern locale rooted in an agricultural economy. instead it presents an imagined but fi rmly entrenched portrait of the midwest as the “all american” home of supposedly unique “heartland” values of close families, hard work, and “authenticity.” the middle comically presents the daily trials and tribulations of the heck family, which comprises working mother frankie heck (heaton), her husband mike (neil flynn) who is the foreman at a quarry, and their school aged children, axl (charlie mcdermott), sue (eden sher), and brick (at- ticus shaffer). at the heart of the show’s conceptualization and its cultural meaning is its construction of a mutually reinforcing social, cultural, and psychological “middleness” that confl ates american geographic central- ity, socioeconomic and cultural middle class identity (or more accurately, that of the petit bourgeoisie), ideological moderation, and middle age sen- sibility. of these elements, the geographic component is the one presented with the least nuance and specifi city. the pilot episode’s opening scene promised an instructive counter- response to the idea of the region as literal and fi gurative “fl yover country” for coastal elites. “some people call this the middle of nowhere,” frankie heck’s voice intones as the show began, “you know, one of those places that you fl y over from somewhere to somewhere else but you wouldn’t live here.” the scene then rapidly transitions to a fl ight attendant on a jet liner far overhead encouraging people to look down on “the great state of indiana,” immediately followed by a quick cut to the airline passengers studiously doing anything but look out the window. heck then chimes in again, as the screen fl ashes a cartoonish map of the plane’s route between boston and lax: “well, look down next time and you’ll see us down here in the middle.” yet over the rest of the episode and continuing through the following six seasons, the show has offered few insights into the actual midwest and its people. certainly there are occasional references to midwestern and espe- cially indiana place markers, such as a heck family trip to chicago, a hal- loween episode that features a corn maze, or local sports teams. (indiana media reviews hoosiers basketball tickets briefl y feature in one storyline, the indianapolis colts as mike’s favorite team in others.) in general, though, there is little to indicate that the program takes place in a defi nitive midwestern locale, or, more to the point, that such a socioeconomic and physical landscape would be any different from a suburban landscape anywhere else in the country. there have been a few exceptional episodes with a more explicit midwestern focus, including one that focuses on the town’s one hundred year anniversary celebration— the “orsontennial” as they call it— in which the hecks drive a fl oat shaped like an enormous cow, as well as another episode about a college tour by sue and mike of actual indiana universities. nonetheless, for the most part, the show could take place in almost any smallish u.s. town outside the immediate area of major metroplexes. this is perhaps not that surprising given that the program is fi lmed on a back lot in burbank. even the opening sequence of fi elds was fi lmed around oxnard, california, and had to have corn stalks digitally added. in the paley interview cited above, cocreator eileen heisler acknowledges that the show was “never intended . . . to be a show where the middle of the country would say, yay! finally i’m recognized.” instead, she explains “you write it for everyone.” yet she and her cocreator deann heline framed this rather differently in an interview with an indianapolis local news reporter who visited the hollywood set in . “it’s a love letter to the midwest,” heline told the reporter, who gushed that the show is “packed with hoo- sier references” and “midwestern expressions,” although the sole example she provided was “pop” for soft drinks. in the interview, heline and heisler claimed that the show celebrates uniquely midwestern qualities; the pair used words such as “genuine” and “hardworking” to describe these re- gional characteristics. after the conclusion of the interview segment, the local anchor back in indianapolis proudly asked the reporter, “they are still holding on to their midwestern roots aren’t they?” the beaming re- porter concurred. heisler and heline indeed were raised in midwestern cities (heisler in chicago, heline outside of cincinnati), and the two met at indiana univer- sity and have worked as cowriters and coproducers ever since graduating college. (they both transferred to new york university). their responses to the indianapolis reporter and her reaction, however, are key to understand- ing the “midwest” the show depicts. the middle does not focus on a actual physical place shaped by real economic and social circumstances, but rath- er a symbolic locale defi ned by a set of romanticized conceptual qualities of middle west review • vol. no. “realness” and “americanness”— largely defi ned as white, christian, and middle class— that are supposedly exceptional to the midwest. in so doing, the show reinforces “heartland” tropes that have long played a central role in the televised national imaginary of postwar america. to be fair, the middle is hardly alone in fi lming its show on a studio lot rather than on location, or in presenting a mythologized america. further- more, to its great credit, the show is a far more honest representation of the daily lives of most americans than are most other televised situation com- edies. the creators eschew the typical multicamera fi lm format and built- in laugh track, and the episodes routinely offer a generally honest (if comi- cally presented) depiction of struggling lower middle class life. the heck home is marked by a constant state of mess but not fi lth, and their lowbrow home décor is represented as brightly colored but functional kitsch. the heck children speak and act like kids actually do rather than simply spouting punchlines, and the show depicts with genuine understanding child- parent relationships and the hardships that face working moms. unfortunately, the middle breaks little new ground in its representa- tion of the midwest itself. the program reinforces the widespread notion that the region’s people are interchangeable and homogenous. thus, al- though the middle ostensibly sets out to challenge a vision of the midwest as a “fl yover” space that is rightfully ignored and forgotten, it has ended up reaffi rming this very “fl yover” mentality. it is therefore yet another popular cultural form that frames the midwest as both hopelessly banal and undif- ferentiated and, at the same time, the home of the “real america.” anthony harkins western kentucky university bowling green, kentucky notes . the middle, abc, sept. , . . ryan reynolds, “roll credits: new abc sitcom won’t use ‘jasper, ind.’ after all,” evansville courier & press, may , . . “abc’s ‘the middle’ loaded with hoosier connections,” wrtv (indianapo- lis), may , , theindychannel.com/entertainment/abc- s- the- middle- loaded- with- hoosier- connections. . for the imagined midwest as both celebrated and derided on american television, see victoria johnson, heartland tv: prime time television and the struggle for u.s. identity (new york: new york university press, ). treatment of locally advanced non-small cell lung cancer - neoadjuvant or adjuvant chemotherapy: e - copyright © by the international association for the study of lung cancer s journal of thoracic oncology • volume , number , supplement , august th world conference on lung cancer . andré f, grunenwald d, pignon jp, et al. survival of patients with resected n non- small-cell lung cancer: evidence for a subclassification and implications. j clin oncol ; : - . albain ks, scott cb, rusch vr, et al.. phase iii comparison of concurrent chemo- therapy plus radiotherapy (ct/rt) and ct/rt followed by surgical resection for stage iiia(pn ) non-small cell lung cancer (nsclc): initial results from intergroup trial (rtog - ). proc am soc clin oncol ; : , (abstr) . albain ks, rusch vw, crowley jj, et al. concurrent cisplatin/etoposide plus chest radiotherapy followed by surgery for stages iiia (n ) and iiib non-small-cell lung cancer: mature results of southwest oncology group phase ii study . j clin oncol ; : - . choi nc, carey rw, daly w, et al. potential impact on survival of improved tumor downstaging and resection rate by preoperative twice-daily radiation and concurrent chemotherapy in stage iiia non-small cell lung cancer. j clin oncol ; : - . betticher dc, hsu schmitz sf, tötsch m, et al. mediastinal lymph node clearance after docetaxel-cisplatin neoadjuvant chemotherapy is prognostic of survival in patients with stage iiia pn non-small cell lung cancer: a multicentric phase ii trial. j clin oncol ; : - . van meerbeeck jp, kramer gw, van schil pe, et al. randomized controlled trial of resection versus radiotherapy after induction chemotherapy in stage iiia-n non-small- cell lung cancer. j natl cancer inst ; : - . van raemdonck de, schneider a, ginsberg rj. surgical treatment for higher stage non-small cell lung cancer. ann thorac surg ; : - . downey rj, martini n, rusch vw, et al. extent of chest wall invasion and survival in patients with lung cancer. ann thorac surg ; : - . rusch vw, giroux dj, kraut mj, et al. induction chemoradiation and surgical resection for superior sulcus non-small-cell lung carcinomas: long-term results of southwest oncology group trial (intergroup trial ). j clin oncol ; : - . sakurada a, sagawa m, usuda k, et al. the significance of surgical treatment for t lung cancer. kyobu geka ; : - . shirakusa t, kawahara k, iwasaki a, et al. extended operation for t lung carcinoma. ann thorac cardiovasc surg ; : - . fukuse t, wada h, hitomi s. extended operation for non-small cell lung cancer invad- ing great vessels and left atrium. eur j cardiothorac surg ; : - . grunenwald dh. resection of lung carcinomas invading the mediastinum, including the superior vena cava. thorac surg clin ; : - . spaggiari l, thomas p, magdeleinat p, et al. superior vena cava resection with pros- thetic replacement for non-small cell lung cancer: long term results of a multicentric study. eur j cardiothorac surg ; : - . mitchell jd, mathisen dj, wright cd, et al. clinical experience with carinal resection. j thorac cardiovasc surg ; : - . yatsuyanagi e, hirata s, yamakazi k, et al. anastomotic complications after broncho- plastic procedures for nonsmall cell lung cancer. ann thorac surg ; : - . demeester tr, albertucci m, dawson pj, montner sm. management of tumor adher- ent to the vertebral column. j thorac cardiovasc surg ; : - . grunenwald d, mazel c, girard p, et al. total vertebrectomy for en bloc resection of lung cancer invading the spine. ann thorac surg. ; : - . komaki r, roth ja, walsh jl, et al. outcome predictors for patients with superior sulcus tumors treated by multidisciplinary approach at the university of texas md anderson cancer center. int j radiat oncol biol phys ; : - . gandhi s, walsh gl, komaki r, et al. a multidisciplinary surgical approach to superior sulcus tumors with vertebral invasion. ann thorac surg ; : - . grunenwald dh, mazel c, girard p, et al. radical en bloc resection for lung cancer invading the spine. j thorac cardiovasc surg ; : - . stamatis g, eberhardt w, stüben g, et al. preoperative chemoradiotherapy and surgery for selected non-small cell lung cancer iiib subgroups: long-term results. ann thorac surg ; : - . grunenwald dh, andre f, le pechoux c, et al. benefit of surgery after chemoradio- therapy in stage iiib (t and/or n ) non-small cell lung cancer. j thorac cardiovasc surg ; : - e - treatment of locally advanced nsclc, mon, sept , : – : treatment of locally advanced non-small cell lung cancer - neoadjuvant or adjuvant chemotherapy vokes, everett e. the university of chicago, chicago, il, usa over recent years combined modality therapy has become firmly es- tablished as a standard in the treatment of patients with locoregionally advanced unresectable non-small cell lung cancer (nsclc). in direct comparison, concomitant chemoradiotherapy was shown repeatedly to be superior to induction chemotherapy. a case can be made that concomitant chemoradiotherapy primarily addresses locoregional dis- ease while induction chemotherapy might be better suited to eradicate distant micrometastatic foci. therefore, continuing sequential and con- comitant therapy might be beneficial and the addition of induction or adjuvant chemotherapy to the concomitant chemoradiotherapy standard have been investigated. calgb compared concomitant chemoradiotherapy using the carboplatin paclitaxel platform versus induction chemotherapy with carboplatin and paclitaxel for two cycles followed by identical chemo- radiotherapy. while a numerical trend favored the induction chemo- therapy arm, there was no significant advantage for overall survival. similarly, the hoosier oncology group evaluated the administration of concomitant chemoradiotherapy using the cisplatin etoposide platform with or without three additional cycles of consolidation chemotherapy with docetaxel. this study was based on promising pilot data generated by the southwest oncology group. again, the study showed no signifi- cant survival advantage from the addition of concomitant chemoradio- therapy. therefore, at the present time, concomitant chemoradiotherapy should be regarded as the standard approach for most patients with unresectable non-small cell lung cancer. induction chemotherapy may have a role for patients with poor perfor- mance status who may not be candidates to undergo aggressive chemo- radiotherapy. certain targeted agents might be appropriate to investi- gate in the consolidation setting under carefully defined experimental conditions. this need is highlighted by the recent experience in swog in which the administration of gefitinib as maintenance therapy was found to decrease survival rates. session e : new technology for diagnosis e - new technology for diagnosis, tue, sept , : – : autofluorescence bronchoscopy and optical coherence tomography lam, stephen british columbia cancer agency and the university of british columbia, vancouver, bc, canada the lung is an internal organ consisting of a complex branching system of airways leading to gas exchange units. lung cancer consists of several cell types instead of a single cell type. different cell types are preferentially located in different parts of the bronchial tree. there is no single method that can detect pre-invasive cancer in the entire bronchial epithelium and allow simultaneous tissue sampling for pathological diagnosis and molecular profiling. while computerized tomography, magnetic resonance imaging and ultrasound can detect objects in the sub-milliliter scale, photonic imaging can detect structural and func- tional changes in cells and tissues down to the micron and sub-micron b - sputum methylation analysis detects patients with lung cancer journal of thoracic oncology • volume , number , supplement , august th world conference on lung cancer copyright © by the international association for the study of lung cancers vival for blacks and whites with localized adenocarcinoma was . % and . %, while survival was . % and . % for blacks and whites with squamous cell carcinoma, respectively. with regard to age group comparisons, there was a significant decrease in -year survival with advancing age among those with localized disease. conclusions: differences in relative survival of lung cancer was strongly dependent upon histologic subtype. indeed, five-year survival for adenocarcinoma of the lung is significantly superior to that of other histologic subtypes. improved survival in adenocarcinoma appears to be related to a more favorable stage distribution, which facilitates cura- tive therapies b - prevention & early detection + epidemiology, tue, : - : sputum methylation analysis detects patients with lung cancer thunnissen, erik shivapurkar, narayan stastny, victor drift, miep v. bolijn, anne hol, bernard yinghui, wang feng, ziding gazdar, adi prinsen, clemens vumc pathology, amsterdam, the netherlands hamon center for therapeutic oncology research, ut southwestern medical center, dallas, tx, usa umcn pulmonology, nijmegen, the netherlands cwz pathology, nijmegen, the netherlands cwz pulmonology, nijmegen, the netherlands fred hutchinson cancer research center, seattle, wa, usa background: sputum methylation analysis is feasible and by some studies suggested to show an increased risk for lung cancer. the aim of this study was to examine methylation markers in sputum of patients with lung cancer and controls. methods: in the canisius wilhelmina hospital nijmegen sputum has been prospectively collected for several years of patients with lung cancer and controls. from this bank cases with lung cancer and controls, most of them with copd dna was retrospectively extracted. samples were subsequently randomized and blinded (tdb). coded samples were exchanged between collaborators unaware of its origin. quantitative msp was performed for apc, cytoglobin, mgmt, -ost- , p , rassf a, and tcf . follow-up was retrieved inde- pendently (mvdd). pathology diagnosis was based on biopsy or only cytology. after submission of methylation data (to zf with subsequent information of tdb) statistical analysis was performed (wy,zf). results: the pathology diagnosis was squamous cell carcinoma (n= ), adenocarcinoma (n= ), sclc (n= ) or nsclc not further specified. the of the controls had a gold score for copd > . sputum samples were collected within months of lung cancer diagno- sis. follow-up of copd cases was minimally years. reproducibility of methylation analysis between two labs was poor for mgmt. rassf a showed hypermethylation in % of the sclc and % of the nsclc lung cancer cases with a specificity of and %, respec- tively. for two markers the sensitivity showed a decrease in sensitivity for sclc and nsclc to %, and %, respectively. the specificity increased for sclc and nsclc to and %, respectively. conclusions: this blinded study shows that hypermethylation in spu- tum is with two markers detects % of the lung cancer patients. session b : nsclc: new paradigm in radiation therapy tuesday, september b - nsclc: new paradigm in radiation therapy, tue, : - : phase iii trial of cisplatin (p) plus etoposide (e) plus concurrent chest radiation (xrt) with or without consolidation docetaxel (d) in patients (pts) with inoperable stage iii non-small cell lung cancer (nsclc): hog lun - /uso- hanna, nasser neubauer, marcus mcgarry, ronald govindan, ramaswamy johnson, cynthia yiannoutsos, constantin breen, tim white, angela ansari, rafat einhorn, lawrence indiana university, indianapolis, in, usa kansas city cancer centers, overland park, ks, usa university of kentucky, lexington, ky, usa washington university, st. louis, mo, usa hoosier oncol- ogy group, indianapolis, in, usa northern indiana cancer research consortium, south bend, in, usa background: concurrent chemoradiation is standard treatment for pts with inoperable stage iii nsclc. a previously reported single- arm, phase ii study by swog (gandara et al jco ) suggested d following ep/xrt further improved survival. we report results from a randomized, prospective phase iii trial comparing ep/xrt with or without consolidation d. methods: eligible pts had inoperable, stage iiia/b nsclc, ps - at study entry, fev- > l, and < % wt loss in the preceding mos. pts (n= ) received p mg/m iv d , , , and e mg/m iv d - , - concurrently with chest xrt to cgy. non-progressing pts (ps - ) were randomized to d mg/m iv every d for cycles vs observation (o). the primary endpoint was to compare os (kaplan- meier analysis). a multivariate parametric accelerated failure time model was performed to identify factors that affected survival. accrual of pts to randomize was planned to demonstrate a difference in mst of vs mos ( % -sided alpha, % power). based upon evidence of futility (predefined as p> . ), a dsmb recommended early termination after an analysis of the initial pts. results: median f/u . mos. pt characteristics (n= ): %/ % f: m; median age ; . %/ . % iiia/b; staged with pet . %; fev- > . %; current smoker . %. g / toxicities during ep/xrt included . % febrile neutropenia (fn), . % esophagitis. of pts ( . %) were randomized to d (n= ) or o (n= ). . % randomized to d received cycles. g / toxicities during d included: . % fn, . % pneumonitis. . % of pts were hospitalized during d (vs . % in o arm) and . % died due to d. factors predictive for im- proved survival included age < , fev- > , and hemoglobin > at baseline. pfs for d was . vs . mos for o (p= . ). the mst for all pts (n= ) was . mos; mst for d was . mos ( % c.i. . - ) vs . mos for o ( % c.i. . - . ) (p= . ). conclusions: the mst with ep/xrt was higher than historical controls; however, consolidation d does not further improve survival, is associated with significant toxicity including an increased rate of hospitalization and premature death. predicting the height growth of oak species (quercus) reproduction over a -year period following clearcutting forest ecology and management ( ) – contents lists available at sciencedirect forest ecology and management journal homepage: www.elsevier.com/locate/foreco predicting the height growth of oak species (quercus) reproduction over a -year period following clearcutting http://dx.doi.org/ . /j.foreco. . . - /� elsevier b.v. all rights reserved. ⇑ corresponding author. e-mail address: jenkinma@purdue.edu (m.a. jenkins). j. travis swaim a,d, daniel c. dey b, michael r. saunders a, dale r. weigel b, christopher d. thornton c, john m. kabrick b, michael a. jenkins a,⇑ a department of forestry and natural resources and hardwood tree improvement and regeneration center, purdue university, west state street, west lafayette, in , united states b usda forest service, northern research station, abnr building, columbia, mo , united states c hoosier national forest, constitution avenue, bedford, in , united states d malheur national forest, highway south, hines, or , united states a r t i c l e i n f o article history: received july received in revised form december accepted january keywords: competition long-term data stand development even-aged central hardwood region a b s t r a c t we resampled plots from a repeated measures study implemented on the hoosier national forest (hnf) in southern indiana in to investigate the influence of site and seedling physical attributes on height growth and establishment success of oak species (quercus spp.) reproduction in stands regenerated by the clearcut method. before harvest, an array of physical attributes were documented for individual stems of advance reproduction. across all surveys, the same characteristics were remeasured in years , , and for all reproduction types (advance reproduction, stump sprouts, and new seedlings). in order to characterize topo-edaphic conditions, soil samples were collected and analyzed in , and slope aspect, slope percent, and slope position were measured in the field. random forest (rf) analysis was used to determine the best physical and environmental predictors of height growth for oak species and their competitors in developing stands. overall, advance reproduction of oak species fared poorly fol- lowing harvests. sprout-origin oak stems proved stronger competitors in developing stands, although their abundance relative to competing species was quite low. advance and sprout origin maple (acer spp.) stems, along with new seedlings of black cherry (prunus serotina ehrh.) and yellow-poplar (liriodendron tulipifera l.), quickly overtopped oak advance reproduction and established dominance in the developing canopy. the height of stems during prior sampling periods was the best overall predictor of stem height in subsequent sampling periods. species was also an important predictor of stem height. comparatively, environmental variables were poor predictors of height growth of individual stems throughout the study, although more mesic aspects, greater cation exchange capacity, and greater soil magnesium saturation were associated with greater height of non-sprout origin stems from species groups other than oak or hickory in year . our results suggest that overstory removal has driven stand demographics towards species favored by infrequent large-scale disturbance events such as clearcutting. without post-harvest treatments to control competitors, oak regeneration on more mesic sites is unlikely to recruit into developing stands. � elsevier b.v. all rights reserved. . introduction during the late th century, silvicultural clearcutting was used extensively across the hardwood forests of the eastern united states (roach and gingrich, ; sander and clark, ; smith, ; walker, ). a major goal of this management technique was to regenerate new stands of shade-intolerant and mid- tolerant species, typically aspen (populus spp.) and oak (quercus spp.) species, respectively. the open conditions created by clearcutting were thought to favor the establishment and persis- tence of oak reproduction over shade-tolerant competitors because oak reproduction rarely persists or grows successfully in the understory of mature hardwood forests due to shading from over- story and subcanopy cover (dey, ). furthermore, clearcutting was favored because it was considered economically advantageous due to high yields and ease of implementation (clark and watt, ; sander and clark, ). subsequent research has shown that the past widespread use of silvicultural clearcutting failed to perpetuate oak dominance, http://crossmark.crossref.org/dialog/?doi= . /j.foreco. . . &domain=pdf http://dx.doi.org/ . /j.foreco. . . mailto:jenkinma@purdue.edu http://dx.doi.org/ . /j.foreco. . . http://www.sciencedirect.com/science/journal/ http://www.elsevier.com/locate/foreco j. travis swaim et al. / forest ecology and management ( ) – especially on productive sites (george and fischer, ; jenkins and parker, ). this is cause for concern considering the eco- nomic and ecological importance of oak species within the central hardwood region (chr). for example, oak species comprise % of all growing stock in indiana and are considered the most important timber species in the state (bratkovich et al., ). in recent years, the desire to maintain oak ecosystems has intensified because oak is recognized as indispensable to addressing non-timber manage- ment objectives including maintaining understory plant diversity, sustaining wildlife populations, and providing habitat and food for neotropical migratory birds (ellison et al., ; mcshea et al., ; groninger and long, ). researchers have hypothesized that the success of oak species in post-harvest, even-aged stands largely depends on the abun- dance of advance reproduction that accumulates in the under- story prior to harvest, as well as sprouts from stumps of harvested overstory trees (sander and graney, ; dey, ; weigel and peng, ). the success of this advance reproduction may depend on the height and diameter of these stems prior to release (belli et al., ), which has been related to the potential of the oak root systems to support vigorous growth after release and allow oak stems to successfully compete with early-seral species (sander, ; sander and graney, ; brose et al., ). research has also shown that upland oak species are better able to establish and presist on less productive sites because they are better adapted to more nutrient-poor or xeric site conditions than their primary competitors (larsen and johnson, ; kabrick et al., , ). under such conditions, oak species are afforded a competitive advantage due to the rapid development of a large taproot, the ability to physiologically function under high water stress, flexibility in maintaining high root:shoot ratios through recurrent shoot dieback, and large genetic variability in drought tolerance within species (abrams, ; pallardy and rhoads, ; parker and dey, ; johnson et al., ). therefore, site quality, in part, determines the abundance of advance oak repro- duction that exists prior to harvest, with a greater abundance of oak species reproduction accumulating as site quality decreases (kabrick et al., ). to examine the competitive relationships between oak and other species, chronosequence studies have been widely used to represent different stages of post-harvest stand development (hilt, ; jenkins and parker, ; brashears et al., ; morrissey et al., , ). generally, the substitution of space for time in chronosequence studies allows the generation of hypotheses related to long-term vegetation dynamics (bakker et al., ), but actually testing these hypotheses depends upon long-term data (pickett, ). in the case of oak species establish- ment, long-term data that tracks the fate of individual stems of advance reproduction and stump sprouts under a range of site con- ditions are critical to testing hypotheses related to the post-harvest establishment and survival of oak species under different site con- ditions. in controlled experiments, the physical attributes of indi- vidual stems have been shown to influence stem growth and survival under different light regimes (dey and parker, ). understanding how growth varies through time in response to competition and environmental conditions provides valuable insight into the optimal timing of post-harvest treatments to release oak stems in developing stands. however, while studies have examined the effects of landtype on regeneration dynamics (kabrick et al., ), and how physical characteristics of under- planted seedlings influence survival (dey et al., ), we found no long-term studies that employed repeated measures to investi- gate the influence of both seedling attributes and the physical environment on height growth of multiple types of natural reproduction in post-harvest stands. in , a long-term, repeated measure study was implemented across six hardwood forest sites on the hoosier national forest (hnf) in southern indiana to examine the competitive ability and development of individual stems of oak and competing species after clearcut harvesting. during a -year sampling period ( – ), morphological/physical attributes of advance repro- duction were measured across six study sites. we also collected additional measurements of all stump sprouts and new seedlings, as well as a suite of physiographic and edaphic variables. based upon these examinations, we address three primary questions and associated hypotheses in this study: . how well does advance reproduction of oak species compete against other species through time during early stages of stand development? we hypothesize that while advance reproduc- tion of oak species will remain a component of developing stands, its relative density will decrease through time due to competition with other species. . how well do pre-harvest physical attributes of advance repro- duction indicate the dominance of individual stems in post- harvest stands? we hypothesize that for any resample (years , , and ), height and root collar diameter during the previ- ous sample will be the strongest predictor of oak reproduction height regardless of origin type. . how well do site conditions predict the growth of advance reproduction in post-harvest stands? we hypothesize that while secondary to the physical characteristics of seedlings, environmental variables associated with low water and nutri- ent availability will be significant predictors of oak reproduc- tion height during stand development. while poor site conditions reduce the growth of all tree species, including oaks, a greater reduction in the growth of mesophytic species results in reduced competition for oak species. . methods . . study sites our six study sites were located within two adjoining natural regions in southern indiana; the highland rim natural region and shawnee hills natural region (homoya et al., ; table ). both natural regions are unglaciated with terrain that varies from steep slopes and narrow ridges in the highland rim to broad ridge- tops in the shawnee hills. soils in both regions are acid silt loams derived from sandstone, siltstone, and loess. uplands are domi- nated by oak species including chestnut oak (quercus prinus l.) in the highland rim and black oak (q. velutina lam.), white oak (q. alba l.), and scarlet oak (q. coccinea muenchh.) in the shawnee hills. hickory species (carya) are present as a secondary canopy component in association with oak species. mesic ravines and slopes harbor species such as american beech (fagus grandifolia ehrh.), northern red oak (q. rubra l.), sugar maple (acer saccharum marsh.), white ash (fraxinus americana l.), and black walnut (juglans nigra l.). . . plot establishment in , six mature oak-hickory stands ranging in size from . to . ha were harvested on the hnf (table ). all stands were clearcut for merchantable timber between the months of april and august. after removal of merchantable timber, all remaining trees p . cm dbh (diameter at breast height) were cut and left on-site.; only trees < . cm remained standing on each site. prior to harvest, overstory stand data were collected from sixty- one m circular plots across the six stands. between april and june , , six to permanent overstory plots were table location and site descriptions of six study sites on the hoosier national forest in southern indiana. site location lat/long (n/ w) size (ha) # of plots topographya parent materialsa soilsa series soil classification basal areab (m ha� ) dominant overstory speciesc . . dissected uplands with steep slopes siltstone, sandstone, shale acid silt loams brownstown-channery silt loam and gnawbone silt loam typic dystrudepts and typic hapludults . ± . chestnut oak, white oak, black oak� . . . broad ridgetops and flats sandstone, loess acid silt loams wellston-tipsaw- adyeville complex ultic hapludalfs . ± . chestnut oak, white oak, black oak� . . . broad ridgetops and flats sandstone, loess acid silt loams apalona-zanesville silt loams oxyaquic fragiudalfs . ± . sugar maple, hickory spp., black oak, white oak � . . . rugged hills sandstone, limestone silt loams wellston-adyeville-ebal complex ultic hapludalfs . ± . hickory spp., white oak, black oak . . . rugged hills sandstone, limestone silt loams wellston-adyeville-ebal complex ultic hapludalfs . ± . yellow-poplar, maple spp., hickory spp., black oak � . . . broad ridgetops and flats sandstone, loess acid silt loams wellston silt loam ultic hapludalfs . ± . white oak, yellow- poplar, hickory spp.� . a from homoya et al. ( ). b preharvest total across all species; mean ± standard error. c dominant species, in order, based upon mean stand basal area (m ha� ) from swaim ( ). j. travis swaim et al. / forest ecology and management ( ) – located and sampled in each stand (table ) using a stratified ran- dom design based upon topographic position. plots were situated within each stand to encompass as many aspect-slope combina- tions as possible (table ) and were positioned a minimum of . m from the edge of the clearcut in order to avoid any edge effects from surrounding mature forest. during the summer of , we resampled of the original plots. two plots in stand were never harvested and therefore were not resampled in . four plots in stand could not be relocated. within each m overstory plot, three . m circular sub- plots were established in , prior to harvest, to assess repro- duction. the m overstory plot was divided into thirds using a randomly chosen azimuth to locate the first dividing line. one hundred twenty degrees were added to this azimuth to locate the second line, and another degrees added to locate the third line. within each third, a permanent . m circular plot was ran- domly located by adding a random number from between and to the azimuth of the dividing line. the . m subplots were not allowed to overlap, therefore they were spaced at least . m table number sampled, aspect-slope position classes, and site index ranges of plots resampled in on the hoosier national forest. stand no. plots sampled aspect-slope position of plotsa site index rangeb sl, sm, wm, nm, el, em, eu – , chestnut oak sm, wl, wm, wu, nm, nu, em – , chestnut oak su, wl, wm, wu – , black oak sl, sm, su, wu, nu, eu – , black oak sl, sm, nl, nm , black oak nm, nu, em, eu – , white oak a u = upper slope, m = mid slope, l = lower slope, n = north ( – �), e = east ( – �), s = south ( – �), w = west ( – �). b ranges are given for most common oak species in stand. values represent mean si for plots where species occurred. stand value based upon one black oak tree sampled. apart. subplots were also spaced so they were at least . m from both the center and the outside edge of the main plot. . . plot sampling all trees p . cm dbh were measured by species in each m overstory plot during the and surveys. in , we also collected replicate soil samples from each plot. a total of four samples per plot were collected from the top cm of the a-horizon m from plot center in each cardinal direction. in each plot, the samples were pooled for analysis. additional site variables measured included aspect, percent slope, slope position, and site index of each plot (table ). within each . m subplot during the , , , and surveys, all reproduction was tallied by species into . m height classes with stems > . m tallied into a single class. a rep- resentative sample of stems < . cm dbh were selected for detailed measurements and mapped by azimuth and distance from the sub- plot center stake. measurements taken prior to harvest in (year ) on advance reproduction stems were form, ground diam- eter class, total height, dbh, and number of live stems (see table for detailed descriptions of measurements). the same individual advance reproduction stems were remeasured in (year ), (year ), and (year ). new seedlings and post- harvest stump sprouts were mapped and measured beginning in year . post-harvest measurements for all permanent stems included origin (advance reproduction vs. stump sprout vs. post- harvest seedling), form, total height, dbh, number of live stems, and crown class. when multiple stems were growing from a single root system, only the tallest stem was selected for measurements. harvest induced damage to advance reproduction stems was documented and advance reproduction was reclassified into two origin classes: ( ) undamaged during harvest or ( ) damaged (cut or broken) and successfully resprouted following harvest (see table for a full description of origins). . . data preparation summary statistics calculated for each overstory plot included density (stems ha� ) and basal area (m ha� ) by species. soil table description of response and predictor variables used in rf analysis. variable code description median mean range response variables h height of individual trees years after clearcut harvests (m) . . . – . h height of individual trees years after clearcut harvests (m) . . . – . h height of individual trees years after clearcut harvests (m) . (all reproduction) . . – . . (advance reproduction only) . . – . biological predictor variables sp species class wo = white oak group, ro = red oak group, hi = hickory group, rm = red maple, sm = sugar maple, yp = yellow- poplar, bc = black cherry, ab = american beech, oc = other commercial species, nc = non-commercial species h height of individual trees years after clearcut harvests (m) . . . – . h height of individual trees years after clearcut harvests (m) . (all reproduction) . . – . . (advance reproduction only) . . – . h height of advance reproduction prior to clearcut harvests; year (m) . . . – . f form class for individual trees years after clearcut harvests = true apical dominance, terminal had not died back, = terminal had died back and a lateral branch had assumed dominance, = no dominant leader, flat topped, bushy, or almost prostrate f form class for individual trees years after clearcut harvests = true apical dominance, terminal had not died back, = terminal had died back and a lateral branch had assumed dominance, = no dominant leader, flat topped, bushy, or almost prostrate f form class of individual trees prior to clearcut harvests; year = true apical dominance, terminal had not died back, = terminal had died back and a lateral branch had assumed dominance, = no dominant leader, flat topped, bushy, or almost prostrate cc crown canopy class years after clearcut harvests = suppressed or overtopped, = intermediate, = codominant, = dominant cc crown canopy class years after clearcut harvests = suppressed or overtopped, = intermediate, = codominant, = dominant or individual stem origin class = advance reproduction, undamaged, = advance reproduction, damaged during harvest and sprouted, = stump sprout after harvest, = new seedling established after harvest gd ground (root collar) diameter class for advance reproduction (cm) = . , = . – . , = . – . , = . – . , = > . environmental predictor variables acode aspect class grouped by azimuth degrees = – �, = – � and – �, = – � and – �, = – � %c soil carbon (%) . . . – . %n soil nitrogen (%) . . . – . %ca soil calcium cation saturation (%) . . . – . %mg soil magnesium cation saturation (%) . . – . %k soil potassium cation saturation (%) . . – . %h soil hydrogen cation saturation (%) . . – ppmp soil phosphorus (ppm) – %om soil organic matter (%) . . – . ph soil ph . . . – . cec soil cation exchange capacity (meq/ g) . . – . tslpos slope position; = plot located at base of slope, = plot located at ridge summit (%) . . – %slope slope (%) . – si site index classa = poor (si = – ), = average (si = – ), = good (si p ) a site index classes derived from carmean, ; carmean et al., . j. travis swaim et al. / forest ecology and management ( ) – samples were analyzed for cation exchange capacity (meq/ g), phosphorus content (ppm), potassium saturation (%), calcium sat- uration (%), magnesium saturation (%), hydrogen saturation (%), organic matter (%), and ph by a&l analytical laboratories in mem- phis, tennessee. carbon (%) and nitrogen (%) were determined using an ecs chnso analyzer (costech analytical technolo- gies, inc., valencia, ca) located in the forest ecology, silviculture, and soils laboratory, purdue university, west lafayette, indiana. aspect was transformed to a linear scale ranging from to , with zero value at southwest [transformed aspect = cos( - aspect) + ; beers et al., ]. transformed aspects were then grouped into four classes ranging from most xeric (class ) to most mesic (class ; table ). slope position was assessed as a relative proportion in relation to a ridge summit or drainage (e.g., ridge/ summit = , mid slope = , drainage = ). when possible, site index class (carmean, ; carmean et al., ) was determined in by measuring total heights and ages of between and dominant and/or codominant black oak trees within each m overstory plot. when there were too few or no black oaks on a plot, scarlet, white, or chestnut oak were substituted. for ease of analysis, and to ensure sampling uniformity across stands, less common species were grouped with closely related species, or with species possessing similar silvical characteristics. for random forest analysis (described below), the white oak group (wo) included white oak, chestnut oak, chinkapin oak (q. muehlen- bergii engelm.), and post oak (q. stellata wangenh.), although our table error statistics for rf runs with all response variables. error for regressions is measured by percent variance explained and the mean of squared residuals (msr; liaw and wiener, ; cutler et al., ). response variables include individual stem heights in years (h ), (h ), and (h ). response variable number of predictor variables number of rf trees number of variables per split % variance explained msr h (advance reproduction only) . . h (all reproduction) . . h (all reproduction) . . h (all reproduction) . . j. travis swaim et al. / forest ecology and management ( ) – plots contained few individuals of these species other than white oak (n = for post oak, n = for chinkapin oak). the red oak group included northern red oak, black oak, and scarlet oak. hickory spe- cies were not separated in previous sampling periods, but in the hickory group (hi) included shagbark [c. ovata (mill.) k. koch], pignut [carya glabra (mill.) sweet], bitternut [c. cordiformis (wan- genh.) k. koch], and mockernut [c. tomentosa (poir.) nutt.] hickory. red maple (rm; a. rubrum l.), sugar maple (sm), american beech, black cherry (prunus serotina ehrh.), and yellow-poplar (lirioden- dron tulipifera l.) were used as individual species in rf analyses. all species with commercial value that are not mentioned above were placed into the commercial group (com). the com group included white ash, blackgum (nyssa sylvatica marshall), bigtooth aspen (populus tremuloides michx.), and elm (ulmus spp.). species of low commercial value were put into the non-commercial group (nc). major species of the nc group included american hornbeam (carpinus caroliniana walter), eastern redbud (cercis canadensis l.), flowering dogwood (cornus florida l.), eastern hophornbeam [ostrya virginiana (mill.) k. koch], and sassafras [sassafras albidum (nutt.) nees]. . . stand density analysis we used tukey hsd multiple comparison tests (a = . ) to examine changes in the stem density of species groups through time at , and years following harvest. stems were grouped into three height classes for analysis; < . m, . – . m, and > . m. for year , we included only advance repro- duction (seedling and sprout origin), whereas for all other years we included only established individuals p . m height. for this analysis, yellow-poplar and black cherry were combined with the com group created for random forest analysis (‘‘other commer- cial” in table ) and red maple, sugar maple and american beech were combined into a single group (‘‘maple and beech” in table ). . . random forest analysis because our dataset included a large number of categorical and continuous variables (table ), we used random forests (breiman, ; liaw and wiener, ), an ensemble classifier that employs multiple decision trees. this nonparametric analysis technique requires few assumptions about data distributions and relation- ships between variables (zhang and singer, ). this and similar monte carlo randomization techniques have become popular in ecological studies because of their high classification accuracy, novel methods of determining variable importance, ability to model complex interactions between predictor variables, and flex- ibility to perform several types of statistical analysis, including regression (cutler et al., ). we used rf to model four response variables: ( ) height of all individual reproduction types in year , ( ) height of individual advance reproduction in year , ( ) height of all individual repro- duction in year , and ( ) height of all individual reproduction in year and (a detailed list of response and predictor variables is provided in table ; a description of error statistics is provided in table ). we analyzed advance reproduction separately in year because only advance reproduction would have a height prior to harvest (h ); therefore, origin could serve as a proxy for height in year . we also examined advance reproduction at year sepa- rately to determine if any site variables were associated with regeneration in pre-harvest stands. all data were analyzed using the r statistical computing pro- gram, version . . (r development core team, ); rf analy- ses were implemented using the packages randomforest (liaw and wiener, ) and party (hothorn et al., a, b; strobl et al., , ). . results . . stand density prior to harvest in , densities of white and red oak group advance reproduction were lower than those of maple-beech, other commercial, and non-commercial species groups (table ). the density of hickory species was lowest overall ( ± stems ha� ), while density of non-commercial species was the highest of any group ( ± stems ha� ). most advance reproduction was small (< . m tall), ranging from % (maple-beech) to % (red oak) of stems in a given species group (table ). in absolute terms, the maple-beech ( stems ha� ) and non-commercial groups ( stems ha� ) comprised most of the larger stems (> . m), combining for roughly % of advance reproduction in that size class. six years after harvest the white oak, red oak, and hickory groups continued to occur in very low densities relative to the maple- beech, other-commercial, and non-commercial species groups. maple-beech was the third most abundant group ( ± stems ha� ), behind other-commercial ( ± stems ha� ) and non-commercial ( ± stems ha� ) groups (table ). the non-commercial, other-commercial, and maple- beech groups continued to dominate the taller height classes (col- lectively % of stems . – . m tall and % of stems > . m tall). the white oak was slightly more abundant than the red oak group in larger size classes in both absolute and relative terms, but both groups were still predominately in the shortest height class (< . m; table ). as stands entered stem exclusion and began to differentiate and self-thin, densities of all species declined in year and again in year . the red oak, white oak, and hickory groups saw the least precipitous decline of all species (collectively- % from year to year ), with a majority of remaining stems progressing into the largest size class (> . m; table ). while non-commercial stems still dominated the stand in year ( ± stems ha� ; % of all stems), their density declined % by year , most notably in the largest size class ( % reduction; table ). by year , maple-beech and other commercial species comprised a majority of the stand ( %) and of the largest stems ( %; table ). table species composition and size distribution (mean ± standard error; per ha) of study sites at harvest ( years) and , and years following harvest. year includes only advance reproduction (seedlings and sprouts), whereas all other years include only established individuals . m or larger. differences of total abundance within year were tested with tukey hsd multiple comparison tests; letters denote differences with p < . among the species groups. height classes are defined as s = < . m, m = . – . m, and l = > . m. other commercial includes yellow-poplar, black cherry, and the com group created for random forest analysis. year ht class species group total red oak white oak hickory maple-beech other commercial non-commercial ± a ± a ± a ± bc ± ab ± c , ± s , m l ± a ± a ± a ± b ± bc ± c ± s m l ± a ± ab ± a ± bc ± c ± c ± s m l ± a ± ab ± a ± c ± c ± bc ± s m l j. travis swaim et al. / forest ecology and management ( ) – . . rf models our rf analyses explained between % (advance reproduction, year ) and % (all reproduction types, year ; table ) of vari- ance. the percent of variance explained increased with year since harvest (table ). mean of squared residuals for all reproduction increased from . in year to . in year . . . . year height of all reproduction the most important predictors for height of all reproduction in year were biological variables origin (or; i.e., new seedlings, advance reproduction and stump sprouts) and species class (fig. a and b). there was little difference in the importance of other predictor variables. the percent saturation of several ions (n, mg, h, c) were the most important environmental predictors, but all were much less significant than origin and species class (fig. a and b). in year , reproduction of oak, hickory, and american beech from stump sprouts (n = , fig. a, node ) were taller when compared to that from advance reproduction of the same species class (n = , node ). the tallest groups in year were stump sprouts of red maple and yellow-poplar (n = , fig a, node ). no new oak, hickory, or beech seedlings were doc- umented in year . mg saturation (%mg) > % and cec > . were associated with greater height of non-stump sprout reproduction (fig a, nodes and ). . . . year height of reproduction originating as advance reproduction the most important predictors for the height of advance repro- duction in post-harvest year were the height of advance repro- duction in year (h ), i.e., before harvest, and species class (sp; fig. a). these predictor variables were � times more important than the next greatest predictor in the model, ground diameter class (gd). there was little differentiation in importance of predic- tor variable other than height in year , species class, and ground diameter class. the most important environmental predictor variable was percent slope, followed by potassium cation satura- tion (%k) and aspect (acode), although all environmental and edaphic variables displayed little importance relative to species class and height in year (fig. a). stems of red oak species, white oak species, hickory species, red maple, and american beech that were . m height before harvest were associated with the shortest overall heights in year (n = , fig. b, node ). the tal- lest reproduction in year six were stems that were > . m height prior to harvest (n = , fig. b, node ). the shortest black cherry and sugar maple advance reproduction in year ( . m; n = , fig. b, node ) were still taller by year than even the tallest oak stems that originated as advance reproduction (n = , fig. b, node ). . . . year height of all reproduction height in year (h ) proved the strongest predictor for height of stems in year (h ; fig. c and d). other important biological predictors included species class, crown class in year (cc ), and origin. environmental variables appeared to have little relationship to the height of stems in year (fig. c and d). oak, hickory, black cherry, and yellow-poplar stems that were the tallest groups in year remained in the tallest groups in year , although samples were smaller (n = , fig. b, nodes and ). yellow-poplar from the second tallest height class in year ( . m) appeared to be highly competitive in year (node ). . . . year height of all reproduction similar to year models, only biological predictors proved sig- nificant when predicting the year heights of all reproduction (fig. e and f). height in year was the best overall predictor, fol- lowed by crown class year (cc ), height in year , and species class. as in previous years, environmental variables were not strongly related to height in year (fig. e and f). the tallest indi- viduals in year were also the tallest individuals in year (nodes and ; fig. c). . discussion . . advance reproduction due to the lack of large oak advance reproduction, clearcut stands in southern indiana have shifted away from the oak- dominated forests that existed before harvest. in our study, most of the advance reproduction of oak species was < . m tall prior to harvest (table ). this pool of small seedlings remained the shortest of all reproduction present in year . a small number of oak stems with heights > . m did exist before clearcutting and fig. . variable importance plots from rf model predicting heights of all reproduction types in year (a and b), year (c and d), and year (e and f). %incmse = average increase in the mean square error when data for that variable are permuted while all others are left unchanged (thompson and spies, ). incnodepurity = average increase in node purity from splitting on the variable; node purity is measured by residual sum of squares (liaw and wiener, ). the best predictors are those that produce the purest nodes. predictor variables are described in table . j. travis swaim et al. / forest ecology and management ( ) – fig. . regression trees for individual tree height in year (a), year (b) and year (c) using the five most important predictor variables as determined by random forests analysis. a monte carlo randomization test was used to assess p-values at each node; the length of individual lines represents strength of split. n = number of stems in category, y = mean height of category. predictor variables are described in table . j. travis swaim et al. / forest ecology and management ( ) – fig. . (a) variable importance plots from rf model predicting heights of advance reproduction in year . predictor variables are described table . %incmse = average increase in the mean square error when data for that variable are permuted while all others are left unchanged (thompson and spies, ). incnodepurity = average increase in node purity from splitting on the variable; node purity is measured by residual sum of squares (liaw and wiener, ). the best predictors are those that produce the purest nodes. (b) regression tree for individual advance reproduction height in year using the five most important predictor variables as determined by random forests analysis. a monte carlo randomization test was used to assess p-values at each node; the length of individual lines represents strength of split. n = number of stems in category, y = mean height of category. predictor variables are described in table . j. travis swaim et al. / forest ecology and management ( ) – appeared to be competing for codominant positions in subsequent surveys. however, as we hypothesized, these larger stems were few in number compared to those of other species in the same canopy positions. without surface fire or surrogate disturbances to limit competition and promote root and subsequent height growth of oak seedlings, oak advance reproduction was unable to successfully compete against larger advance stems of competing species (brose et al., ). in partial support of our hypothesis, pre-harvest ground diame- ter class (gd) was the third most important predictor of advance j. travis swaim et al. / forest ecology and management ( ) – reproduction height in year (fig a), although its predictive value was much less than that of initial height and species. in planted red oak seedlings, dey and parker ( ) found that stem diameter near the root color provided an integrated measure of seedling growth potential because of its strong relationship with both shoot and root characteristics associated with better field performance. in southeastern ohio, sander ( ) found that the most desirable ground diameter at time of clearcut harvest for oak advance repro- duction was . – . cm; when damaged during harvest, sprouts from these larger stems averaged . m height by year and suc- cessfully competed for codominant canopy positions. however, smaller damaged stems were unable to produce competitive sprouts in subsequent stands, likely due to their low biomass of roots prior to damage (dey and parker, ). in our study, % of the advance reproduction we tracked had root collar diameters less than . cm, below the minimum diameter recommended by sander ( ). whether advance reproduction was damaged or not during harvest had little to no importance in predicting height of stems years later (fig. a), suggesting that both damaged and undamaged stems lacked competitive ability in new stands. the tallest stems of reproduction in year were black cherry, sugar maple, red maple, and yellow-poplar (fig. a). the relatively high number of maple advance reproduction in larger size classes before harvest allowed the genus to remain in dominant positions during subsequent stand development (table ). the shade- intolerance of black cherry and yellow-poplar seedlings suggests successful advance reproduction from those species germinated during the spring before harvest and benefitted from the light con- ditions created by complete overstory removal. other studies across the region have also observed dominance of these species in post-harvest stands (standiford and fischer, ; jenkins and parker, ; lhotka, ). for instance, in oak-hickory- dominated stands that were clearcut in southern illinois during the s, groninger and long ( ) found that maple, yellow-poplar, and black cherry were the most abundant species – years after harvest. . . other regeneration types in year , the tallest stems from advance reproduction were black cherry, sugar maple, red maple, and yellow-poplar (fig. a). advance reproduction of red and white oak group species, along with hickory species and american beech, were found in the smallest height class in year . some oak, hickory, and beech stump sprouts were competitive in year (mean height = m), but the sample was small ( out of total stems measured, fig. a). the low number of sprout origin oaks documented in the cur- rent study was likely due to the advanced age and large size of oaks within pre-harvest stands. the low disturbance intensities and long rotations that currently characterize management in southern indiana may have profound effects on oak regeneration since veg- etative reproduction in the form of stump sprouting in oaks decli- nes as rootstocks continue to age and stems grow in size, thus becoming less likely to produce stump sprouts that are competitive (groninger and long, ; johnson et al., ; weigel and peng, ; weigel et al., ). no new oak seedlings were documented in year old stands, likely due to the relatively poor seed dispersal within the genus and the lack of older, seed producing oaks left within harvest units. the tallest stems of all reproduction in year stands were from yellow-poplar and red maple stump sprouts. repeated measure studies from the central and southern appalachian mountains have shown similar results; stems of black cherry, yellow-poplar, red maple and similar species originating from stump sprouts grow quickly (keyser and zarnoch, ) and remain dominant decades after harvest, outcompeting white, northern red, and chestnut oak (wendel, ; beck and hooper, ). while studies have shown that early seral species such as yellow-poplar and black cherry were common in pre-harvest overstories (wendel, ; beck and hooper, ), the shade tolerance of red maple allowed it to establish in the pre-harvest subcanopy in densities much higher than those of other species, providing numerous sprout origin red maple stems in regenerating stands (wendel, ; gould et al., ). as we hypothesized, by year and the heights of stems in the previous sampling periods were by far the strongest predictors of current height. essentially, the height growth winners in year were already among the tallest stems years after harvest and remained in these positions through the end of the study. this result corroborates research demonstrating that intermediate and codominant oaks should be released with pre-commercial thinning – years post-harvest in order to recruit oak species into the overstory (johnson et al., ; ward, ). because prior height and species were the strongest predictors of future height, site preparatory burns that limit competition and promote larger advance oak reproduction could offer an effective management strategy before overstory removal (arthur et al., ). as observed with reproduction in year , basal diameter of advance reproduc- tion in previous samples was a weak predictor of future height. . . environmental factors contrary to our hypothesis, site factors contributed little to the prediction of height growth in post-harvest stands relative to bio- logical factors. however, in year black cherry, sugar maple, and yellow-poplar displayed greater height on richer soils (high %mg and cec) located on north and east facing slopes (aspect code; fig. a). both advance and post-harvest seedlings of these species were able to establish codominant positions years after harvest, unlike oak advance reproduction, which were largely relegated to intermediate canopy positions. the rapid growth of black cherry, sugar maple, and yellow- poplar on productive sites suggests that they have some affinity for topographic and edaphic characteristics associated with more mesic conditions. even though these species do establish on drier, more nutrient-limited sites (hilt, ; jenkins and parker, ), the vigor of post-harvest stems decreased with decreasing site quality, likely limiting their competitive ability. morrissey et al. ( ) showed that slightly older clearcut stands (cut before ) in our study region maintained higher densities of oak in post-harvest conditions. this was likely due to an exceptional drought that occurred from to , which likely limited growth and caused mortality of seedlings from drought sensitive species such as yellow-poplar and black cherry (morrissey et al., ). seedling in this study developed in relatively drought-free conditions, therefore drought sensitive species remained highly competitive in young stands. . conclusions as observed in other forest systems (bailey and covington, ; guyette et al., ), our study sites experienced historic harvesting and subsequent management practices that led to a simplified disturbance regime that drove composition and struc- ture towards dense stands of shade-tolerant species. as observed in our study and many others (e.g., beck and hooper, ; george and fischer, ; jenkins and parker, ; groninger and long, ), this shift in composition and structure has resulted in the oak recruitment failure that currently plagues much of the chr. in our study, large oak reproduction was rare in pre- harvest stands, allowing more aggressive early seral species, such j. travis swaim et al. / forest ecology and management ( ) – as yellow-poplar, and large advance reproduction of shade-tolerant species, like sugar maple, to dominate new stands. initial height was the best predictor of future height growth in all species, fur- ther highlighting the poor competitive status of small oak repro- duction. oak stump sprouts remained competitive in developing stands, but too few of these stems existed to successfully regener- ate stands, likely due to the advanced age and wide dispersion of overstory stems in pre-harvest stands. in our study, the physical environment, including slope position and aspect, were poor predictors of oak height growth in young, clearcut stands. the poor predictive power of site variables sug- gests that competition effects masked any direct site effects in this study. the productivity of southern indiana forests, and lack of fire or surrogate disturbance to control mesophytic competition, has allowed developing stands to shift away from the oak-dominance that existed prior to harvest. because the tallest stems in -year-old stands were the tallest stems in years and , stands should be cleaned as early as pos- sible to release oak reproduction from competition. alternatives to silvicultural clearcutting should also be considered, such as a two- or three-stage shelterwood combined with pre-harvest prescribed burns. these treatments would likely promote the large oak advance reproduction that is needed to successfully reproduce oak ecosystems in southern indiana. acknowledgments we thank charlie jackson, lindsay jenkins, andy meier, and josh shields for their assistance with fieldwork and analyses. we also thank the personnel of the hoosier national forest for their assistance and logistical support. funding for this study was pro- vided by the usda forest service northern research station and hoosier national forest under joint venture agreement -jv- - . references abrams, m.d., . adaptations and responses to drought in quercus species of north america. tree physiol. , – . arthur, m.a., paratley, r.d., blankenship, b.a., . single and repeated fires affect survival and regeneration of woody and herbaceous species in an oak-pine forest. j. torrey bot. soc. , – . bailey, j.d., covington, w.w., . evaluating ponderosa pine regeneration rates following ecological restoration treatments in northern arizona, usa. for. ecol. manage. , – . bakker, j.p., olff, h.j., willems, j.h., zoebel, m., . why do we need permanent plots in the study of long-term vegetation dynamics? j. veg. sci. , – . beck, d.e., hooper, r.m., . development of a southern appalachian hardwood stand after clearcutting. south. j. appl. for. , – . beers, t.w., dress, p.e., wensel, l.c., . aspect transformation in site productivity research. j. for. , – . belli, k.l., hart, c.p., hodges, j.d., stanturf, j.a., . assessment of the regeneration potential of red oaks and ash on minor bottoms of the mississippi. south. j. appl. for. , – . brashears, m.b., fajvan, m.a., schuler, t.m., . an assessment of canopy stratification and tree species diversity following clearcutting in central appalachian hardwoods. for. sci. , – . bratkovich, s., gallion, j., leatherberry, e., hoover, w., reading, w., durham, g., . forests of indiana: their economic importance. usda forest service, northeastern area state and private forestry, na-tp- - , p. breiman, l., . random forests. mach. learn. , – . brose, p.h., dey, d.c., phillips, r.j., waldrop, t.a., . a meta-analysis of the fire- oak hypothesis: does prescribed burning promote oak reproduction in eastern north america? for. sci. , – . brose, p., schuler, t., van lear, d., berst, j., . bringing fire back: the changing fire regimes of the appalachian mixed-oak forests. j. for. , – . carmean, w.h., . site index curves for black, white, scarlet, and chestnut oaks in the central states. usda forest service, north central experiment station, st. paul, mn, research paper nc- . carmean, w.h., hahn, j.t., jacobs, r.d., . site index curves for forest tree species in the eastern united states. usda forest service general technical report nc- . clark, f.b., watt, r.f., . silvicultural methods for regenerating oaks. in: white, d.e., roach, b.a. (eds.), the silviculture of oaks and associated species. usda forest service, northeastern forest experiment station, upper darby, penn., pp. – . cutler, d.r., edwards jr., t.c., beard, k.h., cutler, a., hess, k.t., gibson, j., lawler, j.j., . random forests for classification in ecology. ecology , – . dey, d.c., . the ecological basis for oak silviculture in eastern north america. in: mcshea, w.j., healy, w.m. (eds.), oak forest ecosystems: ecology and management for wildlife. johns hopkins university press, baltimore, pp. – . dey, d.c., parker, w.c., . morphological indicators of stock quality and field performance of red oak (quercus rubra l.) seedlings underplanted in a central ontario shelterwood. new for. , – . dey, d.c., spetich, m.a., wiegel, d.r., johnson, p.s., graney, d.l., kabrick, j.m., . a suggested approach for design of oak (quercus l.) regeneration research considering regional differences. new for. , – . ellison, a.m., banks, m.s., clinton, b.d., colburn, e.a., elliot, k., ford, c.r., foster, d.r., kloeppel, b.d., knoepp, j.d., lovett, g.m., mohan, j., orwig, d.a., rodenhouse, n. l., sobczak, w.v., stinson, k.a., stone, j.k., swan, c.m., thompson, j., houe, b.v., webster, j.r., . loss of foundation species: consequences for the structure and dynamics of forested ecosystems. front. ecol. environ. , – . george, d.w., fischer, b.c., . the occurrence of oak reproduction after clearcut harvesting on the hoosier national forest. north. j. appl. for. , – . gould, p.j., steiner, k.c., finley, j.c., mcdill, m.e., . regenerating mixed-oak stands in pennsylvania: a quarter century retrospective. in: van sambeek, j.w., dawson, j.o., ponder jr. f., loewenstein, e.f., fralish, j.s. (eds.). proc. of the th central hardwood forest conf. us for. serv. gen. tech. rep. nc- . pp. – . groninger, j.w., long, m.a., . oak ecosystem management considerations for central hardwoods stands arising from silvicultural clearcutting. north. j. appl. for. , – . guyette, r.p., muzika, r., voelker, s.l., . the historical ecology of fire, climate, and the decline of shortleaf pine in the missouri ozarks. in: kabrick, j.m., dey, d. c., gwaze, d. (eds.), shortleaf pine restoration and ecology in the ozarks: proceedings of a symposium. general technical report nrs-p- . usda forest service, northern research station, newton square, pa, pp. – . hilt, d.e., . species composition of young central hardwood stands that developed after clearcutting. in: dawson, j.o., majerus, k.a. (eds.), proceedings of the th central hardwood forest conference, – april , champaign, illinois. department of forestry, university of illinois urbana, champaign, ill., pp. – . homoya, m.a., abrell, d.b., aldrich, j.r., post, t.w., . the natural regions of indiana. proc. indiana acad. sci. , – . hothorn, t., bühlmann, p., dudoit, s., molinaro, a., van der laan, m.j., a. survival ensembles. biostatistics , – . hothorn, t., hornik, k., zeileis, a., b. unbiased recursive partitioning: a conditional inference framework. j. comput. graph. stat. , – . jenkins, m.a., parker, g.r., . composition and diversity of woody vegetation in silvicultural openings of southern indiana forests. for. ecol. manage. , – . johnson, p.s., shifley, s.r., rogers, r., . the ecology and silviculture of oaks. cabi publishing, new york. kabrick, j.m., zenner, e.k., dey, d.c., gwaze, d., jensen, r.g., . using ecological land types to examine landscape-scale oak regeneration dynamics. for. ecol. manage. , – . kabrick, j.m., villwock, j.l., dey, d.c., keyser, t.l., larsen, d.r., . modeling and mapping oak advance reproduction density using soil and site variables. for. sci. , – . keyser, t.l., zarnoch, s.j., . stump sprout dynamics in response to reductions in stand density for nine upland hardwood species in the southern appalachian mountains. for. ecol. manage. , – . larsen, d.r., johnson, p.s., . linking the ecology of natural oak regeneration to silviculture. for. ecol. manage. , – . liaw, a., wiener, m., . classification and regression by randomforest. r news , – . lhotka, j.m., . effect of gap size on mid-rotation stand structure and species composition in a naturally regenerated mixed broadleaf forest. new for. , – . mcshea, w.j., healy, w.m., devers, p., fearer, t., koch, f.h., stauffer, d., waldon, j., . forestry matters: decline of oaks will impact wildlife in hardwood forests. j. wildlife manage. , – . morrissey, r.c., jacobs, d.f., seifert, j.r., fischer, b.c., kershaw, j.a., . competitive success of natural oak regeneration in clearcuts during the stem exclusion stage. can. j. for. res. , – . morrissey, r.c., jacobs, d.f., seifert, j.r., kershaw, j.a., . overstory species composition of naturally regenerated clearcuts in an ecological classification framework. plant ecol. , – . pallardy, s.g., rhoads, j.l., . morphological adaptations to drought in seedlings of deciduous angiosperms. can. j. for. res. , – . parker, w.c., dey, d.c., . influence of overstory density on ecophysiology of red oak (quercus rubra) and sugar maple (acer saccharum) seedlings in central ontario shelterwoods. tree physiol. , – . pickett, s.t.a., . space-for-time substitution as an alternative to long-term studies. in: likens, g.e. (ed.), long-term studies in ecology: approaches and alternatives. springer-verlag, pp. – . r development core team, . r: a language and environment for statistical computing, version . . . r foundation for statistical computing, vienna, austria. http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h j. travis swaim et al. / forest ecology and management ( ) – roach, b.a., gingrich, s.f., . even-aged silviculture for upland central hardwoods. forest service, usda forest service, agricultural handbook. p. sander, i.l., . height growth of new oak sprouts depends on size of advance reproduction. j. for. , – . sander, i.l., clark, f.b., . reproduction of upland hardwood forests in the central states. in: agriculture handbook no. . u.s. department of agriculture, forest service. sander, i.l., graney, d.l., . regenerating oaks in the central states. in: oak regeneration: serious problems practical recommendations. usda forest service gtr se- , pp. – . smith, d.w., . the southern appalachian hardwood region. in: jbarrett, j.w. (ed.), regional silviculture of the united states, rd ed. wiley, new york, pp. – . standiford, r.b., fischer, b.c., . fifteen year results of three harvesting methods on composition and development of regeneration in southern indiana upland hardwoods. in: garrett, h.e., cox, g.s. (eds.), proceedings of the rd central hardwood forest conference, – september , columbia, missouri. university of missouri press, columbia, missouri, pp. – . strobl, c., boulesteix, a.l., kneib, t., augustin, t., zeileis, a., . conditional variable importance for random forests. bmc bioinformatics , . strobl, c., boulesteix, a.l., zeileis, a., hothorn, t., . bias in random forest variable importance measures: illustrations, sources and a solution. bmc bioinformatics , . swaim, j.t., . stand development and the competitive ability of oak (quercus spp.) following silvicultural clearcutting on the hoosier national forest. m.s. thesis. purdue university, west lafayette, in, p. thompson, j.r., spies, t.a., . vegetation and weather explain variation in crown damage within a large mixed-severity wildfire. for. ecol. manage. , – . walker, l.c., . the north american forests: geography, ecology, and silviculture. crc press, new york, p. ward, j.s., . intensity of precommercial crop tree release increases diameter growth and survival of upland oaks. can. j. for. res. , – . weigel, d.r., peng, c.y.j., . predicting stump sprouting and competitive success of five oak species in southern indiana. can. j. for. res. , – . weigel, d.r., dey, d.c., peng, c.y.j., . stump sprout dominance probabilities of five oak species in southern indiana years after clearcut harvesting. in: proceedings of the th central hardwood forest conference gtr-nrs-p, p. . wendel, g., . stump sprout growth and quality of several appalachian hardwood species after clearcutting. usda forest service research paper, northeastern forest experiment station. zhang, h., singer, b.h., . recursive partitioning and applications, nd ed. springer, new york, ny, p. http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h 	predicting the height growth of oak species (quercus) reproduction over a -year period following clearcutting 	 introduction 	 methods 	 . study sites 	 . plot establishment 	 . plot sampling 	 . data preparation 	 . stand density analysis 	 . random forest analysis 	 results 	 . stand density 	 . rf models 	 . . year height of all reproduction 	 . . year height of reproduction originating as advance reproduction 	 . . year height of all reproduction 	 . . year height of all reproduction 	 discussion 	 . advance reproduction 	 . other regeneration types 	 . environmental factors 	 conclusions 	acknowledgments 	references crop science, vol. , july– august www.crops.org research genotype ´ environment interaction and stability analysis for watermelon fruit yield in the united states mahendra dia, todd c. wehner,* richard hassell, daniel s. price, george e. boyhan, stephen olson, stephen king, angela r. davis, and gregory e. tolla abstract one of the major breeding objectives for water- melon (citrullus lanatus [thumb.] matsum & nakai) is improved fruit yield. high yielding genotypes have been identified, so we mea- sured their stability for fruit yield and yield components over diverse environments. the objectives of this study were to (i) evaluate the yield of watermelon genotypes over years and locations, (ii) identify genotypes with high sta- bility for yield, and (iii) measure the correlations among univariate and multivariate stability sta- tistics. a diverse set of genotypes was evalu- ated over yr ( , , and ) and eight locations across the southern united states in replicated trials. yield traits were evaluated over multiple harvests, and measured as marketable yield, fruit count, percentage cull fruit, percent- age early fruit, and fruit size. there were strong effects of environment as well as genotype ´ environment interaction (g´e) on watermelon yield traits. based on multiple stability mea- sures, genotypes were classified as stable or unstable for yield. there was an advantage of hybrids over inbreds for yield components in both performance and responsiveness to favor- able environments. cultivars big crimson and legacy are inbred lines with high yield and sta- bility. a significant (p < . ) and positive cor- relation was measured for shukla’s stability vari- ance (si ), shukla’s squared hat (ŝi ), wricke’s ecovalence (wi), and deviation from regression (s d) for all the traits evaluated in this study. m. dia and t.c. wehner, dep. of horticultural science, north carolina state univ., campus box , raleigh, nc - ; r. hassell, clemson university, coastal research and education center, savan- nah hwy, charleston, sc ; d.s. price, georgia county extension, sw district, west th ave, suite c, cordele, ga ; g.e. boy- han, university of georgia, dep. of hort., miller plant science bldg., athens, ga ; s. olson, north florida rec, univ. of florida, research road, quincy, fl - ; s. king, texas a&m university, dep. of hort. sci., research pkwy, ste a , college station, tx ; a.r. davis, usda-ars, east highway , lane, ok ; g.e. tolla, monsanto/seminis veg seeds, state hwy , woodland, ca . received oct. . accepted feb. . *corresponding author (tcwehner@gmail.com). abbreviations: aec, average environment coordinate; ammi, additive main effects and multiplicative interaction; b i , linear regression coefficient; ci, clinton, nc; fl, quincy, fl; ga, cordele, ga; gge, genotype main effects plus genotypic ´ environment interaction effect; ggl, genotype main effects plus genotypic ´ location interaction effect; g´e, genotype ´ environment interaction; g or , au-jubilant; g or , allsweet; g or , big crimson; g or , black diamond; g or , calhoun gray; g or , calsweet; g or , carolina cross# ; g or , charleston gray; g or , congo; g or , crimson sweet; g or , desert king; g or , early arizona; g or , early canada; g or , fiesta f ; g or , georgia rattlesnake; g or , golden midget; g or , graybelle; g or , hopi red flesh; g or , jubilee; g or , king & queen; g or , legacy; g or , mickylee; g or , minilee; g or , mountain hoosier; g or , nc giant; g or , navajo sweet; g or , peacock wr- ; g or , quetzali; g or , regency f ; g or , royal flush f ; g or , sangria f ; g or , starbrite f ; g or , stars-n-stripes f ; g or , stone mountain; g or , sugar baby; g or , sugarlee; g or , sweet princess; g or , tendersweet of; g or , tom watson; g or , yellow crimson; kn, kinston, nc; m, trait mean; ok, lane, ok; pc, principal component; sc, charleston, sc; svp, singular value partitioning; tx, college station, tx; ca, woodland, ca; s i , shukla’s stability variance; ŝ i , shukla’s squared hat; s d , deviation from regression; w i , wricke’s ecovalence; ys i , kang’s stability statistic published in crop sci. : – ( ). doi: . /cropsci . . © crop science society of america | guilford rd., madison, wi usa all rights reserved. published june , www.crops.org crop science, vol. , july– august elite cultivars of watermelon have been developed with high fruit yield, fruit quality, earliness, percent- age marketable fruit, excellent shipping characteristics, and disease resistance. a century of breeding has produced uniform hybrids, seedless triploids, tough rind, high sugar content, dark red flesh, kg picnic watermelon, and kg mini watermelon (gusmini and wehner, ). since , yield has increased approximately % in the united states (usda, ). however, high yield is often associated with decreased yield stability (calderini and slafer, ; padi, ). yield stability is important, but limited studies have been done on watermelon (vasanthkumar et al., ). genotypes with high yield and stability can be identi- fied using trials in multiple years and locations (lu’quez et al., ). genotypes respond differently to environ- mental factors such as soil fertility or the presence of dis- ease pathogens (kang, ). these contribute to geno- type ´ environment interaction (g´e), which has been reported in field and vegetable crops (bednarz et al., ; mekbib, ; riday and brummer, ; fan et al., ; mulema et al., ; miranda et al., ; de vita et al., ; panthee et al., ; rak et al., ). the presence of g´e makes it useful to measure both performance and stability for genotypes being evaluated in breeding programs (magari and kang, ; ebdon and gauch, ). genotype ´ environment interaction may result in low correlation between phenotypic and genotypic values, thereby reducing progress from selection. this leads to bias in the estimation of heritability and in the prediction of genetic advance (comstock and moll, ; alghamdi, ). therefore, the magnitude and nature of g´e deter- mines the features of a selection and testing program. several statistical methods for evaluating stability have been proposed, reflecting different aspects of the g´e. these statistical methods range include univariate models, such as regression slope, deviation from regression, envi- ronmental variance, and kang’s yield-stability, and mul- tivariate models, such as genotype main effect plus geno- type ´ environment interaction (gge) biplot, and addi- tive main effects and multiplicative interaction (ammi) (finlay and wilkinson, ; eberhart and russell, ; yan, ; kang, ; yan and kang, ). however, no single method adequately explains genotype performance across environments. stability statistics (variation) are best used in combination with trait performance (mean). analysis of variance is often used to identify the existence of g´e in multiple-environment trial data. analysis of vari- ance measures the components of variance arising from dif- ferent fixed and random factors (e.g., genotype, location, year, and replication) and their interactions. however, anova has limitations, including the assumption of homogeneity of variance among environments, in its ability to explore the response of genotypes for g´e (zobel et al., ). the most widely used approach for stability analy- sis is based on linear regression: the slope (b i ) or deviation from regression (s d )] of genotype performance relative to an environmental index derived from the average perfor- mance of all genotypes in each environment (finlay and wilkinson, ; eberhart and russell, ; freeman, ; chakroun et al., ). some researchers have found deficiencies in the regression method for evaluation of g´e patterns (zobel et al., ; nachit et al., ; annicchi- arico, ; kandus et al., ; de vita et al., ). the deficiencies are of four types. first, the estimates of best fitted line have high error when only a few low- and high- yielding locations are included in the study (crossa et al., ). second, the average of all genotypes evaluated in each environment (environmental index) is not indepen- dent of each genotype for that environment (freeman and perkins, ). third, the errors associated with the slopes of genotypes are not statistically independent (kandus et al., ). fourth, there is a required assumption of a linear relationship between interaction and environmental means when the actual responses of the genotypes to the environ- ments are intrinsically multivariate (crossa et al., ). shukla ( ) proposed an unbiased estimate of the variance (s i ) of g´e plus an error term associated with genotype, in which a genotype with low s i is regarded as stable. shukla’s stability variance is a linear combination of wricke’s ecovalance (w i ), which is the proportion of g´e variance contributed by each genotype. w i and s i are equivalent in ranking genotypes for stability (kang et al., ). kang’s stability statistic (ys i ) is nonparametric, using both trait mean (m ) and s i , with equal weight on each. genotypes with ys i greater than the mean ys i are stable (kang, ; mekbib, ; fan et al., ). multivariate analysis includes the ammi method, and the gge method with a graphical display (casanoves et al., ; dehghani et al., ). these models are based on principal component (pc) analysis and have the ability to reveal structure in the data. the ammi and gge biplot differ in value for analyzing multi-environ- ment trial data (gauch, ; yan et al., ). the gge biplot was named by yan et al. ( ). it is constructed from the first two principal components (pc and pc ) that explain maximum variability in the data, derived by singular value decomposition of a two-way (genotype ´ environment) data matrix (yan et al., ). the ammi model combines the anova (an addi- tive model) to characterize genotype and environment main effects, with pc analysis (a multiplicative model) to characterize interactions (crossa et al., ). depend- ing on number (n) of pc’s used in study the interaction, ammi models are usually called ammi , ammi , … ammi(n). the ammi biplot separates genotypes accord- ing to their pc scores, making it easy to determine geno- type stability (carbonell et al., ). crop science, vol. , july– august www.crops.org weight divided by total fruit weight. culls included crooked, bottle-necked, and other deformed fruit. one to four harvests were made depending on location and year. most locations had two to three harvests. harvest data were not collected from oklahoma in , georgia in , and florida in . data were not collected on percentage cull fruit from south carolina in , , and ; or from florida in and . a single harvest was made in california in and in georgia in (no data on percentage early fruit for those environments). data analysis data were analyzed for genotype, environment and g´e using the sasgxe program developed by dia and wehner ( ) and sas v . (sas, ). sasgxe computes univariate stabil- ity statistics, input files that are ready to use in r packages for multivariate stability statistics, anova, descriptive statistics, and correlation among stability analysis methods (dia and wehner, ). sasgxe is available at http://cuke.hort.ncsu.edu/cucurbit/ wehner/software.html. years, locations, replications, and geno- types were analyzed as random effects. aanalysis of variance was used to determine the size and significance (f ratio) of genotype ´ environment interactions for the traits of interest. if genotype ´ environment interactions were significant, additional statistics were calculated to determine the stability of each genotype over the environments (location ´ year combinations). the univariate stability parameters used were shukla’s stability variance (s i ), shukla’s squared hat (ŝ i ), wricke’s eco- valence (w i ), regression slope (b i ), deviation from regression (s d ) and kang’s yield-stability statistics (ys i ). least squared means or adjusted trait means (m ) and their lsd for each geno- type were computed over the yr and eight locations for the traits of interest. hereafter, mean will indicate least squared mean or adjusted trait mean. the ammi and gge biplots were computed using the ammi (agricolae) and ggebiplot- gui package, respectively, of r statistical software in rstudio (r development core team, ; cran, ; rstudio, ). the ammi and gge biplot analysis were used to visu- ally assess the presence of g´e and rank genotype based on stability and mean (yan et al., ; yan and kang, ). tests for significance were derived using a t test for each b i and an f test for each s d for statistical differences from one and zero, respectively, at . , . , and . levels of probability. ranks were assigned to each genotype in increasing order for each sta- bility parameter, except percentage cull fruit (selected for low values). simple correlation coefficients (using spearman rank) were calculated for all pairs of stability measures. results the results are presented for the magnitude of g´e, the stability of genotypes, and the correlations among stability measures. genotype ´ environment interaction analysis of variance the combined anova indicated highly significant envi- ronment, genotype, and g´e effects for marketable yield, fruit count, percentage cull fruit, percentage early fruit, and fruit size (table ). except for fruit size, all yield traits we measured watermelon yield components includ- ing marketable yield (mg ha– ), fruit count (thousand fruit ha– ), percentage early fruit, percentage cull fruit, and fruit size (kg fruit– ). the objectives of this study were to (i) evaluate the g´e of watermelon genotypes, (ii) iden- tify watermelon genotypes with high stability for yield, and (iii) estimate the correlations among univariate and multivariate stability statistics. materials and methods germplasm and location forty genotypes of watermelon were evaluated for yr ( , , and ) and in eight locations across the southern united states. locations were chosen to represent major water- melon production regions in the united states: north carolina (kn: kinston, ci: clinton) and south carolina (sc: charles- ton) in the east to georgia (ga: cordele), florida (fl: quincy), oklahoma (ok: lane), and texas (tx: college station) in the south to california (ca: woodland) in the west. forty geno- types were chosen to represent new vs. old releases, small vs. large fruit size, round vs. elongate fruit shape, striped vs. solid rind pattern, anthracnose resistance vs. susceptibility, eastern vs. western adapted, and inbred vs. hybrid type (supplemental tables s and s ). the watermelon genotypes were catego- rized as inbred or hybrid based on information obtained from seed providers (gusmini and wehner, , ). hybrids are identified with f after their name. cultural practices the experiment design was a randomized complete block with four replications, eight locations and yr. seeds of each geno- type were sown in -cell polyethylene flats in the greenhouses at north carolina state university. the seedlings were trans- planted by hand at the two-true-leaf stage. missing or damaged transplants were replaced wk later. plots were planted on raised, shaped beds in rows on . -m centers with plants . m apart. the beds had drip irrigation tubes covered with black polyethylene mulch. production prac- tices were according to the north carolina extension service and southeastern u.s. vegetable crops handbook (sand- ers, ; holmes and kemble; ). data collection and traits at each location, the watermelon genotypes were evaluated for traits including marketable yield (mg ha– ), fruit count (thou- sand fruit ha– ), percentage cull fruit ( × cull fruit yield/total fruit yield), percentage early fruit ( × fruit weight of first har- vest/fruit weight over all harvests), and fruit size (kg fruit– ). fruit were harvested using the guide of number of days to maturity, as well as the indicators of maturity: a brown and dry tendril at the node bearing the fruit, a dull waxy fruit surface, a light-colored groundspot on the fruit, and a dull sound of the fruit when thumped (maynard, ). fruit were weighed indi- vidually, and yield was calculated as total and marketable fruit weight (mg ha– ) and number (thousand ha– ) by summing plot yield over harvests. numbers of cull and marketable fruit were also recorded. percentage cull fruit was calculated as cull fruit www.crops.org crop science, vol. , july– august had large variances due to environment (ranged from – %), with large differences among environments for genotype means causing most of the variation in genotype performance (table ). marketable yield, percentage cull, and percentage early had large genotype ´ environment effect ( , , and % of total sum of squares, respectively) and small genotype effect ( , , and % of total variance, respectively) (table ). for fruit count, the environment effect, genotype effect, and genotype ́ environment effect had similar contribution to total variation. in contrast, fruit size had large genotype effect ( % of total variance), small environment effect ( % of total variance) and genotype ´ environment effect ( % of total variance) (table ). polygon view of gge biplot the polygon (which–won–where) view of the gge biplot divides the biplot into sector via perpendicular lines (rays) passing from the polygon sides (fig. ). the poly- gon is drawn by joining extreme genotypes of the biplot. if environments fall into different sectors, then different genotypes won in different sectors, and a crossover g´e pattern exists. the winning genotype for an environment or set of environments in a sector is the vertex genotype. conversely, if all environments fall into a single sector, a single genotype had the highest yield in all environments. the vertex genotype in a sector where no environment is present is considered to be a poor performer in all test environments. genotypes within the polygon were less responsive to location than the vertex genotypes. a poly- gon view of the gge biplot explained , , , , and % of the genotype and genotype ´ environment varia- tion for the marketable yield, fruit count, percentage cull fruit, percentage early fruit, and fruit size, respectively (fig. : panel a, b, c, d, and e). other than fruit count and fruit size, yield traits had environments in two sectors with different wining genotypes (vertex genotype) in each (fig. : panel a, b, c, d, and e). this confirms the existence of g´e for marketable yield, percentage cull fruit and per- centage early fruit (fig. : panel a, c, and d). genotype main effects plus genotype ´ location interaction effect (ggl) biplots for individual year were constructed and showed that location grouping did not vary across years. results of ggl biplots are not presented here. additive main effects and multiplicative interaction in ammi , the biplot abscissa and ordinate used the first and second principal component terms (pc and pc ), respec- tively. the ammi biplot explained , , , , and % of the genotype and genotype ´ environment variation for marketable yield, fruit count, percentage cull fruit, percent- age early fruit, and fruit size, respectively (fig. : panel a, b, c, d, and e). horizontal and vertical lines passing from the origin ( , ) of the biplot divide it into four sectors. like gge biplot, the genotypes at the vertex of the polygon are the winners for the environments included in that sector. a location close to the biplot origin is a less interactive location and is considered to be good for the selection of genotypes with average adaptation (murphy et al., ). the angle between genotype and environment vectors determined the nature of g´e: it is positive for acute angles, negligible for right angles, and negative for obtuse angles. also, the dis- tance of genotype and environment vectors from the biplot origin indicates the magnitude of g´e exhibited by geno- types over environment or environments over genotype. for marketable yield, fruit count and fruit size, environments fell into all four sectors, and different wining genotypes (vertex genotypes) were observed in each sector (fig. : panel a, b, and e). similarly, percentage cull and percentage early had environments falling into three sectors (fig. : panel c and d). thus, ammi biplot validated the existence of interac- tion of watermelon genotypes with environments for table . significance values and variance (mean squares) for marketable yield (mg ha– ), fruit count, percentage cull fruit, percent- age early fruit, and fruit size (kg fruit– ) of watermelon genotypes tested in yr and eight locations in the united states. source df† marketable yield fruit count percentage cull fruit percentage early fruit fruit size environment (e) ** ** ** ** ** location (l) ** ** ns‡ ** ns year (y) ns ns ns ** ns l ´ y ** ** ** ** ** replication within e ** ** ** ** ** genotype (g) ** ** ** ** ** g ´ e ** ** ** ** ** g ´ l ** ** ** ** ** g ´ y ns ns ** ** ** g ´ l ´ y ** ** ** ** ** ** significant at the . level of probability, † note: degrees of freedom (df) were lower for percentage cull fruit and percentage early fruit due to missing data. ‡ ns = nonsignificant. crop science, vol. , july– august www.crops.org fig. . the polygon (which–won–where) view of genotype main effects plus genotypic ´ environment interaction effect (gge) biplot of watermelon genotypes tested in yr and eight locations for (panel a) marketable yield, (panel b) fruit count, (panel c) percentage cull fruit, (panel d) percentage early fruit, and (panel e) fruit size. the biplots were based on scaling = , centering = , and svp = . www.crops.org crop science, vol. , july– august all the traits evaluated. however, ammi exhibited a dif- ferent crossover pattern than the gge biplot (fig. and : panel a, b, c, d, and e). also, ammi for individual year was analyzed for all traits, and some inconsistencies were observed, since locations fell into different sectors. however, the general pattern of location grouping did not vary across years. results of ammi are not presented here. genotype evaluation the significant g´e justified our evaluation of water- melon genotypes for yield stability over environments. genotype means marketable yield ranged from a high of . to a low of . mg ha– . highest marketable yield was recorded for genotypes big crimson, stone mountain, stars-n-stripes f , and starbrite f (table ). other high yielding genotypes fig. . additive main effects and multiplicative interaction (ammi ) biplot of the first two principal component (pc and pc ) showing the genotype ´ environment interaction (g´e) for (panel a) marketable yield, (panel b) fruit count, (panel c) percentage cull fruit, (panel d) percentage early fruit, and (panel e) fruit size of watermelon genotypes tested in yr and eight locations. crop science, vol. , july– august www.crops.org were fiesta f , regency f , calhoun gray, legacy, and mountain hoosier. genotype golden midget had mar- ketable yield significantly lower than the other genotypes (table  ). fruit count ranged from . to . thousand fruit ha– (table ). highest fruit count was for golden midget, followed by genotype minilee and king & queen (table ). lowest fruit count was for genotype carolina cross # , significantly lower than the other genotypes (table ). large fruit size was correlated with high percent- age cull fruit (table ). genotype nc giant and congo had large fruit and the highest percentage cull fruit ( . and . %, respectively, table ). lowest percentage cull fruit were for minilee, which was similar to high yielding genotypes starbrite f and regency f (table ). geno- types carolina cross# , nc giant, georgia rattle- snake, au-jubilant, and jubilee had the largest fruit size; genotypes golden midget, minilee, and mickylee had the smallest fruit size (table ). genotypes golden midget, early canada, stone mountain, and regency f had the highest percentage early fruit (table ). genotypes navajo sweet, peacock wr- , king & queen, minilee, and tom watson produced the lowest percentage early fruit (table ). regression coefficient or slope according to eberhart and russell ( ), a regression coefficent (b i ) approximating unity along with s d near zero indicate stability. for marketable yield, the b i value for many genotypes was close (p > . ) to unity, except for genotypes big crimson, carolina cross# , charleston gray, congo, early canada, golden midget, graybelle, king & queen, royal flush f , starbrite f , and tom watson (tables and ). similarly, b i for fruit count was close to unity for genotypes sugar baby, stars-n-stripes f , tendersweet of, crimson sweet, fiesta f , desert king, mountain hoosier, legacy, and peacock wr- (table ). genotypes black diamond, congo, and star- brite f had b i significantly different from unity (table ). black diamond and congo had b i of zero. for percentage cull fruit, b i ranged from – . to . (data not shown). genotypes such as congo and hopi red flesh with high percentage cull fruit had high b i . con- versely, genotypes such as stars-n-stripes f , fiesta f , peacock wr , and yellow crimson with low percentage cull fruit were resistant to environmental effects. genotypes minilee, navajo sweet, and calsweet had low percentage cull fruit and negative b i value. the b i value for percentage early fruit for all the genotypes was positive and close to . , except genotype georgia rattlesnake (table ). deviation from regression and shukla’s stability variance the genotypes with the highest marketable yield in this study were starbrite f , stars-n-stripes f , fiesta f , regency f , big crimson, stone mountain, calhoun gray, and legacy. among these high yielding genotypes, two inbreds (big crimson and legacy) and a hybrid (starbrite f ) had signifi- cant s d and high s i for marketable yield (table ). similarly, two high yielding inbreds (big crimson and legacy) had significant s d and high s i for yield components, including fruit count and percentage early fruit (table ). for percent- age early fruit, except two hybrids (fiesta f and starbrite f ), all high yielding genotypes had nonsignificant s d and table . means (corrected by least squares) (m) for marketable yield (mg ha– ), fruit count (thousand fruit ha– ), percentage cull fruit, percentage early fruit and fruit size (kg fruit– ) of watermelon genotypes tested in yr and eight locations (in order by marketable yield). genotype market- able yield† fruit count percent- age cull fruit percent- age early fruit fruit size big crimson . . . . . starbrite f . . . . . stone mountain . . . . . stars-n-stripes f . . . . . fiesta f . . . . . regency f . . . . . calhoun gray . . . . . legacy . . . . . mountain hoosier . . . . . yellow crimson . . . . . desert king . . . . . au-jubilant . . . . . black diamond . . . . . sangria f . . . . . royal flush f . . . . . calsweet . . . . . tendersweet of . . . . . nc giant . . . . . king & queen . . . . . sweet princess . . . . . charleston gray . . . . . tom watson . . . . . georgia rattlesnake . . . . . navajo sweet . . . . . jubilee . . . . . sugarlee . . . . . early arizona . . . . . carolina cross# . . . . . allsweet . . . . . peacock wr- . . . . . hopi red flesh . . . . . congo . . . . . quetzali . . . . . mickylee . . . . . crimson sweet . . . . . minilee . . . . . early canada . . . . . graybelle . . . . . sugar baby . . . . . golden midget . . . . . † lsd of marketable yield, fruit count, percentage cull fruit, percentage early fruit and fruit size is . , . , . , . , and . , respectively. www.crops.org crop science, vol. , july– august low s i (table ). in contrast, for fruit size all high yielding genotypes had nonsignificant s d and low s i , except inbred calhoun gray (table ). hybrids with high marketable yield consistently had nonsignificant s d and low s i for fruit count and percent- age early fruit (table ). the top three yielding hybrids (starbrite f , regency f , and stars-n-stripes f ) had nonsignificant s d and low s i for fruit count and percent- age cull fruit (table ). on the other hand, high yield- ing inbreds had significant s d and high s i for both fruit count and percentage cull fruit (table ). table . significance value of regression coefficient (bi), deviation from regression (s d), shukla’s stability variance (si ), and kang stability statistic (ysi) for marketable yield (mg ha – ), fruit count (thousand fruit ha– ), percentage cull fruit, percentage early fruit and fruit size (kg fruit– ) of watermelon genotypes tested in yr and eight locations. genotype marketable yield fruit count percentage cull fruit percentage early fruit fruit size bi s d si ysi bi s d si ysi bi s d si ysi bi s d si ysi bi s d si ysi au-jubilant . . * Ö‡ . . . . . . . . . . Ö . . . Ö allsweet . . . . . . . . * . . . . . ** . . . . Ö big crimson * ** ** Ö . ** . Ö . *** * Ö . * . . . . Ö black diamond . * ** Ö ** ** . Ö . . . . . * ** Ö . . . Ö calhoun gray . . . Ö . . . Ö * * . . . ** ** Ö . ** . Ö calsweet . ** ** Ö . . . . **† . . Ö . * ** . . *** . Ö carolina cross# ** * * . . ** . . . . ** . . ** ** . *** *** ** Ö charleston gray * *** ** . . *** . . . . . . . . . Ö . * . Ö congo * . . . **† . . . * . ** . . . . Ö . . . . crimson sweet . *** ** . . *** . . . . . Ö . . ** Ö **† * * . desert king . . ** Ö . . . Ö . . . Ö . ** ** . . . . . early arizona . . . . . ** . Ö . . . . . . ** . . * . . early canada * ** . . . . . Ö . . . Ö . . ** Ö . *** . . fiesta f . . * Ö . ** . Ö * * . . . . . Ö . . . Ö georgia rattlesnake . . . . . * . . . . ** . * . . Ö . . . Ö golden midget ** *** ** . . *** ** Ö . *** ** Ö . *** ** Ö *** . . . graybelle * . . . . . . Ö ***† ** . Ö . . . Ö * . . . hopi red flesh . * * . . *** . Ö * *** ** . . . . . . *** . . jubilee . . . . . . . . . * ** . . . ** Ö * . . Ö king & queen **† ** ** . . *** ** Ö *** *** . Ö . * . . * *** . . legacy . *** ** Ö . *** . . **† . . Ö . *** ** Ö . . . Ö mickylee . *** * . . ** . Ö ***† *** . Ö . ** ** . *** ** . . minilee * * . . . . . Ö *** ** . Ö . ** ** . * * . . mountain hoosier . *** ** Ö . *** . . . * . . . . . . . . . Ö nc giant . *** ** . . *** . Ö . ** ** Ö . . * . . ** ** . navajo sweet . *** ** . . *** . . **† *** . . . * * Ö * . . Ö peacock wr- . . . . . * . . * . . Ö . . . . . * . . quetzali . . . . . . . Ö ***† . . Ö . ** ** . . *** . . regency f . . . Ö . . . Ö . . . Ö . * * Ö . . . Ö royal flush f ** * . Ö . . . Ö ** * . Ö . * . . . . . . sangria f . . . Ö . . . Ö ** . . . . ** * . . . . Ö starbrite f ** ** ** Ö * . . Ö * . . Ö . . . Ö . . . Ö stars-n-stripes f . . . Ö . . . Ö . . . Ö . *** ** Ö . . . Ö stone mountain . . ** Ö . . . Ö ** ** . . . ** ** Ö . . . Ö sugar baby . ** . . * . Ö ** ** . Ö . . ** Ö ** *** . . sugarlee . . . . . . . . . . Ö . . . Ö . * . . sweet princess . *** ** . . . . . . . . . . . * Ö . . . Ö tendersweet of . * * Ö . . . . * . * . . . ** . . . . Ö tom watson * . ** . . * . . . * . . . * ** . . ** . . yellow crimson . *** ** Ö . *** . . ** ** . Ö . . . Ö . . . . * significantly different from unity for the regression coefficients or slope (bi) and from zero for the deviation from regression (s d) at . level of probability. ** significantly different from unity for the regression coefficients or slope (bi) and from zero for the deviation from regression (s d) at . level of probability. *** significantly different from unity for the regression coefficients or slope (bi) and from zero for the deviation from regression (s d) at . level of probability. † indicates negative slope. ‡ Ö indicate stable according to kang stability statistic (ysi). crop science, vol. , july– august www.crops.org kang’s stability statistics according to kang’s stability statistics (ys i ), genotypes with ys i higher than the mean ys i are stable. for a trait where lower value is desired (e.g., percentage cull fruit) ys i lower than the mean ys i are stable. the mean ys i for market- able yield, fruit count, percentage cull fruit, percentage early fruit, and fruit size was . , . , . , . , and . , respectively. according to ys i , the top five genotypes with highest marketable yield and stable across all the evaluated traits were stone mountain, stars-n-stripes f , calhoun gray, big crimson, and regency (table ). mean vs. stability and genotype comparison with ideal genotype views of gge biplot the average environment coordinate (aec) view based on genotype-focused singular value partitioning (svp = ) can be referred as the “mean vs. stability” view of gge biplot (yan et al., ). that view facilitates genotype comparisons based on mean performance and stability across environments within a mega-environment. the mean vs. stability view of gge biplot explained , , , , and % of genotypic and genotype ´ environ- ment variation for the marketable yield, fruit count, percentage cull fruit, percentage early fruit, and fruit size, respectively (fig. : panel a, b, c, d, and e). the arrow shown on the aec abscissa points in the direction of higher trait performance of genotypes and ranks the genotypes with respect to trait performance. thus, geno- type starbrite f (g ) had the highest marketable yield and golden midget (g ) had the lowest (fig. : panel a). similarly, genotypes golden midget (g ), nc giant (g ), golden midget (g ), and carolina cross # (g ) had the highest fruit count, percentage cull, percent- age early and fruit size, respectively. genotype mickylee (g ) had the lowest percentage cull, genotype carolina cross # (g ) had the lowest fruit count and percent- age early and golden midget (g ) had the lowest fruit weight (fig. : panel b, c, d, and e). the stability of each genotype was explored by its projection onto the aec vertical axis. the most stable genotype was located almost on the aec abscissa (horizontal axis) and had a near-zero projection onto the aec (vertical axis). thus, genotypes fiesta f (g ), au-jubilant (g ), tom watson (g ), and carolina cross # (g ) were the most stable and yellow crimson (g ), and king & queen (g ) were the least stable for marketable yield (fig. : panel a). according to yan and tinker ( ), stability is meaningful only when associated with high trait mean. therefore, an ideal genotype has both high trait mean and stable performance. an ideal genotype is represented by a circle on the head of arrow on the aec abscissa (horizon- tal axis) (fig. : panel a, b, c, d, and e). for marketable yield, genotypes fiesta f (g ), stars-n-stripes f (g ), regency f (g ), black diamond (g ), and au-jubilant table . means (corrected by least squares) (m), regression coefficient (bi), deviation from regression (s d), shukla’s sta- bility variance (si ), and kang stability statistic (ysi) for mar- ketable yield of watermelon genotypes tested in yr and eight locations. genotype marketable yield m† bi s d si ysi —————————— mg ha– —————————— au-jubilant . . . . allsweet . . . . big crimson . . * . ** . black diamond . . . * . calhoun gray . . . . calsweet . . . ** . carolina cross # . . ** . * . charleston gray . . * . *** . congo . . * . . crimson sweet . . . *** . – desert king . . . . early arizona . . . * . early canada . . * . ** . fiesta f . . . . georgia rattlesnake . . . . golden midget . . ** . *** . – graybelle . . * . . hopi red flesh . . . * . jubilee . . . . king & queen . - . ** . ** . legacy . . . *** . mickylee . . . *** . minilee . . * . * . mountain hoosier . . . *** . nc giant . . . *** . navajo sweet . . . *** . peacock wr- . . . . quetzali . . . . regency f . . . . royal flush f . . ** . * . sangria f . . . . starbrite f . . ** . ** . stars-n-stripes f . . . . stone mountain . . . . sugar baby . . . ** . – sugarlee . . . . sweet princess . . . *** . tendersweet of . . . * . tom watson . . * . . yellow crimson . . . *** . * significantly different from unity for the regression coefficients or slope (bi) and from zero for the deviation from regression (s d) at . level of probability. ** significantly different from unity for the regression coefficients or slope (bi) and from zero for the deviation from regression (s d) at . level of probability. *** significantly different from unity for the regression coefficients or slope (bi) and from zero for the deviation from regression (s d) at . level of probability. † lsd of marketable yield is . . www.crops.org crop science, vol. , july– august fig. . the mean vs. stability view of genotype main effects plus genotypic ´ environment interaction effect (gge) biplot of water- melon genotypes tested in yr and eight locations for (panel a) marketable yield, (panel b) fruit count, (panel c) percentage cull fruit, (panel d) percentage early fruit, and (panel e) fruit size. the biplots were based on scaling = , centering = , and svp = . the ideal genotype is represented by a circle on average environment coordinate (aec)-abscissa which passed through biplot origin. crop science, vol. , july– august www.crops.org (g ) could be regarded as the best genotypes (fig. : panel a). similarly, for fruit count genotypes minilee (g ), golden midget (g ), early arizona (g ), and sugar baby (g ) were best (fig. : panel b). for percentqage cull genotypes early canada (g ), starbrite f (g ), big crimson (g ), and calsweet (g ) were best (fig. : panel c). for percentage early genotypes sweet princess (g ) and legacy (g ) were best (fig. : panel d). for fruit weight genotypes au-jubilant (g ), nc giant (g ), and georgia rattlesnake (g ) were best (fig. : panel e). the “comparison with ideal genotype” view of gge biplot has concentric circles with the ideal genotype in the inner circle and the head of the arrow is the center of the circle (the arrow is highlighted) (fig. : panel a, b, c, and d). the genotypes grouped in the inner circle (ideal geno- types) are more desirable than the others. thus, golden midget (g ) and minilee (g ) were the most desirable genotypes for fruit count (fig. : panel b). however, for marketable yield, percentage early and fruit size, no geno- types were in the inner circle (fig. : panel a, c, and d). therefore, genotypes next to the ideal circle were desir- able. genotypes stars-n-stripes f (g ); golden midget (g ); and nc giant (g ) with carolina cross # (g ) were desirable genotypes for marketable yield, percentage early and fruit size, respectively (fig. : panel a, c, and d). fig. . the genotypes comparison with ideal genotype view of genotype main effects plus genotypic ́ environment interaction effect (gge) biplot of watermelon genotypes tested in yr and eight locations for (panel a) marketable yield, (panel b) fruit count, (panel c) percentage early fruit, and (panel d) fruit size. the biplots were based on scaling = , centering = , and svp = . an ideal genotype is represented by circle within innermost concentric circles on average environment coordinate (aec)-abscissa which passed through biplot origin. www.crops.org crop science, vol. , july– august additive main effects and multiplicative interaction in ammi , the biplot abscissa and ordinate show the trait main effect and first principal component (pc ) term, respectively. the genotypes with pc scores close to zero indicate general adaptation across environments, whereas larger pc scores indicate specific adaptation of genotypes to the environment having the same pc score and sign (murphy et al., ). thus, genotypes regency f ( ), tendersweet of ( ), stars-n-stripes f ( ), starbrite f ( ), stone mountain ( ), big crimson ( ), au-jubi- lant ( ), and calsweet ( ) were stable across environments for marketable yield (fig. : panel a). plant breeders are interested in genotypes having high and stable yield: those with a pc score close to zero and a high trait mean. thus, for marketable yield genotypes peacock wr- ( ), mickylee ( ), calsweet ( ), minilee ( ), and quet- zali ( ) were best (fig. : panel b). for other traits, the best genotypes were golden midget ( ), georgia rattle- snake ( ) and regency f ( ) for fruit count; georgia rattlesnake ( ), golden midget ( ), early canada ( ), allsweet ( ), and regency f ( ) for percentage cull; carolina cross # ( ), peacock wr- ( ), black diamond ( ) and tendersweet of ( ) for percentage early; and hopi red flesh ( ), au-jubilant ( ), carolina cross # ( ), desert king ( ) and tendersweet of ( ) for fruit size (fig. : panel b, c, d, and e). rank correlations for stability measures spearman’s rank correlation spearman rank correlations were computed between the genotype ranks of all evaluated trait means and stability measures. stability measures include s i , shukla’s squared hat (ŝ ii ), w i , b i , deviation from regression (s d ), gge (mean vs. stability view) and ammi (mean vs. stabil- ity view). in mean vs. stability view of gge and ammi biplot for fruit size, the majority of the genotypes were clustered near the biplot origin ( , ). therefore, correla- tion for fruit size could not be computed between gge and ammi and the other stability measures. trait mean was significantly (p < . ) and posi- tively correlated (spearman) with gge and ys i for all traits evaluated in this study (table ). results for ys i are not presented in table . similarly, significant cor- relations were measured between s i with ŝ i , w i and s d for all the traits evaluated in this study; and among all stability measures (except, ammi) for percentage cull fruit (table ). however, nonsignificant correlation was measured between some of the stability statistics (b i , s d , s i , gge, and ammi) with trait mean for marketable yield, fruit count, percentage early and fruit size (table ). thus, stability statistics provide information that cannot be obtained from trait mean (mekbib, ). discussion for all the yield traits evaluated in this study, environ- ment explained most of the variation, and genotype and g´e were small (table ). the partitioning of variance components for environment showed that both predict- able (location) and unpredictable (year and location ´ year) components were important. for marketable yield, fruit count and percentage early, location effects were sig- nificant, suggesting that plant breeders can either develop specialist genotypes for selected environments or general- ist genotypes adapted to a wide range of environments. since location ´ year was significant for all traits evalu- ated, plant breeders should develop stable genotypes that perform well over environments. the ideal genotype should have high mean and high stability. when the genotypes were grouped into high (top ), mid-high ( – ), mid-low ( – ) and low (bottom ), the high yielding hybrids starbrite f , stars- n-stripes f , fiesta f , and regency f had mid-high fruit count, low to mid-low percentage cull, mid-high to mid-low percentage early and high to mid-high fruit size (table ). the high yielding inbreds big crimson, stone mountain, calhoun gray, and legacy had mid-high to mid-low fruit count, high to low percentage cull and percentage early, and high to mid-low fruit size (table ). genotypes with the lowest marketable yield had high fruit count; high to low percentage cull and percentage early; and mid-low to low fruit size. the stability rank correlation of gge with mean marketable yield, fruit count and percentage cull was positive and significant, suggesting that pc of the gge biplot captured a large part of the variation due to geno- type. unlike ammi analysis, which removed genotype as a main effect, gge biplot did not remove genotype as a main effect. thus, gge biplot captured more variation for all watermelon yield traits evaluated in this study. a similar observation was reported by ding et al. ( ), karimiza- deh et al. ( ), and yan et al. ( ). other advantages of gge biplot were interactive graphics for genotype eval- uation and identification of mega-environments and test locations. thus, gge biplot was preferred over ammi analysis when applying stability-selection criteria. how- ever, gge biplot has certain limitations including its abil- ity to explore only a small portion of total genotype effect and g´e effect. that occurs when the genotype effect is smaller than the g´e effect, and the g´e pattern in not constant over years (ding et al., ). however, gge biplot analysis captured a large part of genotype and g´e variation for all yield traits evaluated in this study. similarly, for ys i , the f test value for s i was non- significant for all the traits evaluated. thus, it captured a large portion of the trait mean, causing the rank for m and ys i remain unchanged. significant correlations for s i with ŝ i , w i , and s d for all the traits evaluated in this study crop science, vol. , july– august www.crops.org suggested that these statistics were measuring the same thing (wachira et al., ), and that they could be used interchangeably to select stable genotypes. conversely, nonsignificant correlation was found for some of the stabil- ity statistics (b i , s d , s i , gge, and ammi) and trait mean for marketable yield, fruit count, percentage early and fruit size (table ). thus, stability statistics provide information that cannot be obtained from trait mean (mekbib, ). similarly, the weak correlation of stability statistics (s i , ŝ i , w i , b i , s d , gge, and ammi) with trait mean indi- cated the need to use these statistics simultaneously (kang and pham, ; kang and gauch, , mekbib, ). the negative and nonsignificant correlation of s i and s d with mean for marketable yield indicated that high yield- ing genotypes were often stable (low s i and s d ). fig. . additive main effects and multiplicative interaction (ammi ) biplot showing the main and first principal components (pc ) effects of both genotypes and environments on (panel a) marketable yield, (panel b) fruit count, (panel c) percentage cull fruit, (panel d) percent- age early fruit, and (panel e) fruit size of watermelon genotypes tested in yr and eight locations. www.crops.org crop science, vol. , july– august stone mountain (g ), and calhoun gray (g ). these genotypes had high marketable yield and low s i ; b i equal to unity; nonsignificant s d ; close to ideal genotype and near- zero projections onto the aec (vertical axis) on mean vs. stability view of gge biplot and ys i higher than the mean ys i . genotypes stars-n-stripes f (g ), fiesta f (g ), based on multiple stability measures (s i , b i , s d , ys i ,and gge), watermelon genotypes were classified into three cat- egories. category was genotypes having high market- able yield and high stability. these genotypes are widely adapted across diverse environmental conditions. those were genotypes stars-n-stripes f (g ), fiesta f (g ), table . spearman rank correlations among trait mean (m) and stability parameters (regression coefficient [bi], deviation from regression [s d], shukla’s stability variance [si ], gge, and ammi for watermelon based on genotypes tested in environments. m si ŝi wi bi s d gge ammi marketable yield m si – . ŝi – . . *** wi – . *** . *** bi . . – . . s d – . . *** . *** . *** . gge . *** – . – . – . . – . ammi – . . . . . . – . fruit count m si – . ŝi – . *** . *** wi – . ** . *** bi . . – . . s d – . *** . *** . *** . *** – . gge . *** – . – . * – . . – . * ammi – . – . – . – . – . – . – . percentage cull fruit m si . *** ŝi . *** . *** wi . *** . *** bi . *** . *** . *** . *** s d . *** . *** . *** . *** . *** gge . *** . *** . *** . *** . *** . *** ammi . – . – . – . . – . . percentage early fruit m si – . * ŝi – . ** . *** wi – . ** . *** bi . – . – . – . s d – . ** . *** . *** . *** . gge . *** – . – . * – . * – . – . * ammi – . * . . . – . . – . * fruit size m si – . ŝi – . *** . *** wi – . . *** bi . *** – . – . – . s d – . ** . *** . *** . *** – . gge – – – – – – ammi – – – – – – * significance level at . level of probability. ** significance level at . level of probability. *** significance level at . level of probability. crop science, vol. , july– august www.crops.org stone mountain (g ), and calhoun gray (g ) exhibited consistent high marketable yield in varied environmental conditions. hybrids stars-n-stripes f (g ) and fiesta f (g ) had average high fruit count, percentage early, and fruit size; and low percentage cull. in contrast, inbreds stone mountain (g ) and calhoun gray (g ) had low performance for yield components. category genotypes had long or round fruit shape, medium size fruit ( – kg fruit– ), medium rind thickness, and red flesh color. category genotypes exhibited high marketable yield but low stability, so these genotypes are suited for specific environments. this category includes genotypes big crim- son (g ), starbrite f (g ), legacy (g ) and mountain hoosier (g ). genotypes big crimson (g ) and starbrite f (g ) had high marketable yield and were significantly higher than legacy (g ) and mountain hoosier (g ). for marketable yield, category genotypes had b i greater than unity, significant s d , high s i , and high projections onto the aec (vertical axis) on the mean vs. stability view of gge biplot. these finding suggest that big crimson (g ), starbrite f (g ), legacy (g ) and mountain hoo- sier (g ) were sensitive to environmental change (below average stability) and had greater specificity of adaptabil- ity to high yielding environments. however, according to ys i , category genotypes were better than average and should be considered stable. unlike category genotypes that had mid-low to low percentage early and fruit size, category genotypes had high % early and fruit size. for fruit count and percentage cull, category genotypes had high to mid-low. overall, for marketable yield, percentage cull, percentage early and fruit size, category genotypes were slightly better in performance than category geno- types. plant breeders can use big crimson (g ), starbrite f (g ), legacy (g ) and mountain hoosier (g ) to develop high performers for specific locations. category genotypes had low marketable yield and stability. these genotypes are suitable for breeding for traits other than yield, for example, disease resistance or fruit quality. category genotypes include golden midget (g ), sugar baby (g ), and early canada (g ). these genotypes had marketable yield significantly lower than the other genotypes. for other yield components, category genotypes recorded high fruit count, and low percent- age early and small fruit size. the phenotype of category genotypes included small fruit ( . – . kg/fruit), green or yellow fruit skin, thin rind, and red flesh color. conclusions several watermelon genotypes had significant g´e for yield and yield components, and there was evidence for the advantage of hybrids over inbreds for yield and responsiveness to favorable environments. we classified watermelon genotypes into three categories based on yield and stability. category genotypes had high marketable yield and stability, and are widely adapted across diverse environmental conditions: stars-n-stripes f , fiesta f , stone mountain, and calhoun gray. category geno- types exhibited high marketable yield but low stability and were suited to specific locations: big crimson, starbrite f , legacy, and mountain hoosier. category genotypes had low marketable yield and stability: golden midget, sugar baby, and early canada. the highest performing inbred and hybrid genotypes for watermelon fruit yield and yield components (big crimson and starbrite f ) were not the highest for yield stability. not all genotypes from the three categories were stable for all yield components. therefore, there is room for improvement. category hybrids fiesta f and stars- n-stripes f were more stable for yield components eval- uated in this study than inbreds stone mountain and cal- houn gray. inbreds stone mountain and calhoun gray from category had lower quality than inbreds legacy and big crimson of category . although, legacy and big crimson were unstable for marketable yield, they were stable in performance for some yield components, which were lacking in stone mountain and calhoun gray. the strong positive correlation (p < . ) of s i with ŝ i , w i , and s d (for all the traits) suggested these statistics measure the same aspect of stability (wachira et al., ). therefore, these stability statistics could be used interchangeably to select stable genotypes. the gge biplot captured more variation than ammi biplots for all the watermelon yield traits evaluated in this study. additionally, gge biplot generated more interactive visuals for genotype evaluation. g´e plays a significant role in breeding watermelons for high yield, wide or specific adaptation, and stability. supplemental materials available supplemental material is available with the online version of this article. acknowledgments the authors would like to thank the following individuals for assistance and support with field and laboratory experi- ments: ms. tammy l. ellington at the north carolina state university, raleigh; mr. mark schaffer at clemson univer- sity, charleston, sc; mr. jackie snell, university of florida, gainesville; ms. desiree koehn at texas a&m university, college station; ms. cathy white and ms. dolores safran at monsanto vegetables, woodland, ca; and ms. amy helms and mr. cody sheffield at usda-ars, lane, ok. references alghamdi, s.s. . yield stability of some soybean genotypes across diverse environment. pak. j. biol. sci. ( ): – . doi: . /pjbs. . . www.crops.org crop science, vol. , july– august annicchiarico, p. . joint regression vs. ammi analysis of gen- otype-environment interactions for cereals in italy. euphytica : – . bednarz, c.w., d.c. bridges, and s.m. brown. . analysis of cotton yield stability across population densities. agron. j. : – . doi: . /agronj . x calderini, d.f., and g.a. slafer. . has high stability changed with genetic improvement of wheat stability? euphytica : – . doi: . /a: carbonell, s.a.m., j.a.a. filho, l.a.s. dias, a.f.f. garcia, and l.k. morais. . common bean cultivars and lines interac- tions with environments. sci. agric. (piracicaba, brazil) : – . doi: . /s - casanoves, f., j. baldessari, and m. balzarini. . evaluation of multienvironment trials of peanut cultivars. crop sci. : – . doi: . /cropsci . chakroun, m., c.m. tliaferro, and r.w. mcnew. . genotype- environment interactions for bermuda forage yields. crop sci. : – . doi: . /cropsci . x x comstock, r.e., and r.h. moll. . genotype-environment interactions. in: w.d. hanson and h.f. robinson, editors, statistical genetics and plant breeding. nas-nrc publ. . natl. academy of sci.-natl. res. council, washington, dc. p. – . cran. . the comprehensive r archive network. compre- hensive r archive network for the r programming language. http://cran.r-project.org/web/packages/available_packages_by_ name.html#available-packages-a (accessed september ). crossa, j., h.g. gauch, and r.w. zobel. . additive main effects and multiplicative interaction analysis of two maize culti- var trials. crop sci. : – . doi: . /cropsci . x x dehghani, h., a. ebadi, and a. yousefi. . biplot analysis of genotype by environment interaction for barley yield in iran. agron. j. : – . doi: . /agronj . de vita, p., a.m. mastrangelo, l. matteu, e. mazzucotelli, n. virzi, m. palumbo et al. . genetic improvement effects on yield stability in durum wheat genotypes grown in italy. field crops res. : – . doi: . /j.fcr. . . dia, m., and t.c. wehner. . analysis of genotype x environment interaction (gxe) using sas programming. http://cuke.hort.ncsu. edu/cucurbit/wehner/software.html (accessed mar. ). ding, m.m., b. tier, w. yan, h.x. wu, m.b. powell, and t.a. mcrae. . application of gge biplot analysis to evalu- ate genotype (g), environment (e) and gxe interaction on p.radiata: a case study. n. z. j. for. sci. : – . eberhart, s.a., and w.a. russell. . stability parameters for comparing varieties. crop sci. : – . doi: . /cropsci . x x ebdon, j.s., and h.g. gauch, jr. . additive main effect and multiplicative interaction analysis of national turfgrass perfor- mance trials: i. interpretation of genotype environment inter- action. crop sci. : – . doi: . /cropsci . fan, x.m., m. kang, h. chen, y. zhang, j. tan, and c. xu. . yield stability of maize hybrids evaluated in multi-environment trials in yunnan, china. agron. j. : – . doi: . / agronj . finlay, k.w., and g.n. wilkinson. . the analysis of adaptation in a plant breeding programme. aust. j. agric. res. : – . doi: . /ar freeman, g.h. . statistical methods for the analysis of genotype- environment interactions. heredity : – . doi: . / hdy. . freeman, g.h., and j.m. perkins. . environmental and gen- otype-environmental components of variability. viii. relation between genotypes grown at different environments and mea- sures of these environments. heredity : – . doi: . / hdy. . gauch, h.g. . statistical analysis of yield trials by ammi and gge. crop sci. : – . doi: . /cropsci . - gusmini, g., and t.c. wehner. . foundation of yield improve- ment in watermelon. crop sci. : – . gusmini, g., and t.c. wehner. . fifty years of yield improve- ment for cucumber, melon and watermelon in the united states. horttechnology ( ): – . holmes, g.j., and j.m. kemble. . southeastern us veg- etable crops handbook. the packer. http://www.lsuagcen- ter.com /n r /rdon ly res/ c – cbb - e –ba – cefa / / _sevg .pdf (accessed oct. ). kandus, m., d. almorza, r.boggio ronceros, and j.c. salerno. . statistical models for evaluating the genotype-environ- ment interaction in maize (zea mays l.). phyton (buenos ares) : – . kang, m.s. . simultaneous selection for yield and stability in crop performance trials: consequences for growers. agron. j. : – . doi: . /agronj . x kang, m.s. . breeding: genotype-by-environment interac- tion. in: r.m. goodman, editor, encyclopedia of plant and crop science. marcel-dekker, new york. p. – . kang, m.s., and g.h. gauch. . genotype by environment interaction. crc press llc, denvers, ma. kang, m.s., j.d. miller, and l.l. darrah. . a note on relation- ship between stability variance and ecovalence. j. hered. : . kang, m.s., and h.n. pham. . simultaneous selection for high yielding and stable crop genotypes. agron. j. : – . doi: . /agronj . x karimizadeh, r., m. mohtasham, n. sabaghni, a.a. mahmoodi, b. raoutami, f. seyyedi, and f. akbari. . gge biplot anal- ysis of yield stability in multi-environment trials of lentil geno- types under rainfed condition. notulae sci. biol. ( ): – . lu’quez, j.e., l.a.n. aguirreza’bal, m.e. aguero, and v.r. pereyra. . stability and adaptability of cultivars in non- balanced yield trials: comparison of methods for selecting ‘high oleic’ sunflower hybrids for grain yield and quality. crop sci. : . doi: . /j. - x. . .x magari, r., and m.s. kang. . genotype selection via a new yield stability statistic in maize yield trials. euphytica : – . doi: . /bf maynard, d.n. . watermelons. characteristics, production, and marketing. ashs press, alexandria, va. p. – . mekbib, f. . yield stability in common beans (phaseolus vulgaris l.) genotypes. euphytica : – . doi: . /a: miranda, g.v., l.v. de souza, l.j.m. guimaraes, h. namorato, l.r. oliveira, and m.o. soares. . multivariate analysis of genotype x environment interaction of popcorn. pesp. agropec. bras., brasilia. : – . mulema, j.m.k., e. adipala, o.m. olanya, and w. wagoire. . yield stability analysis of late blight resistant potato selections. exp. agric. : – . crop science, vol. , july– august www.crops.org murphy, s.e., e.a. lee, l. woodrow, p. seguin, j. kumar, i. raj- can, and g.r. ablett. . genotype x environment interac- tion and stability for isoflavone content in soybean. crop sci. : – . doi: . /cropsci . . nachit, m.m., m.e. sorrells, r.w. zobel, h.g. gauch, r.a. fischer, and w.r. coffman. . association of environmental variables with sites’ mean grain yield and components of geno- type-environment interaction in durum wheat. j. genet. breed. : – . padi, f.k. . relationship between stress tolerance and grain yield stability in cowpea. j. agric. sci. camb. : – . doi: . /s panthee, d.r., c. cao, s.j. debenport, g.r. rodriguez, j.a. labate, l.d. robertson, a.p. breksa, iii, e.v.d. knaap, and b.b.m. gardner. . magnitude of genotype x environment interac- tions affecting tomato fruit quality. hortscience ( ): – . rak, k., f.m. navarro, and j.p. palta. . genotype x stor- age environment interaction and stability of potato chip color: implications in breeding for cold storage chip quality. crop sci. : – . doi: . /cropsci . . r development core team. . r: a language and environ- ment for statistical computing. r foundation for statisti- cal computing, vienna, austria. http://www.r-project.org/ (accessed sept. ). riday, h., and e.c. brummer. . persistence and yield stabil- ity of intersubspecific alfalfa hybrids. crop sci. : – . doi: . /cropsci . rstudio. . rstudio: integrated development environment for r (computer software v . . ). rstudio. http://www.rstu- dio.org/ (accessed sept. ). sanders, d.c., editor. . vegetable crop guidelines for the south- eastern u.s. – . north carolina vegetable growers assoc., raleigh. sas. . sas: business analytics and business intelligence soft- ware. sas inst. http://www.sas.com/en_us/home.html (accessed sept. ). shukla, g.k. . some statistical aspects of partitioning genotype- environmental components of variability. heredity : – . doi: . /hdy. . usda. . usda, natl. agric. statistics serv. http://www.nass. usda.gov/quickstats/index .jsp (accessed apr. ). vasanthkumar, a.m. shirol, r. mulge, n. thammaih, and p. kumar. . genotype x environmental interaction in water- melon (citrullus lanatus thumb.) genotypes for yield and quality traits. karnataka j. agric. sci. ( ): – . wachira, f., w. ng’etich, j. omolo, and g. mamati. . gen- otype x environment interaction for tea yields. euphytica : – . doi: . /a: yan, w. . ggebiplot: a windows application for graphical anal- ysis of multi-environment trial data and other types of two-way data. agron. j. : – . doi: . /agronj . x yan, w., l.a. hunt, q. sheng, and z. szlavnics. . cultivar evaluation and mega-environment investigation based on gge biplot. crop sci. : – . doi: . /cropsci . x yan, w., and m.s. kang. . gge biplot analysis: a graphical tool for breeders, geneticists, and agronomists. crc press, boca raton, fl. yan, w., m.s. kang, b. ma, s. woods, and p.l. cornelius. . gge biplot vs. ammi analysis of genotype-by-environment data. crop sci. : – . doi: . /cropsci . . yan, w., and n.a. tinker. . biplot analysis of multi-environ- ment trial data: principles and applications. can. j. plant sci. : – . doi: . /p - zobel, r.w., m.j. wright, and h.g. gauch, jr. . statistical analysis of a yield trial. agron. j. : – . doi: . /agron j . x supplemental material description the supplemental tables provide a listing of pedigree and phenotype of all genotypes used in the study. table s . the watermelon genotypes tested with pedigree information. id genotype year of release pedigree g au-jubilant jubilee x pi g allsweet [(miles x peacock) x charleston gray] g big crimson na† na g black diamond segregation within cannonball or black diamond g calhoun gray calhoun sweet x charleston gray g calsweet na [(miles x peacock) x charleston gray] g carolina cross # na na g charleston gray [{(africa x iowa belle) x garrison} x garrison] x [(hawkesbury x leesburg) x garrison] g congo (african x iowa belle) x garrison g crimson sweet (miles x peacock) x charleston gray g desert king na na g early arizona na na g early canada na na g fiesta f unknown (plant variety protection) g georgia rattlesnake g golden midget new hampshire midget x pumpkin rind g graybelle sugar baby x charleston gray sister line g hopi red flesh na na g jubilee africa , iowa belle, garrison, hawkesbury, and leesburg g king & queen na na g legacy (early gray x little jubilee ) x verona g mickylee texas w , fairfax, summit, and graybelle g minilee texas w , fairfax, summit, and graybelle g mountain hoosier na na g nc giant na na g navajo sweet na na g peacock wr- klondike r x peacock g quetzali na g regency f unknown (plant variety protection) g royal flush f unknown (plant variety protection) g sangria f na unknown (plant variety protection) g starbrite f na unknown (plant variety protection) g stars-n-stripes f na unknown (plant variety protection) g stone mountain na g sugar baby tough sweets selection, inbred yr g sugarlee texas w , summit, charleston gray, fairfax, crimson sweet, and graybelle g sweet princess small-seeded congo type x charleston gray g tendersweet of na na g tom watson na g yellow crimson na na † not available. table s : fruit and seed traits for the watermelon genotypes evaluated. id genotype fruit rind seed flesh color¶¶ shape† size‡ color§ thickness¶ pattern# size†† color‡‡ au-jubilant l m lg s m l r r allsweet l m lg n w s r s big crimson r m mg m w m r r black diamond r s db m s l r r calhoun gray l m g m s m r r calsweet l m lg t w s b s carolina cross # e g lg t n l w r charleston gray l l g t r m r r congo l m dg m r l t r crimson sweet r m lg m m s b r desert king o s sg s s m r o early arizona o s sg n s l r r early canada r s g n r s r r fiesta f l m lg m w s b r georgia rattlesnake l g lg m n l r r golden midget o c y n s l r r graybelle g s g n s s r r hopi red flesh o m sg m s l b r jubilee l l lg t n l r r king & queen o m lg n n m b r legacy l m lg n n m r r mickylee r n lg n r m r s minilee r s g n r s r s mountain hoosier o m sg t s l w r nc giant l g lg t r l r r navajo sweet r s lg m n m r r peacock wr- l s sg m s s r r quetzali r s lg n s m r r regency f o s mg m m s t r royal flush f l m mg m w s b s sangria f o m mg m s s b s starbrite f o m lg l s s r r stars-n-stripes f o m dg t w l b s stone mountain o m sg t s l t r sugar baby r m mb s s s r s sugarlee r s lg m n m r r sweet princess o m g m r t r r tendersweet of e m dg m w l w o tom watson e m mg t s l t r yellow crimson l l lg n s l b c † fruit shape: elongate (e), oval (o), round (r) ‡ fruit size: micro ( < lb.) (c), mini ( – lb.) (n), icebox ( – lb.) (b), small (s), sometimes called pee-wee ( – lb.), medium ( – lb.) (m), large ( – lb.) (l), and giant ( > lb.) (g). § rind thickness: thick ( > mm) (t), medium ( – mm) (m), thin ( < mm) (n). ¶ rind color: light green (lg), medium green (mg), dark green (dg), solid light black (lb), solid medium black (mb), solid dark black (db) golden (g), solid green (sg), gray (r), yellow (y), mottled (m). # rind pattern: wide stripe (w), medium stripe (m), narrow stripe (n), gray (g), solid (s), rattle snake (r) [dark green is dominant, stripe is decided by dark green]. †† seed size: tomato size (t), small (s), medium (m), large (l). ‡‡ seed color: black (b), brown (r), tan (t), dotted (d), white (w). §§ flesh color: scarlet red (s), coral red (r), orange (o), salmon yellow (y), canary yellow (c), or white (w). 	supplemental material description creighton university skip to main content 	 	 		 covid- response: get the latest creighton university information and updates. visit staying safe, staying creighton. faculty, staff and students, register to be tested at our on-campus testing center. learn more about the community vaccine clinics. 		 			 				 											 									 				 											 transaction menu 	apply 	visit 	request info 	give 									 				 											 							 								 							 						 										 							 								 										menu 								 								 									 									 										 	admissions 	academics 	student experience 	athletics 	research 	about 	make a gift 	campus maps 	a-z index schools and colleges 	college of arts and sciences 	college of nursing 	college of professional studies 	graduate school 	heider college of business schools and colleges 	school of dentistry 	school of law 	school of medicine 	school of pharmacy and health professions information for 	future students 	current students 	adult students 	military/veteran students information for 	alumni, donors and friends 	parents and visitors 	faculty and staff 	professional development/ce 	phoenix health sciences campus resources for current students, faculty and staff 	directory 	office 	blueline 	nest 	libraries 	bookstore 	calendar 	mycreighton 	media 										 									 									 								 							 					 				 			 		 		 						 		 				 											 							 								 							 						 					 										creighton university 									 			 header main navigation 	admissions 	academics 	student experience 	athletics 	research 	about 	 				 										 	 	scholarship available to all creighton alumni the graduate school is now offering creighton alumni a scholarship for % off of tuition for a wide range of graduate and certificate programs. 	learn more about this alumni scholarship 	 	visit campus virtually a personalized virtual visit is a great way to see our campus from the comfort of your home. 	sign up for a virtual visit today 	 	explore over academic programs creighton university offers a top-ranked education for undergraduate, graduate and professional students with programs in the arts, sciences, law, business and health professions. 	find your program 	 	transforming health care for the underserved creighton university health sciences faculty and students will play a lead role in a new effort to reduce growing health disparities in phoenix. 	read more about this initiative 	 	one of america's best colleges with over award-winning programs, creighton university’s graduate school will help advance your career. 	see our rankings 	 	cl werner center for health sciences education a transformational gift from cl and rachel werner will expand creighton university’s national reputation as a leader in preparing students for interprofessional, team-based health care. 	read more schools and colleges 	college of arts and sciences 	college of nursing 	college of professional studies 	graduate school 	heider college of business 	school of dentistry 	school of law 	school of medicine 	school of pharmacy and health professions find a major or program: 	 search for programs new applied chemistry major study alongside faculty mentors, with an opportunity to be published in scientific journals and present at national meetings. learn more about new bs major new applied chemistry major internships pay dividends with five internships under his belt, senior jeremy plummer lands a job with a private equity firm in kansas city. read article internships pay dividends meet the jesuits dzao vu and jeff sullivan, ba’ , are spending part of their jesuit formation process at creighton university. read article meet the jesuits inclusive excellence creighton works to create a welcoming and diverse environment for all. learn more about diversity and inclusion inclusive excellence meet our vaccine volunteers the covid- community vaccine clinic at creighton is made possible by student volunteers. read about their impact meet our vaccine volunteers the human element in teaching endowed professors inspire students as agents of change. meet victoria haneman, the frank j. kellegher professor of trusts and estates. learn about her project research the human element in teaching 	 	 facebook twitter instagram youtube 	 	 	 			 				 	creighton’s social science data lab aims to provide research and data services to nonprofit and advocacy groups. 			 				 	hey @creighton students! you can join us on a fall break service & justice trip! 			 				 	sarah walker will head creighton’s diversity and inclusion efforts. read more about her important role. 			 				 	creighton lost a friendly bet with gonzaga university during #marchmadness. 	 	 creighton by the numbers 	 	 the creighton experience creighton university's students share a unique bond. together with faculty and staff, they form a passionate community in which they are driven to do more, challenged to make a positive change in the world, and guided by fundamental values of the jesuit tradition. our students say it best 	 footer transaction menu 	apply 	visit 	request info 	give 		 	 header main navigation select from main menu 	admissions 	academics 	student experience 	athletics 	research 	about 							 					 						 					 				 			 	 		 	 california plaza omaha, ne . . ask a question 	human resources 	handshake 	myit 	safety & compliance 	facebook 	twitter 	linkedin 	youtube 	instagram © creighton university 	 public administration review • november | december norma m. riccucci, editor rosemary o’leary syracuse university when a career public servant sues the agency he loves: claude ferguson, the forest service, and off -road vehicles in the hoosier national forest administrative profi le: claude ferguson rosemary o’leary is distinguished professor and phanstiel chair at the maxwell school of syracuse university, and the author of the ethics of dissent: managing guerrilla government. e-mail: roleary@maxwell.syr.edu claude ferguson, who in his own words “met the test of his lifetime,” deviated from the norms of the u.s. forest service articulated by herbert kaufman in th e forest ranger to became a government guerrilla against the organization he loved. th is profi le highlights several enduring themes: the inherent tensions between democracy and bureaucracy, the many masters of career bureaucrats, how organizational culture can both empower and constrain employees, and what it means to act responsibly, ethically, and with integrity as a public servant. in addition, this case demonstrates how the forest service has evolved since kaufman’s classic study. first, kaufman depicted forest rangers as “valuing the organization more than they value[d] getting their own way,” yet this profi le underscores that public servants do not check their worldviews, mores, or ethics at the door. second, kaufman described the forest service’s eff orts to routinize the decisions of its employees in an eff ort to prevent allegiances to, or co- optation by, local populations. yet in this administrative profi le, ferguson’s hidden strategic tactics co-opted local stakeholders to enlist their support for a cause he deeply felt was right and just. this administrative profi le concerns the ef-forts of a career employee of the u.s. forest service to protest and eventually halt off -road vehicle (orv) trails built in the hoosier national forest in the state of indiana. th is story is about one man, claude ferguson, who in his own words “met the test of his lifetime” and, deviating from forest service norms, became a government guerrilla against the organization he loved. “guerrilla government” is my term for the actions of career public servants who work against the wishes—either implicitly or explicitly communicated—of their superiors (o’leary ). while this case took place in the s, it highlights several enduring themes that are still relevant to- day: the inherent tensions between democracy and bureaucracy, the many masters of career bureaucrats, the ways in which organizational culture can both empower and constrain employees, and what it means to act responsibly, ethically, and with integrity as a public servant. th e culture of the forest service was fi rst brought into the limelight by herbert kaufman in his classic book th e forest ranger ( ). in that work, kaufman examined the for- est service of the s from the ranger district upward. in kaufman’s own words, his book is about how daily decisions and actions at lower echelons make concrete the realities of policy statements and the declared objectives of the leadership. forest rangers are members of forest rangers . . . . have many masters: local residents, timber companies, ranchers, miners, conservationists, members of the general public, congresspersons, and the president. despite the possibilities for fragmentation, the forest service of kaufman’s era was amazingly cohesive and uniform in action. administrative profi le the federal bureaucracy, yet much of their work is carried out in a decentralized fashion in locations far from washington, d.c., in re- motely dispersed locations. th ey have many masters: local residents, timber companies, ranchers, miners, conservationists, members of the general public, congresspersons, and the president. despite the possibilities for fragmentation, the forest service of kaufman’s era was amazingly cohesive and uniform in action. much of kaufman’s book describes the mechanisms by which forest service leaders maintained uniformity and control over the diff use organization, seeking in part to discourage guerrilla government before it had a chance to germinate. as deviation threats increased, for example, central controls multiplied. as impulses toward fragmentation grew, the discretion of fi eld offi cers was contracted. in order to narrow latitude, “preformed decisions” were made at all levels above the rangers (kaufman , ). rangers were thoroughly screened to promote homogeneity, while the forest service “manipulate[d] the intellects and wills” ( ) of its mem- bers. in-service indoctrination and training promoted standardiza- tion. an attempt to diff use diff erences of opinion was made prior to the promulgation of policies. allegiances to local populations were neutralized by frequently rotating rangers throughout the united states. th e result was that the patterns of informal organization in the national forests were rarely at odds with the policies enunciated at higher levels, and centrifugal tendencies were vanquished. forest rangers in the s, in short, tended to “value the organization more than they value[d] getting their own way” ( ). despite these attempts to forge a tightly run forest service and to produce nearly all-obeying forest rangers, kaufman acknowledged that there were exceptions: “in the last analysis” he wrote, “all infl uences on admin- istrative behavior are fi ltered through a screen of individual values, concepts, and images” ( ). claude ferguson: the early years claude ferguson fi rst joined the forest service in , years before the publication of kaufman’s book, as a lookout in the mark twain national forest in missouri while a senior in high school. he later served as a national youth administration crew member before permanent appointment as a fi re control aide with the forest service in the mark twain national forest. he worked closely with the civilian conservation corps in fi re control in three camps. after three years in the u.s. navy, he returned to the mark twain national forest as a forestry aid for four years, where he was engaged in timber management and land acquisition. prior to working in the hoosier national forest, he worked as a forester in the nicolet national forest in wisconsin, as district ranger in the shawnee national forest in illinois, as district ranger in the hiawatha national forest in upper michigan, and as staff forester in the ottawa national forest in upper michigan in charge of lands, recreation, wildlife, soil and water. he later served as chief of the branch of cooperative forestry management and chief of the branch of operations in the forest service regional offi ce in milwaukee. in , he was transferred to bedford, indiana, where he was forest supervisor for both the hoosier national forest and the wayne national forest in ohio. in , he stepped down from the position of forest supervisor at his own request after he married another forest service employee. his successor was a man named donald girton, who will return later in this profi le. the off-road vehicle challenge emerges ferguson’s awareness of an off -road vehicle problem in indiana was sparked by an event that occurred in april , when he was forest supervisor. th e event was dubbed the “buff alo ,” named after john buff alo, an avid motorcyclist who had purchased acres of private land located in the middle of the hoosier national for- est. without asking for permission or notifying the forest serv- ice, buff alo and his friends marked a -mile trail through the hoosier national forest and held a motorcycle race on the federal land. with the district ranger, ferguson fi lmed the damage caused by the motorcyclists, including the tearing up of hiking trails, the destruction of fragile forest land not meant for trails, the trampling of young trees, the ripping down of branches, the destruction of wildlife habitat for both endangered and nonendangered species, littering, and excessive noise levels. several motorcyclists fractured bones when their orvs hit tree stumps and tree limbs. when the organizers of the buff alo were told by ferguson the next year that they could not hold their event again in the national forest, he was warned that they “know how to strike matches,” a not-too-veiled threat that they were willing to burn down portions of the forest if ferguson did not kowtow to their demands. fergu- son told them to talk to the bureau of motor vehicles, as it handled motorized vehicles. his job, ferguson told them, was to protect the national forest. th e eastern national forests were established under the weeks act after public outcry concerning uncontrolled logging and fi res. th e act allowed the federal government to purchase private land east of the mississippi river for the protection of the headwaters of navi- gable streams and for the production of timber. th e boundaries of the hoosier national forest were drawn to embrace the lands of the nor- man and crawford physiographic regions in indiana that were most susceptible to erosion by water. by defi nition, according to ferguson, they would be the last soil types to select for orv use in indiana. “at every staff meeting i attended,” ferguson said, there was not one staff member who supported the use of hoosier national forest lands for orvs. he continued, after one of the many show[s] of hands on this question at a staff meeting, i questioned further consideration of the issue because we had reached a consensus. mr. girton, the for- est supervisor, advised the group that this was an “erroneous consensus,” and that he was speaking to a wealth of interdisci- plinary professional people whom he had hired to counsel and advise him in their fi elds of expertise. girton obviously felt he had to be all things to all people and that it was in our best interests to broaden our base of support beyond hikers, hunt- ers, conservationists, and environmentalists. after several such meetings it became clear that the fi nal orv policy as adopted and implemented would be solely girton’s policy, ghost-writ- ten in part by the american motorcycle association (ama). when the ama published proposed orv trail standards at the tail-end of this process, professional foresters blasted them. ferguson works with environmental interest groups in july , the forest service held many listening sessions in order to glean public sentiment on the issue. many people spoke at public administration review • november | december these sessions, both for and against orvs in the hoosier national forest. despite the fact that the data gathering was not yet com- plete, supervisor girton stated that certainly some portion of the forest would be allotted to orv users, enraging many who felt that he had made up his mind prematurely and without analyzing the bi- ophysical evidence or listening to the majority of the public. when written comments were solicited, they ran to against reopening a portion of the hoosier national forest to orv use. despite the widespread negative sentiment concerning orvs in the forest from both the lay public and professional foresters, girton made the deci- sion to proceed with a policy of orv use in the hoosier national forest and announced that it might occur as early as september . ferguson commented, in my mind this decision was just plain wrong for several reasons: data collection and analyses were incomplete; the public, generally, did not want the trails; professional foresters had counseled against the trails; and the fragile indiana ter- rain could not support the trails. we were kowtowing to one special interest group: the ama, who, by the way, had bought girton’s two children trail bikes. in december , the state of indiana pub- lished a report concluding that orv use was not compatible with the natural resource pur- poses of the state properties, and therefore con- tinued its closure of state properties to orvs. on the federal side, the forest service decided in to write an environmental impact statement (eis) and proposed rules regulating orv usage in limited terrains. frustrated with the closed doors he was facing with girton, ferguson took guerrilla action and submitted comments expressing his professional view against orv trails through the national environmental group citizens for a better environment. ferguson explained, by the time the fi nal eis was written and distributed in , the original orv trail standards had been removed, language had been watered down, and trail standards developed by the ama—which arrived after the cutoff date for respons- es—were published as an appendix to the eis. th e public response to this unpopular decision was as expected—people were outraged and my phone rang off the hook with calls from angry citizens. in response to the eis, the indiana division of the izaak walton league of america (iwl), of which ferguson was a member, fi led an administrative appeal with the regional forester requesting that the policy be set aside. ferguson helped draft their appeal, but initially did not reveal his authorship. before responding to the iwl appeal, in august , the forest service launched a crash project to con- struct and open orv trails in the hoosier national forest by octo- ber , . th e iwl reacted by requesting a stay of construction from the regional forester pending a decision on its administrative appeal. th e regional forester denied the stay on october , . trail construction was commenced without advance approval, as required by forest service regulations, and without the required advance-approved project plans. th e forest service manual in eff ect at the time required advance approval, on an individual project basis, by the regional forester, for all grades with slopes in excess of percent. th ere were several such grades completed on this project prior to any submission of the construction plans for regional forester approval. th e survey and design were conducted simultane- ously with construction, which was contrary to the federal require- ment that plans be completed and approved prior to construction. ferguson explained, i witnessed these trails being built illegally with $ , of public money that was budgeted for routine maintenance of roads and trails. th e forest service’s own guidance on fi nan- cial planning for that fi scal year clearly said, in regard to forest roads and trails appropriation: “forest supervisors have no authority to make fund adjustments between the maintenance and construction activities of the forest roads and trails ap- propriation” (u.s. department of agriculture, forest service guidance on planning, fy and fy ). th e hoosier national forest budgets had been severely reduced already for fi scal years and . th is reduction was particularly severe for the forest roads and trails appropriation. th e budget cuts forced a reduction in force (rif) in engineering personnel and were barely suffi cient for routine maintenance of the then existing roads and trails systems. th e contractor who did the work was told to bill the forest service for “routine maintenance.” to divert these funds for the pet project of one man was clearly against the public will and a violation of law. i was outraged. ferguson decides to sue the forest service on october , , forest supervisor girton and district ranger frank haubry conducted a fi eld trip for the hoosier national forest staff to review the newly constructed orv trails. ferguson was one of those on the fi eld trip. th ey met at an assembly point and proceeded to drive the two-track trail system in four wheel drive vehicles. th ey examined small segments of the one-track system by short hikes on foot. ferguson expressed his reaction: i was appalled. i personally observed violations of the [febru- ary ] president’s executive order [concerning the use of orvs on public lands] as well as fl agrant violations of the forest service’s trail standards. for example, roads and trails were located and constructed to damage soil and watershed on lands acquired for the protection of streams. vegetation had been removed and destroyed during the construction and more damage was eminent from the proposed use on lands acquired for the production of timber. trails were located on land acquired especially for wildlife habitat and public hunt- ing and specially developed by the indiana department of natural resources using hunter fi rearms tax money under the pittman-robertson wildlife restoration act. th ese trails were located in one of the most signifi cant wildlife habitat areas in the state of indiana and in one of the very few wild turkey and ruff ed grouse ranges in the state. th e trails … ferguson took guerrilla action and submitted comments expressing his professional view against orv [off -road vehicle] trails through the national environmental group citizens for a better environment. administrative profi le were located where signifi cant confl icts would result with other existing recreational uses—primarily hunting, hiking, and horseback riding. it was evident that very little consideration had been given to the eff ects of noise and exhaust pollution on the forest, water, and wildlife. i observed violations of regula- tions for reverse curves, trail grades, and minimum safe stop- ping sight distance. i observed stumps that were not fl ush cut in the middle of orv trails. i observed fallen logs, up to six inches in diameter that were left in place across trails. i observed many areas requiring trail surfacing that were not surfaced. my most serious concern, however, was this: given the fact that there was no age limit for orvs on these trails, they posed one of the most serious threats to public safety—espe- cially to the very young—that i had ever witnessed on public land paid for with public funds. liability . . . claims were certain to be fi led against the forest service and lives could be lost—with the blessing of the forest service. i could not live with the thought of a child losing his or her life because of our negligence or inaction. at the conclusion of the fi eld trip the group assembled in a parking lot for commentary. i expressed that i was “profes- sionally sick” with what i had just seen: it was truly the most sickening thing i had seen in my career. i recited some of the violations i had seen and asked how any of us could defend what we had done when queried by the public. mr. girton dismissed my remarks by stating that they were one man’s opinion. i immediately called my contacts at the iwl and told them that if they truly were going to sue the forest serv- ice over the orv trails, as we had discussed at an earlier date, to count me in as a supporter. on october , , ferguson conducted a fi eld trip to the orv trails that several members of the public, including environmental- ists and reporters, attended. th at same day, he also participated in a meeting of the indiana conservation council. th e minutes of that meeting reported, claude ferguson reported on the latest developments in the nebo ridge area. (th e nebo ridge area is a pristine wilder- ness area in indiana that several of us were trying to keep in its natural state.) he mentioned that he would be working with a conservation group in formulating plans to initiate a suit to close the hoosier national forest to orv use. he also mentioned that th e nature conservancy will begin a fund drive to obtain money to purchase key tracts of land in the nebo ridge area so that it can be preserved as a wilderness. (indiana conservation council ) ferguson works through the media and the courts on october , , an article authored by a reporter named don jordan appeared in the local paper, the bedford times-mail. fergu- son was quoted as follows when asked his opinion about the orv trails system: “you can quote me as being professionally sick. . . . th is is the most sickening thing i’ve seen in years of service.” th e following day, ferguson was confronted by girton and asked wheth- er the quote was accurate. ferguson said that it was. girton handed ferguson a copy of part of the forest service manual that deals with involvement in public controversies. ferguson later off ered to send the newspaper the following clarifi cation: th e statements ascribed to me were my own professional opinion and were not intended to, nor did they, represent offi cial condemnation or criticism of any policy of my em- ployer. i regret that they may have been so misconstrued and hereby retract any such connotation. a. claude ferguson, october , . (u.s. department of agriculture ) girton would not allow ferguson to take such action, later explain- ing to a special agent of the u.s. department of agriculture that he felt the disclaimer “would tend to compound the public misun- derstanding about the forest service’s position in the matter” (u.s. department of agriculture ). th e iwl then fi led suit in federal court on october , , seeking a temporary restraining order and an injunction halting the trails. an affi davit that ferguson had authored documenting the damage to the orv trails (complete with evidence collected with the assistance of student volunteers from indiana university) was claude ferguson’s clashing obligations forest service regulation . h ( ): “involvement in public controversies. employees are expected to avoid becoming involved in public controversies on matters of public policy. disagreements as to either fact or policy should not, under any circumstances, be publicly aired through statements to the press or any other medium.” code of ethics for government service ( ): “any person in government service should: put loyalty to the highest moral principles and to country above loyalty to persons, party or government department.” part employees’ responsibility and conduct, title agriculture, subtitle a—offi ce of the secretary of agriculture, subpart e ( )—conduct prohibited conduct—general: (b) ( ) taking any action which might prejudice the govern- ment’s interest in a criminal or civil case; (b) ( ) giving aid or assistance, other than in the discharge of offi cial duties, to any claimant in prosecuting any claim against the united states; (b) ( ) (ii) directly or indirectly condemn or criticize the policies of any government department or agency. excerpt from the affi davit of a. claude ferguson; indiana division, izaak walton league of america endowment, inc., vs. donald girton, supervisor, wayne-hoosier national forest, indi- ana and jay cravens, easter region forester, united states forest service: “at this point i respectfully remind the court that i am an employee of the forest service, united states department of agriculture. i am subject to the employee responsibilities and conduct [rules] as set forth in offi ce of personnel regulations part . . . . i am torn between my responsibilities to citizens (subpart a) for whom i serve as a professional manager of their resources and the prohibition against prejudicing the govern- ment’s interest in a criminal or civil case (subpart b (b) ( )).” public administration review • november | december attached to the request. th e temporary restraining order and injunc- tion were granted on october , . after learning that ferguson had fi led the affi davit in the federal court suit, girton wrote to the regional forester, we are now faced with the immediate task of working with the u.s. attorney to prepare for a hearing within the next days. it will be extremely diffi cult for my staff to work with the u.s. attorney to prepare a case that will involve confi - dential communications when a key member of my staff has signed an affi davit in support of the plaintiff ’s position. (u.s. department of agriculture ) five days later, ferguson was informed by girton that he was being transferred by the regional forester to milwaukee to work in the fire and aviation management group, and that his initial detail would be for three weeks, from november , , through november , . girton later put in writing that he considered this transfer “a possible temporary solution to the problem” and “timely” (u.s. department of agriculture ). ferguson responded as follows: i considered the detail to be harassment and punishment for exercising my constitutional right to free speech, as well as a shallow cover for my fraudulent removal from offi ce. i refused to go. (u.s. department of agriculture ) after the regional forester denied the iwl appeal in january , the american motorcycle association joined the lawsuit on behalf of the forest service. bureaucratic guerrilla warfare was in full force. th e sassafras (indiana) chapter of the audubon society, the national audubon society, the indiana conservation council (the national wildlife federation affi liate in indiana), and the national wildlife federation joined the lawsuit supporting the izaak walton league. bureaucracy versus democracy? on march , , ferguson was given notice that he was to be transferred permanently to milwaukee, where he had been “selected” for a gs- position to begin on april . on april , , he wrote to girton and to carl webb, director of personnel management for the forest service in milwaukee, stating his refusal to be transferred, especially given the fact that the forest service had refused to transfer his wife, who was also a forest service employee. th e iwl appealed the orv case to the board of forest appeals on april , , and waited eight months for a denial of appeal from the board. on april , , girton received a memo from webb. in part, that memo read, claude has made public statements regarding forest service policies that have brought criticism against the forest service. th e public has diffi culty diff erentiating from forest service policy and mr. ferguson’s personal views. since one of his key duties as a forest staff offi cer involves the informa- tion and education program, we feel there is a situation that may result in continued confl ict of interest. th ere have been recent complaints about his statement and actions regarding the proposed eastern wilderness legislation. we conclude that this reassignment is warranted to preclude continuation of this problem. on may , , ferguson received a memorandum from webb indicating that his transfer to milwaukee was being delayed pending an investigation, but that he would be detailed on may , . on may , ferguson received a second memorandum stating that the eff ective date of the transfer to milwaukee had been moved to june , . th at same month, may , ferguson received an award from the sassafras chapter of the national audubon society commending his “extraordinary eff orts in the preservation of our natural resourc- es.” th e commendation read in part, claude ferguson is a seasoned forester who is a staff offi cer in the management of the hoosier national forest. for thirty-two years he has served well the interests of us all in the united states forest service. we salute him today not for any single accomplishment, although many could be cited. rather, we want to commend him for a career-long demon- stration of sensitivity and concern for the preservation of our natural environment. most of us fi nd that our own interests often confl ict with those of other people in our society. mr. ferguson labors within a milieu where those confl icting forces have to be balanced. he performs there with a grace and gentle spirit that must be admired. he maintains both a personal and vocational love for the kind of world that audubon members strive to achieve and pass on to subsequent generations. for this we thank him. an opinion survey released by the forest serv- ice for the midlands area soon after this report- ed that percent of the forest service person- nel who replied to a question about whether orvs should be allowed in the forests stated that they should be prohibited. only percent felt that they should have been allowed (dustin ). ferguson broadened his guerrilla attack, and government offi cials reacted: i then asked the offi ce of inspection within the forest service to investigate my case. th ey ended up investigating only my com- plaints of harassment, fraud and deceit in connection with my detail to the regional offi ce in milwaukee (and found my detail to be “reasonable”). two of my colleagues went on record accusing me of being mentally ill. one used confi dential medical records of an illness i had had years ago as “proof ” that i was mentally unstable [u.s. department of ag- riculture ]. . . . th e investigator was told not to look into my charges of wrongdoing connected to the orv issue because it was a subject being considered in the civil action in court. an opinion survey released [in ] by the forest service for the midlands area … reported that percent of the forest service personnel who replied to a question on whether orvs should be allowed in the forests stated that they should be prohibited. administrative profi le in june , ferguson submitted a report to the offi ce of inspec- tion within the forest service that documented wrongdoing in the development of the orv trails. ferguson received no response to that report. after being suspended from the forest service for two weeks without pay, ferguson was notifi ed that he would be removed from offi ce, eff ective february , , for participating in a lawsuit in which the government had an interest, and for confl ict of interest. th is was months before he would have been eligible for retirement. ferguson explained, i asked for early retirement instead, but my request was denied, putting me in jeopardy of losing up to $ , in pension benefi ts due me for years of service. i immediately held a press conference protesting my treatment and fi led a re- quest with the secretary of agriculture for a full investigation. “as of monday i will join the ranks of the unemployed,” the local paper quoted me as saying. “i regret this loss of protec- tion for my family, but honor has no price tag in our circles.” (snapp ) the public rallies; ferguson wins awards th e city council of bedford, headquarters of the hoosier national forest, passed a unanimous resolution requesting that a fair and impartial investigation of ferguson’s treatment be made (joseph b). th at resolution was sent to john r. mcguire, chief of the forest service; edward h. levi, attorney general of the united states; and u.s. representative phil hayes. an editorial in a local paper commented, “it is a sad state of aff airs when an employee of government is prohibited from speaking out on something that is obviously wrong, and then fi red if he does so” (times-mail ). another newspaper wrote, “claude ferguson, whose career in the u.s. forest service apparently ends tomorrow, belongs to the rare breed of government employe [sic] who recognizes that his ulti- mate responsibility is to his conscience and to the public. . . . what confronted mr. ferguson was a confl ict between what he considered the greater public good and a forest service policy” (courier-journal and times ). additional protests were voiced by the sassafras audubon society, the indiana division of the izaak walton league, the indiana conservation council, the nebo ridge study com- mittee, the sierra club, and the student committee of the sassafras audubon society. “it is a question of whether public servants serve the public or their immediate supervisors,” commented phil schrodt of the student committee of the sassafras audubon society (young ) (see inset for ferguson’s clashing obligations.) ferguson explained further, my family rallied around me. my father contacted congress- man richard ichord of missouri, while my daughter wrote the president of the united states. i continued to write policy makers around the country, including congressman john dingell, agriculture secretary robert bergland, and attorney general bell. when questioned by the press, supervisor girton commented, “i guess it goes back to this old adage—if you must condemn and you must criticize externally, i think it gets to the point where the individual had better just resign his position and pursue his course of action—if he feels that strongly about it” (holwager ). in a separate article, girton commented, “employes [sic] have to refrain from directly or indirectly criticizing the rules of the agency” (jor- dan ). later, girton said that ferguson had always been against the idea of orv use in the forest and that he had “disguised” his beliefs so that he could attack the policy (lindley ). it is important to note that around this time, girton publicly admitted to certain trail deviations (joseph a). congressman hayes asked the public integrity section of the u.s. department of justice to review ferguson’s case. he also sent letters of support to john g. mcguire, chief of the forest service; edward h. levi, attorney general, and l. lucius free, assistant director of the offi ce of investigation of the u.s. department of agriculture (indianapolis star ). when asked to comment on hayes’s actions, ferguson was quoted as follows: when the bright morning sunshine reaches the dark corners in this case, i am confi dent that the dust and cobwebs will be- come highly visible to those whose duty is to keep our house of government clean and in good order. (indianapolis star ). in june , ferguson received a “conservation service citation” from the national wildlife federation “for outstanding and dis- tinguished service in the fi eld of natural resource management.” it was accompanied by a citation that described him as “a professional in every respect.” “a trained, experienced, dedicated forester and wildlife manager,” the citation read, “ferguson found himself in a position where he was caught between his own moral convictions and professional responsibilities on one side and a decision from his own agency on the other. he made the hard choice—to follow his own moral and professional dictates.” th e citation continued, claude was enough of a professional to go through all of the prescribed channels to convince his agency that its course of action was wrong. when this failed to produce results, he turned to the people who, in reality, are the owners of the nat- ural resources involved. th is entailed certain risks to claude and his professional career; brought harassment and personal indignities, offi cial censure and personal hardships, ostracism and substantial loss of income and earned retirement rights. claude ferguson laid his career and his professional well- being on the line for what he thought was right, and we have no doubts that history will prove that he was right. th e citation contained two quotations, the fi rst by abraham lin- coln and the second the credo of sigma delta chi, the national pro- fessional journalism fraternity: “to sin by silence when one should protest makes cowards out of men,” and “he serves best who serves the truth.” at that same meeting, ferguson was given a plaque from the indiana conservation council that read, “for outstanding serv- ice, loyalty, and devotion to conservation in indiana— award” (bloomington herald-telephone ). on july , , at the national convention of the izaak walton league of america in baltimore, ferguson received a conserva- tion award for “steadfast devotion to . . . high personal standards as a professional forester and a defender of soil, woods, water and wildlife, without regard for his own personal comfort and economic public administration review • november | december security” (bloomington herald-telephone ). in august , ferguson received an environmental quality award from the u.s. environmental protection agency for being a “citizen activist”; he had been nominated for the award by the bloomington, indiana, chapter of the sierra club. in their letter of nomination, the sierra club members wrote that by protesting illegal orv trails, fergu- son had put his “loyalty to citizen ideals and professional principles above loyalty to the department which he had served faithfully for years” (bloomington herald-telephone ). hearing on the firing of claude ferguson convened as public opinion grew in his favor, ferguson requested that the u.s. civil service commission hearing concerning his fi ring be held in his hometown of bedford. th e hearing was convened in decem- ber by a hearing offi cer from the civil service commission. ferguson’s attorney, david mosier, city judge of columbus, indiana, and president of the indiana chapter of the nature conservancy, had subpoenaed the forest service chief of personnel from wash- ington d.c., as well as the regional forester, the assistant regional forester, and the forest service chief of personnel from milwaukee, among others. ferguson described the proceedings as follows: after consultations with my attorney, family, friends and some colleagues, i had concluded that the best thing for all would be my reinstatement to the full retirement i had requested when i was fi rst notifi ed of my transfer to milwaukee. forest service regulations allowed such action. i was tired and worn out by the battle, but wanted it to end justly and honorably. more importantly, i did not want to harm my coworkers or bring more controversy to the forest service i loved by being reinstated and then working side-by-side with girton and others who had opposed me. i am certain that heads would have rolled if the hearing into my removal went full course. th at i did not want. th ere were a handful that deserved such a “comeupance,” as grandmother would say, but the majority of those who would have been in- jured were good people who were caught between the prover- bial rock and hard place—victims of the bureaucratic squeeze. following the opening of the hearing, my attorney suggested a recess so that he might have a private conversation with forest service chief of personnel whom he had just met. th e recess was granted by the greatly relieved hearing offi cer. david mosier suggested to the chief of personnel that the two of them take a stroll around the block. away they went leaving a room full of very nervous and apprehensive bureaucrats. after perhaps minutes, the strollers returned. th ey approached the hearing offi cer and the three of them talked quietly for a few minutes. th e hearing offi cer then reopened the hearing to announce that the proceedings were completed. david came over to me and announced in a loud voice, “claude, enjoy your well earned retirement and let’s get out of this place. i need some fresh air!” later david told me he had recited just a little of the evidence we were going to present to the chief of personnel and told him of some of the witnesses ready we had ready to testify. th e top forest service offi cials quietly folded their tent and slipped quietly into the night when they came face-to-face with the reality of the facts. my retirement benefi ts and other fringe benefi ts were restored retroactive to the date of my fi ring. the orv court case continues th e orv court challenge, however, continued to drag on. in may , judge william steckler announced that a pretrial conference in the orv case would be held in one month. th e conference was held, and a trial date was set for the fi rst week of december. by mid- october, information leaked to the press indicated that the forest service was ready to settle the lawsuit by withdrawing the entire policy allowing orvs in the hoosier national forest (ellis ; snapp a). at the end of october , the forest service an- nounced that it would indeed reconsider its orv policy, in a move that would settle the orv case before it went to trial (indianapolis star ; snapp b). th e forest service later completely termi- nated its policy of allowing orvs in the hoosier national forest. elated, ferguson held a press conference with the four plaintiff s in the orv case. ferguson said that it had been a “long, costly and heart- breaking three years that need not have been” (snapp c). he also asked for an investigation of some yet-to-be-answered questions: what about the apparent collusion between certain forest service offi cials and offi cials of the american motorcycle as- sociation in generating evidence to be used against me? what about the contempt of forest service offi cials in failing to respond to questions about this matter? what about the ori- gin, timing, and erroneous content of the letters of complaint against me and my wife that were secured and used by the forest service offi cials from offi cers and members of the citi- zens concerned about the nebo ridge area? what about the attempted character assassination attempt against me by one high forest service offi cial by use of part of my medical record in violation of rules, regulations, and laws? (snapp c) almost simultaneously, congressmen morris k. udall and paul simon, along with senator patrick j. leahy, introduced bills in congress that would create review boards on improper govern- mental actions that would decide whether complaints or reports of government employees were made in good faith, and, if so, the employee making it would be protected for two years from being harassed, fi red, demoted, or given a hardship transfer for speaking out. “i’ve been vindicated,” said ferguson. (see inset for a poem given to ferguson by his father as “a code for his self-guidance.”) moral of the story what is the moral of this story? th is administrative profi le vividly conveys the fact that the tensions between bureaucracy and democracy are here to stay and aff ect the jobs of public manag- ers daily. th is case also off ers a darwinian (or evolved) view of the forest service, as it presents two missing pieces of the puzzle concerning the forest service and other public organizations treated only minimally by kaufman. th is administrative profi le vividly conveys the fact that the tensions between bureaucracy and democracy are here to stay and aff ect the job of public managers daily. administrative profi le first, kaufman described forest rangers as “valuing the organization more than they value[d] getting their own way” ( , ). not so in this case. claude ferguson spoke about “the forest service i loved” but nonetheless battled the organization head-on because of his own personal sense of what was right. in this case, ferguson’s obligation to himself and to his interpretation of the public interest trumped his obligation to his organization and profession. public servants do not check their worldviews, their mores, or their ethics at the door. nor can those worldviews, mores, and ethics be easily changed through professional indoctrination. rather, public serv- ants bring with them their own personal opinions, beliefs, desires, and biases. as norton long wrote in , bureaucrats are driven if, by rudyard kipling given to claude ferguson by his father as “a code for his self- guidance” if you can keep your head when all about you are losing theirs and blaming it on you; if you can trust yourself when all men doubt you, but make allowance for their doubting too; if you can wait and not be tired by waiting, or, being lied about, don’t deal in lies, or, being hated, don’t give way to hating, and yet don’t’ look too good, nor talk too wise; if you can dream—and not make dreams your master; if you can think—and not make thoughts your aim; if you can meet with triumph and disaster and treat those two imposters just the same; if you can bear to hear the truth you’ve spoken twisted by knaves to make a trap of fools, or watch the things you gave your life to broken and stoop and build ’em up with worn-out tools; if you can make one heap of all your winnings and risk it on one turn of pitch-and-toss, and lose, and start again at your beginnings and never breathe a word about your loss; if you can force your heart and nerve and sinew to serve your turn long after they are gone, and so hold on when there is nothing in you except the will which says to them: “hold on” if you can talk with crowds and keep your virtue, or walk with kings—nor lose the common touch; if neither foes nor loving friends can hurt you; if all men count with you, but none too much; if you can fi ll the unforgiving minute with sixty seconds’ worth of distance run yours is the earth and everything that’s in it. and—which is more—you’ll be a man my son! by their own parochial views, interests, and values. ferguson, like many public servants, brought a very personal sense of right and wrong with him to the forest service. second, kaufman described one rationale of the forest service’s eff orts to routinize the decisions of its employees in an eff ort to prevent allegiances to, or co-optation by, local populations. yet in this administrative profi le, an opposite phenomenon can also be seen: ferguson’s actions behind the scenes included strategic eff orts to co-opt the local population in order to enlist support for a cause he deeply felt was right and just. th is was done by forging alliances with interest groups, issuing press releases, writ- ing letters to politicians, and encouraging newspaper articles in his favor. in today’s era of open systems and networks spanning many organizations and including many people, this phenom- enon is likely to grow. references bloomington (indiana) herald-telephone. . forester ferguson on epa “honor roll.” august . courier-journal and times (louisville, kentucky). . claude ferguson: a whistle-blower gets the axe (editorial). february . dustin, th omas e. . letter to j. carl ferguson, july . ellis, leslie. . off -road vehicles facing ban in hoosier forest? courier journal, october . holwager, joe. . forester fired in vehicle trails dispute. louisville courier-journal, february . indiana conservation council. . newsletter, october, . indianapolis star. . hayes is requesting review of ferguson case. february . ———. . forest service restudies off -road vehicle policy. october . jordan, don. . suit to be filed by isaac [sic] walton league. times- mail (bedford, indiana), october . ———. . girton defends ferguson firing: “employes [sic] shouldn’t criticize.” bloomington herald-telephone, february . joseph, gloria. a. forester fired: supervisor admits to trail devia- tions. indiana daily student, february . ———. b. forest service worker fired. indiana daily student, february . kaufman, herbert. . th e forest ranger: a study in administrative behavior. baltimore: johns hopkins university press. lindley, tom. . forester embroiled in controversy. evansville sun- day courier and press, february . long, norton e. . power and administration. public administration review ( ): – . national audubon society, sassafras, indiana chapter. . commen- dation of a. claude ferguson. national wildlife federation. . citation honoring a. claude ferguson. o’leary, rosemary. . th e ethics of dissent: managing guerrilla government. washington, dc: cq press. snapp, ray. . ferguson fired by u.s. forest service. times-mail (bedford, indiana), february . ———. a. attorney has no comment in orv policy case. times- mail (bedford, indiana), october . ———. b. forest service withdraws off -road-vehicle policy. times-mail (bedford, indiana), october . public administration review • november | december ———. c. forester applauds orv policy withdrawal; says questions still unanswered in his case. times-mail (bedford, indiana), november . times-mail (bedford, indiana). . ferguson firing (editorial). february . u.s. department of agriculture. – . forest service guidance on planning. ———. . investigation report, june . webb, carl e. . memorandum to donald girton, april . young, jack. . ferguson firing. times-mail (bedford, indiana), february . foundations of public administration let par be your guide to discovering the treasures in the ongoing and historical work of public administration…. th e foundations of public administration series is a collection of articles written by experts in content areas, providing introductory essays and recommending top articles in those subjects. introducing: american federalism, intergovernmental relations, and intergovernmental management by: deil s. wright and carl w. stenberg university of north carolina at chapel hill and by: chung-lae cho ewha womans university th e foundations series for human resources also includes a sample syllabus and select readings from par on american federalism, intergovernmental relations and intergovernmental management login at http://www.aspanet.org and click on “public administration review”, and ‘foundations of public administration’. select ‘american federalism, intergovernmental relations and intergovernmental management’ wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ reviews rural fictions, urban realities: a geography of gilded age american literature by mark storey (new york: oxford university press, . pp. viii, . notes, works cited, index. $ . .) when caroline meeber steps onto the train station platform in chicago at the beginning of theodore dreiser’s sister carrie, readers are clearly meant to understand that she is taking a step forward—in space, but also in time. true, dreiser’s novel then goes on to equivocate for more than five hundred pages over the question of whether moving forward necessarily consti- tutes a move for the better. but there is never really any doubt about what carrie’s initial journey from small- town wisconsin to the heart of the windy city is intended to represent. put simply, carrie’s arrival in chicago symbolizes her arrival at the doorstep of american modernity. such grand metaphorical gestures served dreiser and a number of his contemporaries extraordinarily well. not only do we continue to read their books today, we cite them as exemplars of literary “realism” precisely because we see in “urban” novels like sister carrie the kind of dark and disen- chanted world that weberian theorists of modernity promised we would. but as mark storey rightly notes in rural fictions, urban realities, modernity touched virtually everything about american life during the gilded age—including the pages of “rural fiction,” a rich and yet largely unap- preciated genre of writing that storey persuasively argues was anything but provincial in its character or concerns. on the contrary, storey finds in the pages of american late nine- teenth- and early twentieth-century rural fiction many of the same themat- ic preoccupations typically associated with the most avowedly metropolitan of gilded age novels. he simply finds them off the beaten path: on the train as it speeds across the rolling coun- tryside; at the circus; on the road with a country doctor; amidst the lynch mob; even in the middle of nowhere, if you take storey’s thoughtful recon- much earlier. janney’s impressively researched and comprehensive book, spanning more than seven decades, indicates that civil war memory will continue to be a rich and contested field of study for many years to come. kathleen clark is associate professor of history at the university of georgia and the author of defining moments: african american commemoration and political culture in the american south, - ( ). i n d i a n a m a g a z i n e o f h i s t o r y sideration of american utopianism at its enigmatic word. on some level, the persuasiveness of storey’s argument depends on his scrupulously close reading: his eye catches everything, down to the me- chanical gait of a plow horse laboring under the combined metaphorical weight of both garden-variety pastoral- ism and incipient industrial capitalism. but the book also functions as well as it does because of storey’s innovative and unexpected take on the subject of american modernity, an approach that simultaneously challenges and corroborates much of what we already claim to know. on one hand, it is hardly obvious that novels about late nineteenth-century rural and small- town life would prove to be “a vital entry point to the themes, concerns, currents, and transformations of urban modernity,” as storey persua- sively demonstrates (p. ). but then again, how could any artifact from the gilded age have remained innocent of modernity? the answer is that none did, and rural fictions, urban realities demonstrates this beautifully. ida mckinley: the turn-of-the-century first lady through war, assassination, and secret disability by carl sferrazza anthony (kent, ohio: kent state university press, . pp. . notes, bibliography, index. $ . .) comparative analyses of first ladies routinely dismiss the influence of ida saxton mckinley ( - ), por- traying her as inconsequential during her husband william mckinley’s ten- ure as president from until his assassination in . while mystery shrouded her medical condition, the readers with a particular interest in indiana history will appreciate storey’s thoughtful engagement with a number of hoosier authors, including edward eggleston, booth tarkington, and maurice thompson, all of whom receive far less attention from serious critics than they deserve. but even readers who come to his book with a specific agenda will admire the sheer number of american authors storey is able to draw together by way of his adroit analysis. in sum, rural fic- tions, urban realities is a smart book, and one that is argued with an uncom- mon degree of care and sensitivity—a decidedly modern reconsideration of a misunderstood body of literature that has remained at the periphery of debates about american modernity. colin r. johnson is associate profes- sor of gender studies and adjunct associate professor of american studies, history and human biology at indiana university bloomington. he is the author of just queer folks: gender and sexuality in rural america ( ). this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. understanding st. louis’ love for hoosier daniel duncan department of linguistics, new york university, new york, usa while the american english demonym hoosier refers to indiana residents, it means ‘poor, rural, white trash’ in st. louis, missouri (murray, ). this paper uses discourse analysis of several texts across a range of registers and formalities to explore why its use persists despite less- localized alternatives (redneck, etc.) and why it has become enregistered (agha, ) as a feature of the local dialect. findings show hoosier is used to police the behavior of the target. unlike similar slurs, its use requires knowledge of st. louis’ social geography. hoosier allows speakers to demonstrate localness while positioning themselves and st. louis as cosmopolitan compared to the derided target. as such, hoosier asserts positive values for st. louisans who use it. keywords slurs, enregisterment, localness, demonyms, st. louis introduction in general american english, a common demonym for residents of indiana is hoosier. indiana is known as the hoosier state (metcalf, ), and the indiana university athletic teams are known as the hoosiers. it is unclear what the exact etymology of the demonym is, although graf ( ) summarizes several theories—among them, that the term derives from the phrase ‘who’s here?’ or a regional dialect term meaning ‘rustic, country bumpkin.’ the latter possibility appears, however, to be derived from the demonym, rather than providing a source. as will be discussed in this paper, hoosier is commonly used in the local dialect of st. louis (stl), missouri, as a slur that means ‘poor, southern, rural, white trash.’ according to the dictionary of this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. american regional english, the use of hoosier to mean ‘a rustic or countrified person’ is concentrated in missouri, with / speakers using this term native to the state, and the rest scattered across the united states (cassidy, ). this suggests that the st. louis usage may have been highly localized for some time. etymologists local to st. louis claim the usage derives from union struggles during which non-union workers from indiana were brought to the city during strikes (mcgonigle, ). however, this is as unverifiable as the source of the demonym itself. the focus of this paper will be on the use of hoosier in st. louis, rather than its etymology. hoosier has become enregistered (agha, ), that is, given social meaning, as a feature of stl english, at least among middle/upper-middle class whites. for these groups, hoosier strongly indexes stl. such enregisterment is seen, for example, in hoosier’s prominent position in the title of merkel’s ( ) local-interest book hoosiers and scrubby dutch: st. louis’s south side. use of hoosier is salient enough that sociolinguistic studies of stl often refer to it as a motivation for phonological changes to the dialect (murray, ; murray, ; goodheart, ). in this view, phonetic features that hoosiers use are stigmatized enough that speakers actively avoid them. this focus overlooks two related questions. first, why would stl need hoosier when other slurs with similar connotations (hick, redneck, etc.) are available? secondly, why would this linguistic feature become enregistered in the first place? in her study of how the enregisterment of dialect features local to pittsburgh has led to the commodification of those same features as ‘pittsburghese,’ johnstone ( ) observes a sense of local pride in them. most pittsburghese features, like monophthongized /aʊ/, as in downtown (johnstone and kiesling, ), index local identity while remaining neutral in semantic content; however, these features this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. also include the slur jagoff. like jagoff, hoosier is certainly not neutral—why would a slur inspire local pride? this paper probes the use and enregisterment of hoosier. middle/upper-middle class whites, who are the speakers who use and are most affected by a classist slur, are the focus. discourse analysis of four texts will show that to use hoosier appropriately, the speaker must be intimately familiar with stl. speakers use the slur to police behavior in the area, casting stl as a cosmopolitan city that is cultured and sophisticated in comparison to its surroundings. besnier ( ) observes that although ‘locality’ is frequently defined in opposition to traits like ‘cosmopolitanism’, these are combined as often as they are not. in his study of performances in the transgender miss galaxy pageant in tonga, he finds that use of english is used to position oneself as cosmopolitan, within both local and non-local contexts. similarly, it may be argued that the combination of locality and cosmopolitanism enables the enregisterment of hoosier. while derogatory toward the addressee of the slur, hoosier reflects a positive identity and set of values toward the city. methodology hoosier is treated here as a key expression (see duranti, ) whose study may lead to a broader insight into stl. the context, intended meaning, and work performed by the term is examined within several texts. blommaert’s ( ) theorizing of sociolinguistic scales, in which layered complexity in linguistic features is related to differences between levels of scales, will be used to flesh out the lexical meaning of the slur. particular attention will be placed on who or what is being targeted by hoosier, both in terms of identity and message. at the same time, it will this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. be useful to note what the speaker is conveying about her- or himself in addition to his or her target. as hoosier is ubiquitous within stl, four text types were chosen to reflect a range of registers and modes of discourse: . a sociolinguistic interview with mary s., a highly educated young white woman who grew up in a suburb of stl, represents casual conversation. this conversation centered on mary’s attitudes toward stl. . sixty-three articles, representing formal writing, from the st. louis post-dispatch, the main newspaper of the region, and its subsidiary papers were collected. archives are available from - and -present. all uses of hoosier are from the latter archive, appearing regularly in opinion columns and quotes. given the genre of newspaper writing, even columns and quotes constitute a formal environment. . the song ‘hoosier love,’ by the local ska-punk band mu , represents a text that is performed and mediated. that is, such speech is performed for an audience and may be highly stylized as a result. . the radio segment headline hooshe ([huʒ]), a derisive ‘weird news’ segment that appears daily on the local alternative rock station, is a performance as well, albeit less planned. like musical artists, the radio hosts are performing for their audience. on monday through thursday, the segment mocks one story from around the united states. the friday edition (analyzed here) recaps the week’s stories and solicits listener votes on the most hoosier story. in this way, the performance is somewhat interactive: while only the hosts give their opinions on air, the listening audience’s collective opinion is also given representation. this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. the newspaper articles were coded for explicit references to hoosier’s definition, specific locations, etc. the person, thing, or action described as hoosier was also labeled. the other texts were transcribed and the sections surrounding use of hoosier were extracted for analysis. st. louis and its vicinity in focusing on the use of hoosier in discourse, it is helpful to consider the regional context in which it appears. stl and the rest of missouri have traditionally maintained a strong urban/rural divide. for example, in goodheart’s ( ) study of phonological change in the st. louis area, she finds that most of her interviewees identify with stl, but not the rest of missouri. in the sociolinguistic interview conducted for the current study, mary s. illustrates this when she describes where she would say she was from if asked. note that she never claims to say she is from missouri: mary: generally i say st. charles unless i’m out of state and then i just say st. louis […] it’s all respective to where i am at the moment. like if i am out of the country i say the us. if i’m you know, over in new jersey, i tend to say st. louis. in some respects, this suggests that for st. louisans, the urban/rural divide is such that ‘missouri’ is a place that is exclusive of stl. the divide has persisted despite dynamic change along the urban/rural border. suburbanization, spurred by ‘white flight’, has drastically reshaped the landscape in the metropolitan area surrounding stl. as a result, formerly rural towns and counties are now well-developed regions of urban sprawl. this process decimated the city itself and resulted in stl’s population decreasing by % since (see gordon, ). this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. results the previously described connotations of hoosier are often made explicit in the post-dispatch (table ). as such, issues of class, race, and regional identity are wrapped together by definition in hoosier. table . explicit references to hoosier's meaning in st. louis post-dispatch articles explicit description articles referring rural white trash poor synonymous slur redneck although intersectional, these properties may be accessed relatively independently from one another. it is thus useful to conceive of the meaning of hoosier using the family-resemblance approach to category membership that croom ( ) suggests be applied to slurs. that is, an item is a member of a category (c) if it shares some properties (pn) with a prototype. for hoosier, this yields the following: c. h(hoosier) p . x is poor (poverty) p . x is from a rural area (rurality) p . x is southern (southerness) p . x is white (whiteness) p . x is trashy (trashiness) this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. under this conception, if some nominal x may be described with some or all of the above properties, it may be described as hoosier. that some properties are highlighted at the expense of others is illustrated in the headline hooshe segment, which emphasizes southernness. in it, patrico, one of the radio hosts, imitates a southern accent through /aɪ/ monophthongization: moon: i have this visual picture of these two in a cop car speedin’ away and him goin’ ‘yeah amber, whoo-hoo!’ patrico: yeah and her goin’ ‘[a] got you baby! [a] love you so much!’ rurality is also easily accessed, as the conversation with mary reveals: dan: what’s your opinion of the more rural parts of the area? mary: like the hoosier parts? d: yeah. [laughs] m: does that sum up my ideas? […] oh rural missouri. there’s corn and racism and tornadoes. and i don’t want to be there. of course, the above excerpts do not purely refer to one property of hoosier, but multiple properties. accessing these properties, as shown below, is not done haphazardly. rather, the properties of x necessary to be labeled hoosier vary with the locality of x. this is interpreted here in terms of sociolinguistic scales. in an effort to help sociolinguists better analyze language use in a globalizing society, blommaert ( ) theorizes that linguistic phenomena display layered complexity as one moves between levels of a scale. such a scale ranges from local/micro levels to global/macro levels. blommaert’s insight is that shifting meanings and indexical relations of linguistic phenomena are accompanied by a shift in level on a scale. that is, at any particular level of a scale, the meaning is more fixed. using a scale based on locality, we find that hoosier’s meaning displays the layered complexity discussed by blommaert. the relevant this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. scale here has the following levels: stl  missouri  the us. within each level, different properties of hoosier’s meaning are highlighted or diminished. consider mary’s use of it with respect to missouri. here, rurality is strongly brought to the forefront. because mary knows that rural missouri is predominantly poor, white, and southern, these properties are secondary. compare this with the headline hooshe segment, whose stories come from around the us and generally involve low-level criminal acts. as such, the foremost properties of hoosier at this level are poor trash. because these negative properties index southernness among the hosts and audience, this is secondary and made explicit as a result. when used at the level of stl, hoosier again highlights poor trash; however, the secondary meaning changes. this is due to where in stl hoosiers are described as being from. ‘hoosier love,’ for example, refers to “south side city hoosier love” in the refrain. similarly, table shows that the most prominent place explicitly attached to hoosier in the st. louis post-dispatch is south stl. this area is predominantly white due to hypersegregation (see massey and denton, ), and speakers know this. thus, hoosier takes on white as a secondary meaning with respect to stl. table . place associations of hoosier in st. louis post-dispatch articles referent articles referring south st. louis missouri arnold, mo. dong and blommaert ( : ) observe that moving across levels of a scale “presupposes access to particular resources, and such access is often subject to inequality”. we see this especially when moving to the level of stl. the speaker needs a detailed knowledge of the this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. social geography of stl. it is not enough to know that hoosiers are poor and white; one has to know where poor whites live in stl in order to wield hoosier appropriately. in essence, this means only a st. louisan can use hoosier appropriately. therefore, use of hoosier represents a claim of being a st. louisan. normative use any stigmatized thing, behavior, or preference can be described as hoosier (table ). as such, the social judgments in these uses of hoosier serve a normative function. table . non-personal stigmatizing uses and examples of hoosier in st. louis post-dispatch articles modifies examples articles referring thing hoosier weeds; hoosier hot tub behavior leaving christmas tree up all year; supporting george w. bush for president preference hoosier rock; selecting branson, mo for vacation spot it appears that this normative function of hoosier extends to policing inappropriate behavior. the headline hooshe segment illustrates this when the hosts debate which of four stories is the most hoosier: . a woman stole a police vehicle while her husband was being arrested for a dui. . a man wrapped his home in tinfoil to protect himself from aliens. this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. . one man threw a barbequed beef brisket at another during an argument. . a woman attacked her boyfriend with a poop scoop during an argument. the hosts’ debate centers on whether some stories are actually hoosier: jeff: to me it’s easy, it’s number one for me because you have the dui, you get arrested, your wife comes to free you, and steals a cop car? that’s hoosier all over the place, number two there’s just something going on in the guy’s head with the foil-wrapped house. and uh, number three and number four, i mean, they’re just arguments to me. that’s just my opinion. moon: two horse race for me, uh, between the uh, the alien guy because not only is that hooshe, but he’s made his entire neighborhood hooshe now j: yeah but i think he’s just something wrong in his head. rizzuto: if somebody, if one person could lower the property value in a neighborhood. if one person and it’s a-, he foil-wrapped his house cause he’s afraid of an alien attack that gets my vote. m: that’s pretty good. but i am actually voting for number one. r: i thought that’d be in last place. patrico: no honestly to me that’s what i would go for. strikingly, they debate whether the man’s actions in ( ) are actually hoosier. as moon notes, decorating one’s house such that property values decline is indeed socially unacceptable behavior. however, jeff claims that such unacceptable behavior is excused, as it results from mental illness. in other words, only one who intentionally and competently engages in socially unacceptable behavior can be hoosier. the man in ( ) was not capable of this. this argument leads moon and patrico to join jeff in believing story ( ) to be the most hoosier. the audience this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. agreed when given input; story ( ) received a plurality of votes. in short, it is not merely the stigma of certain behaviors that makes them hoosier, rather, one’s behavior is hoosier if it can be corrected. in this way, use of hoosier polices behavior in stl. in this light, ‘hoosier love’ may be seen as a cautionary tale when the narrator, speaking as a hoosier, proclaims, we don’t need no high school i think we’re too cool we’ll have kids at seventeen getting laid at dairy queen these behaviors, the band says, are unacceptable. as such, the song not only makes fun of people who engage in such activities, but warns the listener not to as well. a claim to cosmopolitanism thus far, use of hoosier appears to fall into the first of three categories of use—paradigmatic derogatory use, non-paradigmatic derogatory use, and non-derogatory in-group use—that croom ( ) distinguishes for slurs. this section problematizes such categories by considering what usage represents with respect to the speaker. although hoosier is clearly used in a derogatory manner that others the intended target, by derisively asserting that behavior is socially unacceptable, hoosier also asserts that the user conducts his- or herself in a socially acceptable manner. it seems that the uses of hoosier described above promote the speaker as cosmopolitan in comparison to the derided target. consider hoosier’s use in policing preferences, for example. a speaker maligning hoosier music, for instance, presumably likes and listens to non-hoosier music. as such, using hoosier indexes the speaker’s own claim not just to acceptable behavior, but acceptable taste. a cosmopolitan person is the opposite of a hoosier. use of hoosier, then, is this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. a performance which “create[s] images of the self and the other” (pagliai, : ). by recognizing and pointing out hoosier behavior or taste, the user elevates their own social standing. this function of the slur is not one described in croom’s ( ) categorization of slurs. another way to assert that one is cosmopolitan through use of hoosier is to deny that one is a hoosier. this is far more dangerous for the speaker, because a vehement denial gives the impression the speaker is defending some behavior that is likely hoosier. done well, it conveys that one is indeed cosmopolitan. however, failure to substantiate the claim risks the speaker’s being branded a hoosier. two post-dispatch articles illustrate this. in the first, a well-off neighborhood in stl’s suburbs drafted regulations targeting a new resident, who owned two pit bulls among smaller annoyances such as not having garbage cans of the same color as his neighbors (shinkle, ). in the course of being interviewed for the piece, the resident claimed that he was not a hoosier. however, in the context of the piece, such an assertion merely draws attention to the list of his perceived misdeeds, inviting the reader to evaluate this claim themselves. by contrast, a column discussing the growth and development of stl’s suburbs in st. charles county, missouri is more successful at claiming cosmopolitanism. while the author begs the reader to “spare us the hoosier treatment, and stop telling us to put the rifle in the rack in the truck and clean the dog crap off our boots” (sonderegger, ), he also makes an effort to do extra work at asserting cosmopolitanism by offering several ways in which he believes the county is cosmopolitan. what these examples show is that unlike how directing hoosier at a target constitutes a speaker’s unassailable claim to cosmopolitanism, denying that one is hoosier requires proof of cosmopolitanism. even when offered, there is no guarantee that the proof will be accepted—and therein lies the danger for the speaker. this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. one final use of hoosier worth considering here is its role in self-deprecation. speakers will sometimes claim to be a hoosier themselves or profess to like hoosier things. a reasonable understanding of this would be to read self-described hoosiers as constituting a counter-public (warner, ) in which such speakers recognize their subordinate status, yet at the same time communicate an anti-cosmopolitan identity. such use of hoosier might then be treated as non- derogatory in-group use of the slur (croom, ) and analyzed as an instance of appropriation (following, perhaps, the analysis proposed in bianchi, ). i suggest, however, that this is not the case. speakers claiming to be hoosiers in the st. louis post-dispatch include a local artist, the editor of the newspaper’s lifestyle section, and a tenured archaeology professor. far from being hoosiers, these speakers are clearly cosmopolitan. as such, speakers who claim to be hoosiers or admit to hoosier behavior are not members a counter-public and engaging in non- derogatory in-group use. rather, their occupations suggest that they are authoritatively cosmopolitan. instead of appropriating the slur, these speakers are claiming that they are so cosmopolitan that they are allowed to engage in hoosier-like behavior. discussion we have seen that hoosier is derogatory, layered, and normative. what does this tell us about stl and hoosier’s persistence and enregisterment? consider that hoosier strongly indexes localness. using it correctly involves managing the complex intersectionality of several stigmatized groups across a scale of regionalness. doing so requires intimate knowledge of the social geography of stl. the user then must be a native st. louisan or long-lasting transplant. as such, hoosier is tied to locality in stl more deeply than a second- or third-order indexical relation, which johnstone and kiesling ( ) describe as arising from speakers correlating a this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. linguistic feature with a place. rather than merely correlate with stl, hoosier encodes stl into its pragmatic content. in this way, hoosier acts as a verbal symbol (cavanaugh, ) of stl. this suggests that hoosier persists because its seemingly viable alternatives are not as viable as they first appear. the locality encoded in hoosier also impacts its normative use. through its encoded locality, middle/upper-middle class whites use hoosier to proclaim what unacceptable behavior is for a st. louisan. likewise, it asserts that the speaker is or wants to be perceived as a cosmopolitan st. louisan. implicit here is the claim that stl itself is a cosmopolitan city and as such, use of hoosier makes a proud, positive claim about stl. for its users, hoosier is thus a verbal symbol of the local, associated with positive values. while derogatory, it is used proudly to strengthen the speaker’s identity. this differs from the reclamation or appropriation of slurs for in-group use (see croom, ; bianchi, ; inter alia), as it remains derogatory. this observed use and valuation of hoosier may explain its enregisterment, but also raises the question as to why stl, or any group within it, needs to assert itself as cosmopolitan. here it is useful to consider stl’s relationship with the surrounding area. the urban/rural divide persists despite the fluid urban/rural border introduced by suburbanization and sprawl, which introduces new questions for residents: who is urban? who is rural? how do the suburbs fit in? modan ( ) observes that the changing social geography of a washington, dc, neighborhood triggers the contestation of place in the neighborhood. in particular, she notes how the moral geography, or linking of a moral framework to a geographical area, of the neighborhood, is contested. she observes how values are assigned to the “city” vs. the “suburb,” and how residents creatively position themselves in this moral space through discourse. the changing landscape and social geography of the stl metropolitan area may be viewed similarly. this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. this relates to the need that middle/upper-middle class white st. louisans feel to position themselves as cosmopolitan. at the turn of the twentieth century, stl was the fourth largest city in the us, meaning its sophistication, grandeur, and urban values went unchallenged. this has dramatically changed: the city has roughly , residents and is not even the largest city in missouri. it is an open question, then, whether stl has maintained its prior status. their use of hoosier indicates that this is not an idle question for middle/upper-middle class white st. louisans, who are fighting to claim relevance and sophistication which they once took for granted. in discussing how the bergamasco dialect of italian is perceived as uneducated and hardworking despite the middle- class status of many bergamascos, cavanaugh ( : ) observes, “a bergamasco accent can also symbolize the disjuncture between who bergamascos think they used to be, and who they think they are—and want to be—now.” we see the same disjuncture in stl’s use of hoosier. at the same time, the blurred urban/rural distinction makes it less clear who a st. louisan is, now that once rural towns could possibly stake a claim to urbanity. as such, stl is wrestling with the politics of place and who an authentic community member is. hoosier offers a way to separate central members from more marginalized members (modan, ): it asserts who is and is not an authentic st. louisan, and proclaims that there are standards to achieving this. it gives cosmopolitan suburbanites a way to become viewed as authentic st. louisans, while writing off those who do not conform. conclusion in many respects, this paper explores why hoosier is, as a st. louis post-dispatch editorial observed (editorial board, ), “the last socially acceptable classist thing you can say” in this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. stl. the study focuses on how middle/upper-middle class white speakers use hoosier. this is important to emphasize; the approach taken above lends itself to overgeneralization when done carelessly (duranti, ), and as such, while broader insights may be drawn from how these speakers use the slur, the results may not extend to other groups. as has been shown through its use in discourse, for the group in question, locality is encoded into hoosier. despite its derogatory, normative use, hoosier asserts a positive identity and values for stl. given this, it is not surprising that locals have such an affinity for it. depletion of the urban core through ‘white flight’ to the suburbs and urban sprawl in general changed the social geography of stl, and residents are in a position of grappling with the politics of place. as long as that continues, we can expect hoosier to maintain its salient position in local speech. notes . this is not meant to suggest this is the only possible scale, nor, for that matter, that the relevant scale for hoosier only has these three levels. one could conceive of the scale having more local levels—the neighborhood, for instance—or more global levels. the given levels are those that emerge based on the texts analyzed in this paper. . this, of course, is putting aside the fact that the st. louis area has a large african american population, in addition to other minorities. because hoosier predominantly concerns whites, i have done so as well in this paper. however, the events in the suburb of ferguson show that such issues in st. louis extend far beyond class or an urban/rural divide. ethnicity is clearly one more arena in which st. louis is wrestling with these questions. acknowledgements this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. this paper was developed in a seminar on language, place and space taught by rudi gaudio, and was presented at the ans annual meeting. comments from isaac bleaman, olga verlato, mary robinson, the nyu sociolinguistics lab, and two anonymous reviewers were especially helpful in development. the author also wishes to thank the ans emerging scholar award committee for their helpful comments, and particularly jan tent for his role in ushering the paper along. bibliography agha, asif. . “the social life of cultural value.” language & communication : - . besnier, niko. . “transgenderism, locality, and the miss galaxy beauty pageant in tonga.” american ethnologist : - . bianchi, claudia. . “slurs and appropriation: an echoic account.” journal of pragmatics : - . blommaert, jan. . “sociolinguistic scales.” intercultural pragmatics ( ): – . cassidy, frederic. . dictionary of american regional english. cambridge, ma: harvard university press. cavanaugh, jillian r. . “accent matters: material consequences of sounding local in northern italy.” language & communication ( ): - . croom, adam m. . “how to do things with slurs: studies in the way of derogatory words.” language & communication : - . dong, jie, and jan blommaert. . “space, scale and accents: constructing migrant identity in beijing.” multilingua : - . duranti, alessandro. . “indexical speech across samoan communities.” american anthropologist ( ): - . editorial board. . “happy anniversary, you old dog!” st. louis post-dispatch, june . goodheart, jill c. . “i’m no hoosier: evidence of the northern cities shift in st. louis, missouri.” ma thesis, michigan state university. gordon, colin. . mapping decline: st. louis and the fate of the american city. philadelphia: university of pennsylvania press. this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. graf, jeffrey. . “the word hoosier.” accessed february , . http://www.indiana.edu/~librcsd/internet/extra/hoosier.html johnstone, barbara. . “pittsburghese shirts: commodification and the enregisterment of an urban dialect.” american speech ( ): - . johnstone, barbara & scott f. kiesling. . “indexicality and experience: exploring the meanings of /aw/-monophthongization in pittsburgh.” journal of sociolinguistics ( ): - . massey, douglas s., and nancy a. denton. . “hypersegregation in us metropolitan areas: black and hispanic segregation along five dimensions.” demography ( ): - . mcgonigle, pat. . “here’s why ‘hoosier’ is an insult in st. louis.” ksdk. accessed november , . http://www.ksdk.com/story/homepage/ / / /historiansetymologistspointtotransplan tedindianaworkersatchrysleranheuserbusch/ / merkel, jim. . hoosiers and scrubby dutch: st. louis’s south side. st. louis: reedy press. metcalf, allan. . “hoosiers, suckers, badgers, and wolverines.” lingua franca. accessed january , . http://www.chronicle.com/blogs/linguafranca/ / / /hoosierssuckersbadgersandwol verines/ modan, gabriella g. . turf wars: discourse, diversity and the politics of place. malden, ma: wiley-blackwell. mu . . “hoosier love.” press. monte sereno, ca: asian man records. murray, thomas e. . the language of st. louis, missouri: variation in the gateway city. new york: peter lang. murray, thomas e. . “‘you $#^%?*&@ hoosier!’: derogatory names and the derogatory name in st. louis, missouri.” names ( ): - . pagliai, valentina. . “lands i came to sing: negotiating identities and places in the tuscan contrasto”. in sociolinguistics: the essential readings. ed. c. b. paulston & g.r. tucker, - . malden, ma: blackwell. shinkle, florence. . “westwood’s new ordinance on dogs raises hackles and heated disputes.” st. louis post-dispatch, february . sonderegger, john. . “take a look around—we’re all grown up and exceeding expectations.” st. charles county post, april . warner, michael. . “publics and counterpublics (abbreviated version).” quarterly journal of speech ( ): - . notes on contributor http://www.indiana.edu/~librcsd/internet/extra/hoosier.html http://www.ksdk.com/story/homepage/ / / /historiansetymologistspointtotransplantedindianaworkersatchrysleranheuserbusch/ / http://www.ksdk.com/story/homepage/ / / /historiansetymologistspointtotransplantedindianaworkersatchrysleranheuserbusch/ / http://www.chronicle.com/blogs/linguafranca/ / / /hoosierssuckersbadgersandwolverines/ http://www.chronicle.com/blogs/linguafranca/ / / /hoosierssuckersbadgersandwolverines/ this is a pre-publication draft; the article has been accepted for publication in names: a journal of onomastics. daniel duncan is a ph.d. candidate specializing in sociolinguistics in the department of linguistics at new york university. his dissertation concerns the relation of suburbanization to language variation and change, using the st. louis metropolitan area as a case study. correspondence to: daniel duncan, new york university, department of linguistics, washington place, new york, ny, , usa. email: dad @nyu.edu density, frequency and the expressive phonology of children with phonological delay* judith a. gierut and michele l. morrisette indiana university, bloomington abstract the effect of word-level variables on expressive phonology has not been widely studied, although the properties of words likely bear on the emergence of sound structure (stoel-gammon, ). eight preschoolers, diagnosed with phonological delay, were assigned to treatment to experimentally induce gains in expressive phonology. erred sounds were taught using stimulus words that varied orthogonally in neighborhood density and word frequency as the independent variables. generalization was the dependent variable, defined as production accuracy of treated and untreated (erred) sounds. blocked comparisons showed that dense neighborhoods triggered greater generalization, but frequency did not have a clear differential effect. orthogonal comparisons revealed graded effects, with frequent words from dense neighborhoods being optimal for generalization. the results contrast with prior literature, which has reported a sparse neighborhood advantage for children with phonological delay. there is a suggestion that children with phonological delay require greater than usual cue redundancy and convergence to prompt expressive phonological learning. introduction contemporary research has convincingly demonstrated that statistical regularities in the input are available to, and used by, children as a possible bootstrap to language acquisition (e.g. garlock, walley & metsala, ; jusczyk, luce & charles-luce, ; saffran, aslin & newport, ). reliable input cues have been isolated at the level of the word and also at the level of the sounds that make up the word. thus, dual sources of information, deriving from lexical (word) and sublexical (phonological) structure, contribute to the language acquisition process. in terms of lexical structure, two widely studied variables are neighborhood density and word frequency. density estimates the phonetic similarity of word forms based on one- phoneme substitutions, deletions or additions (luce, ), whereas frequency estimates the occurrence of a given word in the language. in terms of sublexical structure, the variable that has received most attention is phonotactic probability, which is defined as the likelihood of occurrence of a given sound and/or sound sequence in the language (vitevitch & luce, ). to date, research has largely focused on these lexical and sublexical variables in the context of children’s acquisition and recognition of new words, with primary evidence coming from spoken word perception (e.g. metsala, ; *this research was supported in part by a grant from the national institutes of health (dc ) to indiana university. we thank dan dinnsen and jill hoover for their input, and members of the learnability project for their assistance in data management and analysis. © cambridge university press address for correspondence: judith a. gierut, department of speech and hearing sciences, indiana university, south jordan avenue, bloomington, in - , usa. gierut@indiana.edu. nih public access author manuscript j child lang. author manuscript; available in pmc september . published in final edited form as: j child lang. september ; ( ): – . doi: . /s . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript storkel, ; swingley & aslin, ). this line of inquiry has generally supported frequent patterns (words or sounds) and dense neighborhoods as most influential to children’s word learning. in an invited review recently published in jcl, stoel-gammon ( ) brings to the forefront an alternate, but complementary perspective. specifically, she argues that the same regularities that assist children in word learning also assist their acquisition of expressive phonology. she reasons that, if word learning capitalizes on cues at dual levels of structure, then it is likely that the effects are bidirectional : words and sounds converge to enable word learning and, likewise, phonological learning. studies of the effects of statistical regularities on expressive phonology are emerging (beckman & edwards, ; munson, ), but preliminary results suggest that children’s sensitivity to cues for production might not match those for perception (stoel-gammon, ). while asymmetries between production and perception have long been recognized in language acquisition (gerken, ), stoel- gammon ( ) offers a possible reconciliation within this contemporary context. she notes that, although children might perceptually attune to, and discriminate statistical regularities associated with lexical and sublexical structure, they might not have specified and/or organized that same information to the level or degree needed to support accurate production. the apparent asymmetries and the paucity of research prompted stoel- gammon’s ( ) call for further study of statistical variables relative to expressive phonology. the present article responds to this call by evaluating density and frequency in phonological learning by children with phonological delay (pd) enrolled in clinical treatment. density and frequency were examined for insight to the ways that lexical variables impact sublexical learning, thereby honing in on word-to-sound influences. children with pd were of interest given methodological and theoretical advantages inherent to their study. clinical treatment afforded the experimental manipulation and served as the platform for inducing change in the phonologies of the children. by way of background and motivation, we begin with an overview of the characteristics of children with pd, and then turn to the two studies that have examined density and frequency relative to expressive phonology in this clinical population. the review intentionally narrows in on the effects of statistical regularities for the purpose of phonological learning, as exemplified by gains in accuracy of production and as induced through explicit training. children with phonological delay children with pd constitute a subgroup of preschoolers who present with severely reduced phonemic inventories relative to the ambient language and to children of the same age, in the absence of overt causes and with few other apparent developmental lags (gierut, b). because these children are highly unintelligible, they warrant clinical treatment to induce phonological learning. they are likely to be first identified as preschoolers (shriberg, kwiatkowski & gruber, ), so it is possible to engage them in phonological and learning tasks that would otherwise tax the skills or attention of a younger child. treatment itself is designed as an experiment, where instructional stimuli or methods serve as independent variables, and gains in expressive phonology serve as dependent variables. because treatment accelerates learning, it provides an opportunity to glimpse phonological gains on a longitudinal but compressed timescale, which might otherwise be protracted. while the causal factors behind pd are unknown, it has long been hypothesized that a possible source of children’s errors in production lies in the nature of their underlying lexical representations of words (stoel-gammon, ). this follows directly from linguistic accounts of typical phonological acquisition, which posit that learning the sound system of a language requires change (dinnsen, ), elaboration (rice & avery, ), and/or reorganization (gnanadesikan, ) of phonological properties associated with lexical representations. other psycholinguistic models (walley, ) have advanced similar gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript proposals about sublexical change in children’s representations (although these are not without controversy; e.g. dollaghan, ). one claim of the psycholinguistic models is that phonological elaboration of lexical representations is motivated by dense neighborhoods comprised of frequent words, such that phonetically similar forms necessitate phonemic distinctiveness. the relevance is that when the clinical, linguistic and psycholinguistic hypotheses are taken together, the ideas may be arranged as a series of propositions, which generate testable predictions for children with pd. specifically, (i) the locus of pd may lie in the nature of children’s lexical representations; (ii) the sublexical structure of representations may be vulnerable given children’s restricted phonemic inventories; (iii) word-level variables prompt sublexical change in expressive phonology; (iv) density and frequency are two specific word-level variables that motivate sublexical change; and (v) dense neighborhoods composed of frequent words provide a ripe context for sublexical change. if true, then experimental manipulations of word-level variables in the treatment of pd are expected to result in differential phonological learning, such that frequent words from dense neighborhoods are predicted as optimal for growth of the expressive phonology. density, frequency and phonological treatment to our knowledge, there are two studies that considered word-level variables in the design of phonological treatment for purposes of documenting acquisition of the expressive phonology by children with pd (gierut, morrisette & champion, ; morrisette & gierut, ). in one study (gierut et al., ), twelve children were enrolled in a single-subject alternating treatments manipulation to assess gains in production of the treated sound. in the complementary study (morrisette & gierut, ), eight children participated in a multiple baseline manipulation to assess system-wide gains in the expressive phonology. in both experiments, children were taught sounds in stimulus words that systematically varied in density or frequency value based on adult lexical counts (nusbaum, pisoni & davis, ). stimuli were blocked to yield either dense versus sparse or frequent versus infrequent conditions. within a condition, stimuli were further balanced for the complementary variable. that is, when blocked by density, half the treated words were frequent items and half infrequent items. likewise, when blocked by frequency, half the treated words were from dense and half sparse neighborhoods. this ensured that there was just one independent variable (density or frequency), with the alternate property being controlled. generalization to treated and untreated erred sounds was measured, with results converging across studies. namely, words from sparse neighborhoods promoted greater generalization accuracy than dense, and frequent words induced greater generalization accuracy than infrequent. the best-of-the-best conditions for phonological learning were thus sparse neighborhoods or frequent words. while this work demonstrated that word-level variables affect phonological learning by children with pd, it was not without caveat. on the side of frequency, the findings were in sync with typical phonological acquisition (stoel-gammon, ), but on the side of density, they were not. in pd, sparse neighborhoods were optimal for production, but in typical development, it was the reverse, with dense neighborhoods being most beneficial for expressive phonology. there are at least two possible reasons for the observed asymmetry associated with density. one possibility is that children with pd do not, or cannot, capitalize on dense structure to the same extent, or in the same way, as in typical development (storkel, b; storkel & hoover, b). another possibility is that the density asymmetry was entwined with the methodology. recall that density was manipulated independent of frequency in prior treatment studies (gierut et al., ; morrisette & gierut, ). while the latter is a conventional research tack, every word does have both a density and a frequency value, such that, in naturalistic circumstances, the two properties co-occur and also co-vary. also, it is now well established that density is positively correlated with gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript phonotactic probability (vitevitch & luce, ), which was left free to vary in the aforementioned studies. both points must now be taken into account if it is to be determined which cues affect expressive phonology, and how these may or may not converge for children with pd. the purpose of this study was to reassess the effects of density and frequency by orthogonally manipulating the stimulus words of treatment, with further control of other properties. the goals were twofold: to sort out possible population differences from methodological nuances associated with density, and to fill apparent gaps in the literature on statistical regularities and the expressive phonology. methods the children eight children were recruited by public announcement to area schools, daycare facilities, and early childhood programs. to participate, children had to be identified as having pd, characterized by a reduced consonantal inventory relative to english and to expected phonological development. this was established by performance at least one standard deviation below the normative mean reported for age- and gender-matched peers on the goldman-fristoe test of articulation– (goldman & fristoe, ), with a minimum of six sounds produced in error across contexts on this measure. children also had to be within three and six years of age because this is an approximate time frame for accelerated speech sound normalization (shriberg et al., ). inclusionary criteria further required that children perform within typical limits on a battery of diagnostic tests (gierut, b: ), which evaluated hearing acuity, oral-motor structure and function, non-verbal intelligence, working memory, expressive and receptive vocabulary, and language. exclusionary criteria included literacy, bilingualism and concurrent enrollment in treatment for speech and/or language disorders. in all, three boys and five girls met these criteria ; their mean age was ; (years; months) (range ; – ; ), shown in table . expressive phonological samples and analyses descriptions of the expressive phonologies of the children were developed. production data were elicited using the phonological knowledge protocol (pkp; gierut, b). this is a conventional structured probe that samples each english sound in each relevant word position in multiple exemplars. it provides for representative sampling of the statistical patterns of english by including, for example, words from dense/sparse neighborhoods, and frequent/infrequent words (gierut, b: – for statistics). it further provides for elicitation of minimal pairs and morphophonemic alternations as evidence of the phonemic status of sounds in a child’s expressive phonology. pkp data were collected using a spontaneous picture-naming task. a child’s responses were digitally recorded by the experimenter, and subsequently transcribed by independent trained listeners using narrow notation of the ipa. transcribers had no contact with the children or families, and were blinded to the experiment. phonetically transcribed responses were entered into an archival database by a separate team of research assistants who were also blinded to the experiment. archival entries were used to compute percentages of accurate production and to develop standard phonological analyses (dinnsen, ). phonological analyses established each child’s phonetic and phonemic inventories, distribution of sounds and rule-governed alternations. the phonemic inventory was central to this study because it delineated which sounds were used by a child to differentiate meaning, and which were not. sounds that were present in the phonemic inventory did not warrant treatment because they were already functional (i.e. gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript contrastive) in a child’s expressive phonology. by comparison, sounds excluded from the phonemic inventory were non-contrastive and produced with near % accuracy on the pkp (gierut & o’connor, ). these warranted treatment to bring a child’s expressive phonology on track with typical development and in line with the english language. linguistically, sounds excluded from the phonemic inventory reflect phonotactic constraints on the grammar (dinnsen, : ). there are three general types of phonotactic constraints: inventory constraints, which ban specific sounds from occurring (e.g. velar fricatives are prohibited in english); positional constraints, which ban specific sounds from occurring in particular contexts (e.g. velar nasals occur word-finally, but are prohibited word-initially in english); and sequence constraints, which ban certain sequences of sounds (e.g. consecutive sonorant– sonorant consonants such as /mr-/ are prohibited onsets of english). in this study, inventory constraints informed the independent variable because sounds excluded from the phonemic inventory were the focus of treatment for all children. likewise, inventory constraints were measured as the dependent variable to establish which sounds excluded from the inventory evidenced gains in accuracy of production. on average, children of the study excluded ten sounds from the phonemic inventory (range – ; table ). experimental design a single-subject, staggered, multiple-baseline (mbl) across-subjects design was used. single-subject designs are based on the premise that a child serves as his or her own control (mcreynolds & kearns, ). by their very nature, single-subject designs take into account potential individual differences associated with internal or external factors, e.g. age, severity of pd, attention or family dynamics. the mbl is one particular singlesubject design that is implemented as a period of no treatment followed by treatment for each child. in a staggered application of the mbl, children are enrolled in treatment sequentially, rather than concurrently. in an across-subjects application of the mbl, each child serves as one leg of a given experimental condition, such that effects are culled from multiple children, rather than multiple behaviors. legs of the mbl further provide for replication as a reflection of the generalizability of treatment effects. for phonological treatment, generalizability is typically achieved by demonstrating that the effects of treatment are not sound-specific. as applied in this study, the no-treatment or baseline phase of the mbl ensured stability of a child’s performance prior to the instatement of treatment, namely, that sounds excluded from the phonemic inventory showed little fluctuation in production accuracy (± % variation, following mcreynolds & kearns, ). to assess stability of performance over time, each child was given more than one baseline (table ). to assess stability across legs of the mbl, the number of baselines was increased by one as successive children enrolled. based on principles of single-subject design, baseline stability helps to establish a causal relationship between experimental treatment and change in performance. the ongoing stability of time-lagged baselines further demonstrates that change in performance is not attributable to maturation. if maturation were at work, then children next to enroll would be expected to evidence gains in the absence of treatment because the extended period of baseline monitoring would have afforded them the time to mature. as applied herein, the treatment phase of the mbl supplied the independent variable, namely, treatment of a sound excluded from a child’s phonemic inventory using stimulus words that orthogonally varied in density and frequency. four children were taught a sound in words from dense neighborhoods, and four others in words from sparse neighborhoods. within and across conditions, frequency was further varied, such that two of four children were treated on frequent, and two others on infrequent words. hereafter, we refer to the conditions as dense–freq, dense–infreq, sparse–freq and sparse–infreq. through these gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript manipulations, it was possible to examine the blocked effects of dense/sparse neighborhoods and frequent/infrequent words as in prior studies, in addition to the combined effects of density and frequency in keeping with the central aim of this study. the dependent variable was generalization, operationalized as the percentage accuracy of production of sounds excluded from a child’s baseline phonemic inventory as measured on the pkp. generalization data were collected longitudinally, with samples obtained at baseline, at the completion of each treatment step (outlined below), and immediately following completion of treatment. generalization data were also collected throughout treatment on a variable ratio schedule averaging two sessions. on average, pkp samples (range: – ) were obtained from each child. consistent with the enrollment procedures above, each pkp sample was digitally recorded and phonetically transcribed, with transcriptions entered into the archival database. the data were used to compute separate percentages of accuracy for each sound excluded from a given child’s phonemic inventory at each sampling point in time. thus, accuracy was documented for every sound excluded from the phonemic inventory of a child at approximately thirteen points in time, corresponding to the thirteen pkp samples that were administered on average. the full set of longitudinal data was then evaluated in data analysis. because accuracy data were derived from descriptive phonetic transcriptions, it was necessary to document reliability. transcription reliability was established for % of the longitudinal pkp data collected from each child. independent transcriptions from two judges were compared point-to-point, with mean agreement being % based on , consonants transcribed. treated stimuli the overarching goal of treatment was the elimination of inventory constraints as the unified linguistic characterization of all children’s error patterns. each child was taught one sound excluded from his or her phonemic inventory. the treated sound was pseudo-randomly assigned to accommodate individual differences in children’s inventories while also providing for replication. half the children were taught a fricative, and half a liquid, balanced by condition. the treated fricative was either /f/ or /s/, and the liquid was either /l/ or /r/; these are shown in table by condition and by child. the treated sound was taught in the initial position of eight treated words that conformed to the density/frequency characteristics of a child’s experimental assignment; the treated words are shown in the ‘appendix’. treated words were picturable, of suitable content for a child, and non- overlapping with pkp items so as to keep treated stimuli distinct from generalization stimuli. several considerations went into the selection of the treated sounds and treated words, as detailed below. treated sounds the sound /f/, /s/, /l/ or /r/ was selected for treatment for five reasons. first, these sounds were taught in the two previous studies of density and frequency in treatment of pd (gierut et al., ; morrisette & gierut, ). to maintain consistency with the literature and in the interest of replication and comparison, they were again chosen here. second, these sounds are vulnerable to developmental errors of production (smit, hand, freilinger, bernthal & bird, ). third, treatment studies typically demonstrate the generalizability of effects by manipulating an error pattern that is common to all participants, but uniquely instantiated by individual children (e.g. rvachew & nowak, ; tyler, edwards, & saxman, ). the consequence is that children are taught different sounds, which are tailored to their phonological needs, but which fall under the larger umbrella of a commonly shared error pattern. single-subject design explicitly provides for individual variation in the gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript treated sound because each participant serves as his or her own control. in the present study, we adopted the same tack. inventory constraints constituted the common pattern to be remediated in treatment, with the treated sound being specific to the expressive phonology of an individual child. fourth, the use of different treated sounds has the further effect of bolstering the ecological and clinical validity of the findings. because treated sounds varied, the impact of density and frequency on the elimination of inventory constraints could be established more broadly. last, and most critical, the selection of treated sounds was constrained by the planned manipulations of density and frequency relative to the actual (but asymmetric) distribution of words in the english language. recall that each treated sound was taught using eight stimulus words that conformed to a child’s experimental assignment, while also being picturable, appropriate and non-overlapping with items of the pkp. given this, some sounds with the potential for treatment had to be set aside outright because there were simply too few words in the english language to yield an adequate set of eight stimuli. to illustrate, there are seven word-initial /z/ forms that occur in dense neighborhoods of english (nusbaum et al., ), one of which is sampled on the pkp, leaving just six (not eight) possible stimuli for treatment. similarly, there are no word-initial /z/ forms that occur frequently in english (nusbaum et al., ), thereby further precluding /z/ as a possible treated sound. by comparison, the selection of /f/, /s/, /l/ or /r/ guaranteed roughly forty word-initial items for each sound and each category, dense/sparse and frequent/infrequent. this provided an adequate pool from which to select treated words. treated words treated words were selected based on density and frequency values, as retrieved from the hoosier mental lexicon (nusbaum et al., ). this is an online database that supplies the density of , words of the merriam-webster dictionary, with corresponding frequencies from kučera and francis ( ). an adult corpus was used for consistency with the two prior treatment studies involving the manipulation of density and frequency in pd (gierut et al., ; morrisette & gierut, ), which the present study aimed to extend and replicate. some have suggested that child corpora may provide a more accurate reference (dollaghan, ). yet others argue that adult corpora may be extended to the study of children for reasons associated with the normalization of standard distributions (kelly & martin, ), correlations showing the equivalence of corpora (storkel & hoover, a) and behavioral demonstrations of similar effects across corpora (gierut & dale, ). this debate notwithstanding, an adult corpus was used herein for ease of translation and uniformity across studies of pd, and to enable future comparisons between the performance of children and adults as a means of examining developmental continuity. to initiate stimulus selection, conventional operational definitions based on the raw values of density and frequency (luce, ) were applied, such that words from dense neighborhoods were to have ten or more similarity neighbors, and frequent words were to have a raw count of or more occurrences per million. in this study, treated words from dense neighborhoods had a mean of twenty-one neighbors, whereas those from sparse neighborhoods had a mean of four neighbors. treated words that were classified as frequent had a mean of occurrences and infrequent, twenty-one occurrences. in all, four sets of treated words were identified in accord with the crossing of density and frequency by experimental condition. because density and frequency are correlated with other properties of word structure, it was necessary to consider and control for alternate (and potentially confounding) variables in stimulus selection. specifically, for density, there is an established negative correlation with word length (pisoni, nusbaum, luce & slowiaczek, ): dense neighborhoods are largely comprised of short words. indeed, in this study, the treated words from dense neighborhoods averaged . segments, whereas those in sparse neighborhoods averaged . segments in gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript length. this reflected the expected negative correlation between density and length (r( )= − . , p< . ), but it also resulted in a statistically significant difference in length between the dense/sparse stimulus sets (t( )=− . , p< . ). typically, laboratory studies handle such correlations by holding length constant in stimulus selection; however, this may not be feasible, ecologically valid or necessarily desirable when naturalistic stimuli are used and when children are involved as participants (see storkel, a, for review). such was the case in the present study. accordingly, the recommended alternative (p. ) is to transform raw density values using z-score conversion and/or median transformation for length. transformed density values are computed relative to all words that are of the same length in a given corpus (i.e. the hoosier mental lexicon). in this way, length is taken into account to minimize potential confounds. importantly, transformed density values appear to be better predictors of the behavioral patterns to emerge from experimental manipulations of density (p. ). we followed this standard by applying both the procedure for density z-score conversion and density median transformation as outlined by storkel ( a: – ). particular attention was given to the data provided about the hoosier mental lexicon. specifically, storkel (table , p. ) reported the mean and standard deviation for neighborhood density by word length for all words of the hoosier mental lexicon; these values were used in computation of density z-scores. likewise, storkel (table , p. ) reported the median and interquartile range for neighborhood density by word length for all words of the hoosier mental lexicon; these values were used in computation of density median transformations. with referent values in place, density z-scores were calculated herein using the equation: (obtained raw density value – m)/sd (storkel, a: ). that is, for each treated word in each condition, the obtained raw density value was subtracted from the mean density value reported by storkel (table , p. ) for all words of the hoosier mental lexicon that were of the same length. this value was then divided by the standard deviation, also reported by storkel (table , p. ) for all words of the hoosier mental lexicon that were of the same length. to illustrate, funny was a treated word associated with the dense condition (see ‘appendix’). it consists of four segments, and has a raw density value of eleven. the mean density of all four-segment words in the hoosier mental lexicon is . (sd= . ) (table , p. ). thus, the density z-score of funny was . (i.e. ( – . )/ . ). by comparison, lazy was a treated word associated with the sparse condition (see ‘appendix’). it too consists of four segments, but has a raw density value of seven. its density z-score was . (i.e. ( – . )/ . ). this procedure was followed to derive the mean density z-scores for treated words used in the dense/sparse conditions. the average density z-score for the dense condition was . , and the sparse condition − . , resulting in a statistically significant difference in density by condition (t( )= . , p< . ). importantly, there was no correlation between the density z-scores and word length (r( )= − . , p= . ). density median transformation scores were also computed, using the equation: (obtained raw density value – mdn)/ / iqr) (storkel, a: ). procedures were much the same as above, with median transformation scores generated for each treated word used in the dense/sparse conditions. values reported by storkel (table , p. ) again supplied the median and inter-quartile range for density based on all words of the hoosier mental lexicon by length. the resulting mean density median transformation for treated words of the dense condition was . , and for treated words of the sparse condition . . density median transformation values were statistically significant by condition (t( )= . , p< . ), with no correlation between the transformed density values and length (r( )= − . , p= . ). gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript together, the density conversions provided the needed experimental safeguard to demonstrate that the treated stimuli were distinct in density by condition, even when length was taken into account. this established that density, not length, was a primary independent variable. (for completeness, the length of the treated words sorted by frequency was not statistically different (t( )=− . , p= . ). average length of frequent versus infrequent treated words was . versus . segments, . versus . biphones and . versus . syllables, respectively.) besides length, density is positively correlated with phonotactic probability (vitevitch & luce, ): words in dense neighborhoods are largely comprised of common sounds and sound sequences. thus, it was necessary to establish that treated words from dense/sparse neighborhoods did not differ in phonotactic probability as a spurious variable. we followed standard procedures recommended in the literature (storkel, a), computing the sums of segment and biphone frequencies for each stimulus word using a publicly available calculator (vitevitch & luce, ), with z-score conversion for length. resulting values were binarily coded using the rubric that positive phonotactic values reflect common sound sequences, and negative values rare sound sequences. chi-square analyses revealed no differences in commonality or rarity of the phonotactic patterns associated with treated stimuli from dense versus sparse neighborhoods. this held for sum of segment frequencies (χ ( , n= )= . , p= . ) and sum of biphone frequencies (χ ( , n= )= . , p= . ). the proportion of common:rare words was . : , yielding a nearly equivalent distribution of phonotactic probability by density. density, not phonotactic probability, was thus retained as a primary independent variable. frequency is likewise correlated with other statistical cues. specifically, there is a negative correlation between frequency and age-of-word-acquisition (aoa; gilhooly & watson, ). aoa is a reliable and behaviorally valid subjective estimate of when a given set of words was first acquired, as determined by child (walley & metsala, ) or adult judges (gilhooly & logie, ). the observed correlation is that high-frequency words tend to be earlier acquired, although the directionality of the effect remains a point of debate (gilhooly & watson, ). once again, it was necessary to eliminate aoa as a potential confound, rendering frequency as a primary independent variable. this was accomplished by retrieving the aoa values of stimulus words in reference to the normative counts of gilhooly and logie ( ) and bird, franklin and howard ( ). resulting aoa values were binarily coded, with words acquired before age six defined as early acquired in keeping with the inclusionary criteria for participation in this study. chi-square analyses revealed no difference in the aoa values of frequent versus infrequent stimuli (χ ( , n= )= , p= . ). overall, the ratio of early:late aoa words was . : , with near equal distribution of aoa by frequency. as above, frequency, not aoa, was established as a primary independent variable. imageability was a related consideration because it is negatively correlated with aoa (bird et al., ): earlier acquired words tend to be highly imageable, keeping in mind too that the treated words used in this study were to be picturable. above, we demonstrated that treated words were equivalent in aoa. it might be expected then that they would be comparable in imageability, given the reported correlation. nonetheless, this needed to be confirmed. subjective imageability ratings were retrieved from bird et al. ( ) for the treated words. mean imageability ratings of treated words were versus for dense versus sparse neighborhoods, respectively, and versus for frequent versus infrequent stimuli, respectively. there was no significant difference in the imageability ratings of treated words sorted by density (t( )= . , p= . ), or sorted by frequency (t( )=− . , p= . ). gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript beyond these controls, developmental phonological characteristics of the treated words were taken into account, given the interest in expressive phonology. the classification scheme of shriberg and colleagues ( ) was used to code each sound of each treated word by its expected developmental order of emergence. this scheme sorts sounds into early- /m n p b d w j h/, mid- /ŋ k g t f v ʧ ʤ/ and late- /θ ð s z ʃ ʒ r l/ sets. the distribution of early-, mid- and late- segments and clusters in treated words from dense neighborhoods was , , and , respectively; thus, treated words in dense neighborhoods were largely composed of late- sounds. for sparse neighborhoods, the distribution of early-, mid- and late- segments and clusters was , , and , respectively; again, treated words in sparse neighborhoods were largely composed of late- sounds. chi-square analyses showed no significant difference in the segmental and cluster composition of treated words based on the expected developmental order of emergence when sorted by density (χ ( , n= )= . , p= . ). likewise, there was no significant difference in the segmental composition of treated words based on developmental order when sorted by frequency (χ ( , n= )= . , p= . ). as before, treated words sorted by frequency consisted mostly of late- sounds. these analyses ensured that the segmental make-up of treated words was comparable on developmental phonological grounds. two final phonological considerations were taken into account in treated word selection. to the extent possible, treated stimulus sets were comprised of unique words that were not also neighbors (minimal pairs) of each other: % of treated stimuli ( of words) were non- overlapping in phonological form. the intent was to constrain a child’s exposure to the neighborhood structure of the treated words per se, without further enhancement of that structure in the form of minimal pair contrasts. also, to the extent possible, treated stimuli did not form neighbors (minimal pairs) with pkp words: % of relevant pkp words ( of words) were non-overlapping in phonological form with treated words. the intent was to minimize the possible inflation of generalization effects by keeping the treated stimuli distinct from those used to measure generalization. it must be recognized that any number of added factors might have been taken into account in stimulus selection, e.g. featural distinction, stress and grammatical or semantic class. following magnuson and colleagues ( ), these were left free to vary. our objective was to capture the range of variation in english, while at the same time block for the primary variables of interest. in this regard, density and frequency were the key properties varied within a stimulus set as independent variables, while word length, phonotactic probability, aoa, imageability, developmental order of segmental emergence and minimal pairs were controlled across stimulus sets. treatment procedures children were provided treatment three times weekly in individual sessions, each one hour in duration. treatment followed a conventional protocol (gierut, a) consisting of two steps. the first was imitation of the treated sound in treated words. during imitation, a child viewed a picture of the treated word displayed on a computer monitor; the clinician said the depicted word; the child repeated that word. each treated word was presented as a discrete trial, and feedback about accuracy of production was provided on a continuous schedule. accuracy of production was assessed for the treated sound in the treated word-initial position; accurate production of the word form was not also required, in keeping with conventional principles of phonological treatment (hodson & paden, : ). in cases of inaccurate production of the treated sound, instructions about articulatory placement were provided, and an additional model was presented. imitation continued until a child achieved % accuracy of production of the treated sound in the treated words over two consecutive sessions or until seven total sessions were completed, whichever occurred first. following this, treatment shifted to spontaneous naming. the child again viewed the picture of the gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript treated word on a computer monitor, but was to independently name that item with accurate production of the treated sound. feedback and instruction were provided in the same manner outlined for imitation. this continued until a child achieved % accuracy of production over three consecutive sessions or until twelve total sessions were completed, whichever occurred first. in all, children were afforded a maximum of nineteen treatment sessions (i.e. nineteen hours of training). they had the further opportunity to complete the protocol more rapidly, if they achieved performance levels of % and % accuracy of production over consecutive imitation and spontaneous sessions, respectively. thus, time- and/or performance-based criteria determined time in treatment for a given child. in this study, those assigned to the dense conditions tended to follow time-based advancement through the protocol, whereas those assigned to the sparse conditions tended to follow performance-based advancement. overall, the average number of sessions provided to any given child was fifteen, and the average by condition was nineteen for dense–freq, nineteen for dense–infreq, seven for sparse–freq and fourteen for sparse–infreq. across conditions, there was no statistically significant difference in the number of sessions provided in each step of treatment (χ ( , n= )= . , p= . ). overall, mean accuracy of the treated sound at completion of treatment was %. mean accuracy by condition was % for dense–freq, % for dense– infreq, % for sparse–freq and % for sparse–infreq. it must be underscored that performance during treatment showed only that children learned what was taught. the critical question was how children then transferred what had been learned to the expressive phonology. treatment thus served as the conduit from which children were left free to generalize in keeping with the dependent variable. data analysis four sets of analyses of longitudinal generalization data were completed. recall that, for each child, at each pkp sampling point in time, percentages of production accuracy were computed for each sound excluded from the phonemic inventory. this included the treated sound as sampled across contexts and words, as well as other untreated (erred) sounds. any sound that evidenced any gain at any sampling point in time was entered into analysis, and the full set of longitudinal data associated with that sound was evaluated. approximately , productions were examined for generalization, with each child contributing at least , productions over time. the first two evaluations were complementary in establishing which experimental conditions induced generalization from qualitative and quantitative vantages, respectively. the analyses were identical to prior studies of density/frequency in treatment of pd (morrisette & gierut, ), and for comparison purposes, were again applied herein. qualitatively, a dichotomous yes/no coding of generalization was applied to each experimental condition. the coding of yes/no generalization was based on the accepted criterion cut-off of % or greater mean generalization accuracy relative to baseline performance for children of a given condition (elbert, dinnsen & powell, ). this analysis offered preliminary insight to the experimental conditions affiliated with generalization. the quantitative analysis followed, using the rank procedure for replicated ab multiple baseline design to establish generalization as a statistically reliable effect (morrisette & gierut, : ). briefly, the procedure makes use of the combined sign test (marascuilo & serlin, ), organizing sounds that generalized into three sets or ‘factors’ for each child by condition. the three factors were the treated sound, sounds from the same manner class as the treated (within-class generalization) and sounds from different manner classes than the treated (across-class generalization). the number of factors varied depending on which sounds generalized, e.g. if a given child showed no generalization to the treated sound, then that factor was not instantiated (see table ). mean baseline performance gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript and longitudinal generalization accuracy were computed and ranked for each factor, with the difference between ranks totaled by experimental condition. a binomial distribution, where p= . , was then applied to determine whether the resulting generalization was statistically reliable. the third evaluation stands apart in its calculation of effect size for single-subject experimental design. this is a relatively new addition to the single-subject literature that has been introduced as a way of quantifying treatment efficacy, with an eye toward meta- analyses of small-n studies (beeson & robey, ). effect size captures the magnitude of generalization for a given treatment condition, with values compared across conditions to identify which is relatively most efficacious. in this study, standard mean difference (d; beeson & robey, ) was computed. mean accuracy of baseline data was calculated, along with mean accuracy of longitudinal generalization data. the difference between means was obtained and divided by the standard deviation of the baseline, corrected for continuity (gierut & morrisette, b), to achieve the effect size, d. while benchmarks for interpretation of d have been established for between-group designs, it is inappropriate to extend these to small-n studies (beeson & robey, ). standards for what constitutes a small, medium or large effect must be developed empirically, but before this can be established, reports of effect size in single-subject studies must accumulate. to our knowledge, there are only four studies enrolling clinical populations in some form of linguistic treatment that have cited effect sizes, with d values in the range of . to . (beeson & robey, ; gierut & morrisette, a; b). data from the present study added to this base as a secondary contribution. the three aforementioned analyses (i.e. criterion cut-off, rank procedure, effect size) were applied to the generalization data in comparisons blocked by density (dense vs. sparse neighborhoods) and by frequency (frequent vs. infrequent words), as well as in the orthogonal comparison of density × frequency. together, the results revealed the triggers of generalization, namely, which type of treated words prompted generalization in the expressive phonology of children with pd. for the fourth analysis, attention shifted from an evaluation of percentage accuracies to a descriptive characterization of the lexical properties of the pkp words that actually generalized. while certain words might trigger generalization, it was reasonable to also ask which words were the targets of generalization or the recipients of that change (morrisette, ). consider, for example, that treatment in the dense–freq condition might only result in generalization to other frequent words in dense neighborhoods. to address this, the density and frequency values of pkp words that showed improvement in production accuracy from baseline to posttreatment were retrieved from the hoosier mental lexicon (nusbaum et al., ). chi-square analyses were computed to determine whether there were differences in which words generalized based on neighborhood or frequency characteristics. the collective set of analyses thus identified both the triggers and targets of generalization. results the results emphasize the experimental manipulation of density and frequency as triggers of phonological generalization in treatment, in keeping with the central purpose of the study. the presentation is organized to summarize blocked, then orthogonal comparisons of density and frequency, as summarized in table . this is followed by a description of the targets of generalization, with a summary of the density/frequency characteristics of pkp words that generalized as a consequence of treatment. gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript triggers of phonological generalization in the blocked comparison of density, both dense and sparse treatment conditions induced generalization based on the % criterion cut-off. this is shown in table , which reports the mean percentage accuracy by condition and then codes that as yes/no generalization based on the % criterion cut-off (elbert et al., ). the rank procedure confirmed that generalization associated with words from dense neighborhoods was statistically reliable, where p= . , whereas that associated with sparse neighborhoods was not. comparisons of effect size further established that the magnitude of generalization that derived from teaching words from dense neighborhoods was three times greater than sparse neighborhoods (d= . vs. . , respectively). thus, when frequency was removed from consideration, treatment of words from dense neighborhoods induced greater phonological generalization, with convergence across analyses. in the blocked comparison of frequency, table again shows that the % mean generalization cut-off was achieved in treatment of both frequent and infrequent words. the rank procedure confirmed that frequent words resulted in statistically reliable effects, but the same did not hold for infrequent words. in terms of effect size, treatment of frequent words resulted in a slightly greater magnitude of gain than infrequent words (d= . vs. . , respectively). thus, frequent words had a modest advantage over infrequent for phonological generalization, when density was removed from consideration. orthogonal comparisons rendered a more detailed picture of the interface of density and frequency relative to generalization. table shows that children achieved the % criterion cut-off in three of four experimental conditions: dense–freq, dense–infreq and sparse– infreq. of these, only the dense–freq condition resulted in reliable effects based on the rank procedure because children of this condition generalized to treated and untreated sounds from treated and untreated manner classes (see ‘data analysis’ above for a description of the rank procedure and variance in the number of factors). effect sizes associated with the four experimental conditions are plotted in figure , from which three observations may be gleaned. first, treatment of words from dense neighborhoods, whether frequent or infrequent, consistently resulted in a greater effect size than sparse neighborhoods (see also table ). second, the relative effect sizes were graded, such that dense–freq>dense– infreq> sparse–infreq>sparse–freq in magnitude of generalization. the optimal trigger of phonological generalization thus appeared to be treatment of frequent words from dense neighborhoods. a final observation hints at a possible interaction between density and frequency. that is, frequency reversed in direction of generalization effects, depending on neighborhood density. notice in figure that the dense–freq condition outranked the dense–infreq (d= . vs. . , respectively), whereas it was just the reverse for the sparse–infreq condition relative to the sparse–freq (d= . vs. . , respectively). this suggests that the influence of a word’s frequency may shift depending on its residence in a neighborhood. targets of phonological generalization chi-square analysis showed no significant difference in the density of pkp words that showed improved production accuracy as a result of treating words from dense versus sparse neighborhoods (χ ( , n= )= . , p= . ). pkp words that generalized were almost equally split across dense and sparse neighborhoods, on the order of . : . likewise, there was no significant difference in the frequency of pkp words that generalized as a result of treating frequent versus infrequent words (χ ( , n= )= . , p= . ). again, pkp words that generalized included frequent and infrequent items, on the order of : . . these results suggest that improvements in the expressive phonology of children with pd took place across a variety of words with ranging properties of density/frequency. gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript discussion this study set out to reassess how density and frequency affect acquisition of the expressive phonology by children with pd enrolled in treatment. the main finding was a differential effect of density on phonological learning. dense neighborhoods were consistently affiliated with greater gains in the phonologies of children with pd as a consequence of treatment. the effect held independent of frequency, and was confirmed through converging analytic perspectives. by comparison, there was not a clear differential effect of frequency: frequent and infrequent words led to near equivalent phonological gains. moreover, the way that frequency contributed to generalization appeared to depend on its pairing with density, indicative of a potential interaction between variables. the results hinted of graded generalization effects, such that the dense–freq condition was most efficacious as a trigger of phonological learning. while the dense–freq condition might have triggered the greatest learning, improvements in production accuracy were not limited to just those same kinds of words; rather, accuracy extended to untreated words with varied lexical characteristics, including dense/sparse and frequent/infrequent items. this suggests that the dense–freq condition infused system-wide change that fanned across sounds of the phonology, and also words of the lexicon. taken together, the main findings speak to at least four outstanding issues that frame the discussion. these are (i) the utility of statistical cues for children with pd; (ii) the function of density in shaping phonemic contrasts; (iii) an asymmetry in words that are triggers versus targets of phonological learning; and (iv) a possible interaction of density and frequency in phonological learning. population trait or methodology? recall that a motivation for the present study was the finding that children with pd benefit from sparse neighborhoods in phonological learning (gierut et al., ; morrisette & gierut, ). this stood apart from typical development, giving rise to the hypothesis that a sparse neighborhood advantage may be a diagnostic hallmark of pd (storkel, b; storkel & hoover, b). in light of our findings, this hypothesis was not supported: children capitalized on words from dense neighborhoods to promote expansion of the phonemic inventory, consistent with typical phonological acquisition (stoel-gammon, ). consequently, density might not be the parameter that uniquely distinguishes pd from typical development. instead, it is likely that methodological differences account for the disparity in density effects across studies of pd. the difference can be traced to the independent versus concurrent manipulation of density relative to frequency. in the present study, density and frequency were fully crossed in the experimental conditions, whereas previously, density was held constant but frequency was balanced across conditions. an implication is that the function of a given statistical variable may wax and wane depending on the other cues that might be operating concurrently. here, density was bolstered by orthogonal manipulations, whereas frequency appeared to play less of a differential role in generalization. this observation is of interest because other developmental studies, also involving orthogonal manipulations of density and frequency, have reported similar effects (garlock et al., ; krull, choi, kirk, prusick & french, ). in each case, when density moved to the forefront, the effects of frequency receded (see also gierut & morrisette, a, for similar findings in the orthogonal variation of frequency and aoa). this has led to a suggestion that the relationship between statistical variables may be multiplicative, not additive (krull et al., ). in future research, it will be important to develop this point further by examining the combined contributions of input variables to the expressive phonology. work along these lines is beginning, with emphasis on the disambiguation of correlated variables, namely, density and phonotactic probability (storkel & hoover, b), or frequency and aoa (gierut & morrisette, a). by building this line of study, it may be possible to determine gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript if there are precedents, weightings, and/or trade-offs among statistical regularities relative to the expressive phonology. there is also another possible explanation for the waxing and waning of density and frequency that warrants consideration. in particular, the prominence of behavioral effects may vary depending on the reference norms that are used. consider that in this, and other developmental studies which co-varied density and frequency (garlock et al., ), an adult lexical corpus was consulted in stimulus development. yet there are likely differences in the size and structure of a child’s lexicon relative to the adult (walley, ). for a child, neighborhoods appear to be denser (coady & aslin, ), and words of the lexicon are shorter (charles-luce & luce, ) and more frequent (storkel & hoover, b). moreover, differential effects have been reported in connection with the use of different corpora, child versus adult and reception versus expression (goodman, dale & li, ; krull et al., ; but see gierut & dale, , for consistency of effects across corpora in pd). in future research, it will be necessary to test the application of different lexical corpora against acquisition of the expressive phonology. this may be accomplished by direct replication, or by confirmation that the coding of experimental stimuli as dense/sparse or frequent/infrequent is the same across corpora (storkel & lee, ). applications of child corpora in particular may have further methodological advantages in that known relationships between statistical variables might be easier to control. for example, because children’s neighborhoods are denser but words are generally shorter, it may be possible to identify ample stimuli of the same length that vary in density in a child-based corpus. this would eliminate the need to use transformed values in stimulus selection (storkel, a), which may have the undesired consequence of inflating behavioral effects. research along these lines would address two limitations of the present study, namely, the use of an adult corpus in stimulus selection and transformed density values in the operational definition of the independent variables. as a possible broader outcome, the results to emerge from such research may help settle debates about the suitability of different corpora in developmental research (cf. charles-luce & luce, ; ; dollaghan, ). the present findings also have possible applied consequences for the design of instructional programs for pd, but it will necessary to proceed with caution in the translation of research to practice. clinically, teaching erred sounds in words from dense neighborhoods is likely to be effective in promoting the expressive phonology, particularly when the treated words are also frequently occurring. however, it must be underscored that this recommendation is relevant only to children whose primary deficit is a severely reduced phonemic inventory, who are enrolled in instructional programs to eliminate inventory constraints. in future research, it will be necessary to demonstrate the effects of dense neighborhoods for children who exhibit other phonotactic errors associated with positional or sequence constraints. an intriguing possibility is that the words that are optimal for generalization may vary depending on the nature of the phonotactic error being treated. this, in turn, may reveal the unique function of statistical regularities in the acquisition of expressive phonology. density and expressive phonology a second point of discussion is to consider how words from dense neighborhoods gave rise to the greatest system-wide phonological generalization. to evaluate this, we return to the hypothesis that the representation of words is a possible source of pd. if true, then it is possible that dense structure served multiple functions, all of which converged to magnify sublexical structure in treatment. consider that, because treatment was phonological in nature, it might have cast a ‘floodlight’ on the sound system. in turn, dense neighborhoods might have drawn a child’s attention to the phonological structure of the treated words, with that structure being revealed in several ways. gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript specifically, because words from dense neighborhoods have many phonetically similar counterparts, these too would have been activated in treatment. consequently, the same phonological details would have been realized again and again. a treated word like rope, for example, activates neighbors road, robe, hope and soap, among others. notice the shared points of overlap involving the body (/ro/) and rime (/op/), respectively. one possibility is that resemblances in segmental and syllabic structure were reinforced in treatment of words from dense neighborhoods. another way that dense neighborhoods might have brought sublexical structure to the forefront was through the occurrence of minimal pairs. dense neighborhoods essentially define lexical pockets of minimal pairs, which themselves are revealing of phonemic distinctions. as such, treated words from dense neighborhoods might have ‘spotlighted’ (peters & strömqvist, ) the range of phonemic contrasts exemplified in a given neighborhood. thus, neighbor activation may have reinforced overlapping phonological structure while, at the same time, highlighted unique points of difference among minimally distinct forms in dense neighborhoods. it is also the case that minimal pairs are central to the differentiation of meaning in language, and meaning, in turn, is the foundation of the lexicon. possibly, exposure to words from dense neighborhoods strengthened the critical association between form and meaning. other aspects of dense neighborhoods might have been conducive to the process of generalization itself. in particular, treated words from dense neighborhoods were all members of the same cohort, where cohort is defined as words that begin with the same initial consonant (norris, ). it is relevant that words of a cohort are not necessarily in the same neighborhood (magnuson et al., ). for example, sun and sat are in the same cohort, but sun has as its neighbors bun or fun, whereas sat has as its neighbors bat or fat. given this, treatment may have revealed two sides of lexical organization associated with cohort and neighborhood structure, respectively. some have suggested that when children are exposed to a broader problem space such as this, it enables the generalization process (gerken, ). thus, treatment directed toward production of words from dense neighborhoods with shared cohorts may have facilitated greater generalization in children with pd. ultimately, the input provided to a child in treatment might have converged to guide the discovery of basic phonological constructs. this idea would be consistent with macwhinney’s ( ) claim that functional cues make abstract linguistic structure known. perhaps, through treatment of words from dense neighborhoods, a child inferred certain underlying properties of phonological systems, such as what is a phone, phoneme, minimal pair or well-formed syllable. it might have been precisely this convergence of more than one cue that tripped phonological generalization. an implication that emerges is that statistical regularities in the input must be made especially redundant in order for children with pd to extract crucial sublexical details and advance the expressive phonology. future studies that examine multiple cues in tandem are again needed to test this hypothesis. through this sort of work, it may be possible to delineate whether cue redundancy is necessitated in the case of pd, and to establish which and how many cues must converge as primary evidence in treatment. research along these lines would have the added benefit of addressing another limitation of the present study, namely, the small sample size. replications through enrollment of increased numbers of children with pd, who present with varied error patterns being taught sounds that vary in developmental order of emergence, would confirm the robustness of the density effects. a likely outcome would be the characterization of individual differences relative to the effects of statistical variables on expressive phonology. gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript triggers versus targets of phonological gains the results of this study bring up a distinction between the properties of words that trigger phonological learning as opposed to the properties of words that are the targets, or recipients, of improved production accuracy. at first glance, it might be expected that the types of words to induce generalization would be the same as those that undergo generalization in : correspondence. for example, if words from dense neighborhoods were taught, then it might be expected that untreated words, also from dense neighborhoods, would show improved production accuracy. this, however, was not borne out in the present study, nor has it been found in the literature on pd (morrisette, ), typical phonological development (leonard, newhoff & mesalam, ) or fully developed linguistic systems generally (labov, ). rather, the primary way that phonological innovations are introduced in child or adult grammars is through lexical diffusion (chen & wang, ); that is, sound change takes place gradually on a word-by-word basis. while many have tried to find the driving force behind lexical diffusion, this line of research has not met uniform success. some have considered statistical factors in diffusion (bybee, ), whereas others have entertained sociolinguistic (labov, ) or perceptual/ articulatory (janson, ) factors. still others have suggested that diffusion might be cyclic (morrisette, ), with change defined by alternations in the value of word-level variables (e.g. change high-frequency words, then change low-frequency words, then return to change more high-frequency words, and so on). despite efforts, there has been no clear resolve in identifying or predicting precisely which words undergo diffusion when new sounds are introduced to the phonemic inventory. this notwithstanding, the present results are wholly consistent with lexical diffusion. treated words from dense neighborhoods triggered phonological gain, but the targets of change cross-cut organizational properties of the lexicon to encompass all kinds of words. of importance, broad lexical change went hand-in-hand with broad phonological change in the composition of children’s phonemic inventories. this dual benefit to lexical and sublexical structure underscores the bidirectional influences between phonology and the lexicon that stoel-gammon ( ) emphasized. for the future, it might be appropriate to consider the other side of the coin, namely, circumstances under which the process of lexical diffusion stalls. it may be that, for children with pd, certain word types resist improved production accuracy even in the face of long- term intervention. words that resist change might align based on their statistical properties, such that certain word-level variables might be found to block or inhibit phonological generalization. interactions of density and frequency as a final point of discussion, we consider a possible interaction between density and frequency, similar to that reported in other developmental work involving orthogonal evaluations of these two variables (krull et al., ). as shown in figure , there was a possible interaction given the graded generalization by condition, such that the impact of frequency seemed to vary based on the value of density: dense–freq>dense–infreq, but sparse–infreq>sparse–freq. the advantages of density for phonological learning have already been discussed, and the importance of frequency relative to density was shown to be consistent with the literature (stoel- gammon, ). of interest here is the finding that the sparse–infreq condition was relatively more effective than the sparse–freq condition in promoting generalization. based on the literature (stoel-gammon, ), neither sparse neighborhoods nor infrequent words are expected to impact the expressive phonology in a major way. there are, however, two possibilities that might describe the results; namely, gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript treated words associated with the sparse conditions may have acted like nonwords or lexical islands (garlock et al., ). with regard to nonwords, in the sparse–infreq condition, children might have viewed the treated stimuli as novel forms because they had few resemblents and rarely occurred. it has been reported that lexical novelty attracts children’s attention (storkel & hoover, b), that nonwords activate sublexical structure (vitevitch & luce, ) and that children with pd experience greater generalization following exposure to nonwords (gierut, morrisette & ziemer, ). together, these are possible reasons why positive gains may have occurred in the sparse–infreq condition. with regard to lexical islands, in the sparse–freq condition, children might have had little motivation to change the expressive phonology. these treated words occurred often in the input, but there were few other competitors in the similarity neighborhood. consequently, a child would have had little difficulty differentiating treated words from other phonetically similar words because the forms were already distinct. this may be a reason why the sparse–freq condition was least effective in triggering generalization. while intriguing, these accounts of an apparent interaction should be viewed with caution. keep in mind that the presumed interaction between density and frequency was identified based on graded effect size values, and not relative to interpretive standards associated with small, medium or large effects. as noted, benchmarks for interpretation of effect size in single-subject design have yet to be established empirically, necessitating first the accumulation of effect size data; the present findings contribute in this regard. once benchmarks are eventually in place, it is possible that the graded effect sizes reported herein may fall into a single interpretive category, blurring the potential interaction. still, it must be recognized that even the smallest effect size differences between treatments can have significant real-world consequences for the functional outcome of children with pd, thereby highlighting a distinction between experimentally and clinically significant gains in expressive phonology (bain & dollaghan, ). nevertheless, future studies may be interested in the contributions of nonwords and lexical islands to the expressive phonology for independent reasons. for nonwords, it will be important to disentangle the effects of novelty from those of statistical regularity (saffran & thiessen, ) and to further disentangle novelty in phonological form from novelty in meaning (gierut et al., ). to our knowledge, there have been no evaluations of lexical islands in treatment and thus their effect on the expressive phonology remains unknown. conclusion this study capitalized on stoel-gammon’s ( ) observation that regularities in word-level variables affect children’s acquisition of expressive phonology. the approach adopted was novel in its inclusion of children with pd, experimental manipulation of treatment to induce change in the expressive phonology, and orthogonal variation of density and frequency. results showed that children with pd made the greatest gains in expressive phonology following treatment of frequent words from dense neighborhoods. this finding aligned with typical phonological development, but stood apart from prior studies of pd that found a sparse neighborhood advantage. differences were reconciled by considering the effects of multiple converging cues on phonological generalization, with the hypothesis that children with pd might warrant greater than usual cue redundancy to extract sublexical information. the findings also documented the most-to-least influential cues for growth of the expressive phonology and an asymmetry between triggers and targets of phonological learning as exemplified by the process of lexical diffusion. new directions for research were offered to gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript further tease apart the complexities of statistical regularities relative to expressive phonology, with emphasis on the concurrent manipulation of more than one variable. through continued work, it should be possible to delineate how lexical and sublexical structure contributes to phonological learning in particular, and language learning in general, by garnering complementary evidence from expression and reception. references bain ba, dollaghan ca. the notion of clinically significant change. language, speech and hearing services in schools. ; : – . beckman, me.; edwards, j. lexical frequency effects on young children’s imitative productions. in: broe, m.; pierrehumbert, j., editors. papers in laboratory phonology v. cambridge: cambridge university press; . p. - . beeson pm, robey rr. evaluating single-subject treatment research: lessons learned from the aphasia literature. neuropsychology review. ; : – . [pubmed: ] bird h, franklin s, howard d. age of acquisition and imageability ratings for a large set of words, including verb and function words. behavior research methods, instruments, & computers. ; : – . bybee, jl. phonology and language use. cambridge: cambridge university press; . charles-luce j, luce pa. similarity neighbourhoods of words in young children’s lexicons. journal of child language. ; : – . [pubmed: ] charles-luce j, luce pa. an examination of similarity neighbourhoods in young children’s receptive vocabularies. journal of child language. ; : – . [pubmed: ] chen m, wang wsy. sound change: actuation and implementation. language. ; : – . coady ja, aslin rn. phonological neighbourhoods in the developing lexicon. journal of child language. ; : – . [pubmed: ] dinnsen, da. methods and empirical issues in analyzing functional misarticulation. in: elbert, m.; dinnsen, da.; weismer, g., editors. phonological theory and the misarticulating child (asha monographs no. ). rockville, md: american speech- language-hearing association; . p. - . dollaghan ca. children’s phonological neighbourhoods: half empty or half full? journal of child language. ; : – . [pubmed: ] elbert m, dinnsen da, powell tw. on the prediction of phonologic generalization learning patterns. journal of speech and hearing disorders. ; : – . [pubmed: ] garlock vm, walley ac, metsala jl. age-of-acquisition, word frequency and neighborhood density effects on spoken word recognition by children and adults. journal of memory and language. ; : – . gerken, la. child phonology: past research, present questions, future directions. in: gernsbacher, ma., editor. handbook of psycholinguistics. new york: academic press; . p. - . gerken la. decisions, decisions : infant language learning when multiple generalizations are possible. cognition. ; :b –b . [pubmed: ] gierut, ja. fundamentals of experimental design and treatment. in: dinnsen, da.; gierut, ja., editors. optimality theory, phonological acquisition and disorders. london: equinox; a. p. - . gierut, ja. phonological disorders and the developmental phonology archive. in: dinnsen, da.; gierut, ja., editors. optimality theory, phonological acquisition and disorders. london: equinox; b. p. - . gierut ja, dale ra. comparability of lexical corpora: word frequency in phonological generalization. clinical linguistics & phonetics. ; : – . [pubmed: ] gierut ja, morrisette ml. age-of-word-acquisition effects in treatment of children with phonological delays. applied psycholinguistics. a gierut ja, morrisette ml. effect size in clinical phonology. clinical linguistics & phonetics. b gierut ja, morrisette ml, champion ah. lexical constraints in phonological acquisition. journal of child language. ; : – . [pubmed: ] gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript gierut ja, morrisette ml, ziemer s. nonwords and generalization in children with phonological disorders. american journal of speech-language pathology. ; : – . [pubmed: ] gierut ja, o’connor km. precursors to onset clusters in acquisition. journal of child language. ; : – . [pubmed: ] gilhooly kj, logie rh. age-of-acquisition, imagery, concreteness, familiarity, and ambiguity measures for , words. behavior research methods, instruments and computers. ; : – . gilhooly kj, watson fl. word age-of-acquisition effects : a review. current psychological reviews. ; : – . gnanadesikan, ae. child phonology in optimality theory : ranking markedness and faithfulness constraints. in: stringfellow, a.; cahana-amitay, d.; hughes, e.; zukowski, a., editors. proceedings of the th annual boston university conference on language development. vol. . somerville, ma: cascadilla press; . p. - . goldman, r.; fristoe, m. goldman–fristoe test of articulation. . circle pines, mn: american guidance service; . goodman jc, dale ps, li p. does frequency count? parental input and the acquisition of vocabulary. journal of child language. ; : – . [pubmed: ] hodson, bw.; paden, ep. targeting intelligible speech : a phonological approach to remediation. . austin, tx: pro-ed; . janson t. sound change in perception and production. language. ; : – . jusczyk pw, luce pa, charles-luce j. infants’ sensitivity to phonotactic patterns in the native language. journal of memory and language. ; : – . kelly, mh.; martin, s. domain-general abilities applied to domain-specific tasks : sensitivities to probabilities in perception, cognition, and language. in: gleitman, l.; landau, b., editors. the acquisition of the lexicon. cambridge, ma: mit press; . p. - . krull v, choi s, kirk ki, prusick l, french b. lexical effects on spokenword recognition in children with normal hearing. ear and hearing. ; : – . [pubmed: ] kučera, h.; francis, wn. computational analysis of present-day american english. providence, ri: brown university; . labov, w. principles of linguistic change : internal factors. cambridge, ma: blackwell; . leonard lb, newhoff m, mesalam l. individual differences in early child phonology. applied psycholinguistics. ; : – . luce, pa. neighborhoods of words in the mental lexicon (technical report no. ). bloomington, in: speech research laboratory, indiana university; . macwhinney, b. the competition model. in: macwhinney, b., editor. mechanisms of language acquisition. hillsdale, nj: lawrence erlbaum; . p. - . magnuson js, dixon ja, tanenhaus mk, aslin rn. the dynamics of lexical competition during spoken word recognition. cognitive science. ; : – . [pubmed: ] marascuilo, la.; serlin, rc. statistical methods for the social and behavioral sciences. new york: w. h. freeman; . mcreynolds, lv.; kearns, kp. single-subject experimental designs in communicative disorders. baltimore, md: university park press; . metsala jl. an examination of word frequency and neighborhood density in the development of spoken word recognition. memory & cognition. ; : – . morrisette, ml. unpublished doctoral dissertation. indiana university; . lexical influences on the process of sound change in phonological acquisition. morrisette, ml. recursive change in phonological acquisition. in: slifka, j.; manuel, s.; matthies, m., editors. from sound to sense : + years of discoveries in speech communication. boston: mit; . p. - . morrisette ml, gierut ja. lexical organization and phonological change in treatment. journal of speech, language, and hearing research. ; : – . gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript munson b. phonological pattern frequency and speech production in adults and children. journal of speech, language, and hearing research. ; : – . norris d. shortlist : a connectionist model of continuous speech recognition. cognition. ; : – . nusbaum, hc.; pisoni, db.; davis, ck. sizing up the hoosier mental lexicon : measuring the familiarity of , words (progress report no. ). bloomington, in: speech research laboratory, indiana university; . http://neighborhoodsearch.wustl.edu/neighborhood/home.asp peters, am.; strömqvist, s. the role of prosody in the acquisition of grammatical morphemes. in: morgan, jl.; demuth, k., editors. signal to syntax : bootstrapping from speech to grammar in early acquisition. mahwah, nj: lawrence erlbaum; . p. - . pisoni db, nusbaum hc, luce pa, slowiaczek lm. speech perception, word recognition and the structure of the lexicon. speech communication. ; : – . rice, k.; avery, p. variability in a deterministic model of language acquisition : a theory of segmental elaboration. in: archibald, j., editor. phonological acquisition and phonological theory. hillsdale, nj: erlbaum; . p. - . rvachew s, nowak m. the effect of target-selection strategy on phonological learning. journal of speech, language, and hearing research. ; : – . saffran jr, aslin rn, newport el. statistical learning by -month-old infants. science. ; : – . [pubmed: ] saffran jr, thiessen ed. pattern induction by infant language learners. developmental psychology. ; : – . [pubmed: ] shriberg ld, kwiatkowski j, gruber fa. developmental phonological disorders ii : short-term speech-sound normalization. journal of speech and hearing research. ; : – . [pubmed: ] smit ab, hand l, freilinger jj, bernthal je, bird a. the iowa articulation norms project and its nebraska replication. journal of speech and hearing disorders. ; : – . [pubmed: ] stoel-gammon c. relationships between lexical and phonological development in young children. journal of child language. ; : – . [pubmed: ] storkel hl. learning new words: phonotactic probability in language development. journal of speech, language, and hearing research. ; : – . storkel hl. methods for minimizing the confounding effects of word length in the analysis of phonotactic probability and neighborhood density. journal of speech, language, and hearing research. a; : – . storkel hl. the emerging lexicon of children with phonological delays : phonotactic constraints and probability in acquisition. journal of speech, language, and hearing research. b; : – . storkel hl, hoover jr. an online calculator to compute phonotactic probability and neighborhood density on the basis of child corpora of spoken american english. behavior research methods. a; : – . [pubmed: ] storkel hl, hoover jr. word learning by children with phonological delays : differentiating effects of phonotactic probability and neighborhood density. journal of communication disorders. b; : – . [pubmed: ] storkel hl, lee sy. the independent effects of phonotactic probability and neighborhood density on lexical acquisition by preschool children. language and cognitive processes. ; : – . [pubmed: ] swingley d, aslin rn. spoken word recognition and lexical representation in very young children. cognition. ; : – . [pubmed: ] tyler aa, edwards ml, saxman jh. clinical application of two phonologically-based treatment procedures. journal of speech and hearing disorders. ; : – . [pubmed: ] vitevitch ms, luce pa. a web-based interface to calculate phonotactic probability for words and nonwords in english. behavior research methods, instruments, & computers. ; : – . gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript http://neighborhoodsearch.wustl.edu/neighborhood/home.asp walley ac. the role of vocabulary development in children’s spoken word recognition and segmentation ability. developmental review. ; : – . walley ac, metsala jl. young children’s age-of-acquisition estimates for spoken words. memory & cognition. ; : – . appendix dense frequent infrequent same race fairy rake saw reach fan rat serve real fast reef set red fence rich side rest fist rope size right foal round soon road fog rug sound role funny rush sparse frequent infrequent family labor launch circus field language lazy saddle figure level lemon sailor final little lens siren first living lion soccer foreign local lizard sofa forward longer llama soldier further lost lumber supper gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript fig. . effect size associated with phonological generalization resulting from the orthogonal manipulation of density and frequency in treatment. gierut and morrisette page j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript gierut and morrisette page ta b le p ar ti ci pa nt a nd e xp er im en ta l ch ar ac te ri st ic s p ar ti ci p an t ch ar ac te ri st ic s e xp er im en ta l ch ar ac te ri st ic s g en d er a ge ( ye ar s; m on th s) g f t a - p h on em es e xc lu d ed n ei gh b or h oo d s tr u ct u re w or d f re q u en cy b as el in es t re at ed s ou n d m ; f v θ ð s z ʃ ʧ ʤ l r h d en se f re qu en t s f ; ŋ θ ð ʃ ʧ ʤ l r d en se f re qu en t r f ; f v θ ð l w j d en se in fr eq ue nt f f ; ŋ k g θ ð l r d en se in fr eq ue nt r f ; ŋ p t k g f v θ ð s z ʃ ʧ ʤ l r s pa rs e f re qu en t f m ; ŋ k g θ ð ʃ ʧ l r s pa rs e f re qu en t l m ; ŋ v θ ð ʃ ʧ ʤ l r j h s pa rs e in fr eq ue nt l f ; f v θ d s z ʃ ʧ ʤ r s pa rs e in fr eq ue nt s n o t e : g f t a - = g ol dm an -f ri st oe t es t o f a rt ic ul at io n– (g ol dm an & f ri st oe , ) s ta nd ar d sc or e. j child lang. author manuscript; available in pmc september . n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript n ih -p a a uthor m anuscript gierut and morrisette page ta b le s um m ar y of b lo ck ed a nd o rt ho go na l re su lt s ba se d on t he c ri te ri on c ut -o ff o f % m ea n ge ne ra li za ti on a cc ur ac y in c od in g ye s/ no g en er al iz at io n (e lb er t et al ., ), t he r an k pr oc ed ur e fo r re pl ic at ed a b d es ig n (m ar as cu il o & s er li n, ) , a nd e ff ec t si ze f or s in gl e- su bj ec t de si gn ( b ee so n & r ob ey , ) c on d it io n c ri te ri on c u t- of f r an k p ro ce d u re e ff ec t si ze m % a cc u ra cy c od in g n /n p d b lo ck ed d en se % y es / . . s pa rs e % y es / . . f re qu en t % y es / . . in fr eq ue nt % y es / . . o rt ho go na l d en se –f re q % y es / . . d en se –i nf re q % y es / . s pa rs e– f re q % n o / . s pa rs e– in fr eq % y es / . n = n um be r of f ac to rs e vi de nc in g po si ti ve g ai n; n = n um be r of p os si bl e fa ct or s ex am in ed f or p os it iv e ga in ( i. e. n um be r of s ub je ct × t re at ed a nd u nt re at ed s ou nd s fr om t he s am e an d di ff er en t m an ne r cl as se s; se e m or ri se tt e & g ie ru t, : ). i na de qu at e po w er d ue t o th e nu m be r of f ac to rs e vi de nc in g po si ti ve g ai n by c hi ld re n en ro ll ed i n th es e co nd it io ns . j child lang. author manuscript; available in pmc september . crus- - - .. journal of crustacean biology, ( ): – , conservation of imperiled crayfish—orconectes (faxonius) indianensis hay (decapoda: cambaridae) thomas p. simon and roger f. thoma (tps) u.s. fish and wildlife service, south walker street, bloomington, in - , u.s.a; and crustacean collection, aquatic research center, indiana biological survey, south fairfax road, bloomington, indiana , u.s.a. (thomas_simon@fws.gov); (rft) crustacean range, museum of biodiversity, the ohio state university, kinnear drive, columbus, ohio - , u.s.a. (bartonii@adelphia.com) introduction common name.—indiana crayfish (simon, ). conservation status.—rare (taylor et al., ). based on the species’ limited distribution (, , km ) and the loss of populations, as reported in this study, o. indianensis has been classified as vulnerable, vu b b(iv), following iucn ( ). identification.—the indiana crayfish (fig. ), orconectes indianensis, was first described by hay ( ) from the patoka river near the town of patoka, gibson county. the species is a member of the subgenus faxonius and displays the subgenus’ characteristic first form male gonopod having both a short mesial process and central projection. the two rami are strongly divergent in both first and second form males, the rostrum is fixed with a median, distal carina, a strongly developed cervical spine is found on each side of the cervical groove, and the annulus ventralis displays bilateral anterior knobs. natural history distribution.—the geographic range of orconectes in- dianensis is restricted (fig. ) and has been considered ‘‘rare’’ or ‘‘special concern’’ (page, ; page, ; simon, ; taylor et al., ). hobbs ( ) reports that the species is endemic to southeastern illinois and south- western indiana, but it has been extirpated from a large portion of its historic range in illinois (page, ; page, ) and has been reported from very few sites in indiana (hay, ; eberly, ; page and mottesi, ). rhoades ( ) described the historic and current range of o. indianensis in relation to glaciation. the species was considered to have been present throughout the entire wabash river drainage and the dominant crayfish in the white river prior to the illinoian glacier. the farthest advance of the illinoian glacier filled the lower white river drainage causing changes at the ice front as it formed three marginal lakes that supported o. indianensis populations. effluents later connected the three lakes and formed the patoka river and other illinoian drainages, the little pigeon and anderson river drainages. the wisconsin glacier did not have an effect on the distribution of the species (rhoades, ). the first records of o. indianensis from illinois were reported by rietz ( ). however, some of the localities cited in his study are believed based on misidentification as one of two other species (page, ), i.e., orconectes stannardi page, and o. propinquus (girard, ). records for the south fork saline river in gallatin county and the north fork saline river in saline county are considered valid, but additional records for the north fork saline river remain unverified. brown ( ) provided additional distribution records for o. indianensis in cypress ditch, gallatin county. page ( ) found a similar distribution for the species reported by brown ( ) during a statewide survey in illinois from to . additional records were found for williamson, johnson, saline, pope, and gallatin counties, and ohio river drainages in hardin county. additional records for sugar creek (saline river drainage) in williamson county, and brushy slough (ohio river drainage) in white county were found in museum collections. in , page ( ) sampled the same sites in illinois where historical collections had been made. the species was found at all but three sites sampled during as they were during - . the species was not collected from rock creek (hardin county), rock branch battle fork creek (saline county), and brushy slough (white county). page ( ) stated that although the illinois population does not appear to be declining, the reduction of the species’ historical range and threats to the main population requires state protection. in illinois, the species is stable but remains in a precarious state, since the majority of populations are found in the saline river system, now degraded by oil and gas exploration, coal mining, and acid mine drainage (smith, ; page, ; simon et al., ). orconectes indianensis significant range reduction is a result of changing land use (smith, ; page, ). in indiana, orconectes indianensis is found in extreme southwestern indiana in the patoka river system, a tributary of the wabash river; in the black river, a small tributary of the wabash river in posey county; and in ohio river tributaries from pigeon creek, vanderburgh county to anderson river in perry county (page, ; page and mottesi ). page ( ) indicated that the patoka river was one of the last remaining watersheds that possessed populations of o. indianensis. several studies indicate that this watershed is at risk of degradation similar to that in the saline river watershed in illinois (simon et al., ; simon et al., ). we confirmed that o. indianensis distribution is currently limited to the patoka and black river systems (wabash river drainages) and pigeon creek and anderson river (ohio river drainage) in southwestern indiana and the saline river and honey creek systems (ohio river drainage) in southeastern illinois. these six basins encompass approximately km total drainage area. hay ( ) first described the species from the patoka river, gibson county, indiana. rhoades ( ) reported collecting specimens in the patoka river from gibson, pike, dubois, and orange counties; the little pigeon river, warrick county; and anderson river, perry county, indiana. specimens were also found by page ( ) in the united states national museum, the ohio state museum of biodiversity, and the illinois natural history survey. these new museum collections validate records for spencer, posey, and vanderburgh counties. page ( ) reported that o. indianensis was extirpated from gibson and dubois counties; however, our surveys during - found that the species is still extant in these areas. page ( ) believed that the species status was stable in indiana. surveys conducted in indiana during - found additional localities in the upper patoka river on public lands (fig. . ). individuals of o. indianensis were found at sites ( . % of sites collected); however, only two of the historic sites surveyed by page ( ) retained populations. abundance.—the relative abundance of o. indianensis is high ( . individuals/m ) in streams with cobble, rubble bottoms (r ¼ . ) and low sediment loads, and absent or low abundance having only - individuals in heavily sedimented streams (fig. . ). smaller headwater and intermittent streams produced the largest numbers of crayfish. the largest number of o. indianensis collected at a single site was individuals ( . crayfish/m ). habitat and ecology.—orconectes indianensis inhabits streams with coarse substrates primarily of large rocks or woody debris, and slow to moderate current (page, ; page, ). our studies found orconectes indianensis to prefer high gradient streams having riffles, runs, and pools with clear water of moderate nutrient levels. presence of o. indianensis is strongly correlated with the presence of rock cobble or boulder habitat (table ). the preferred stream size includes perennial headwater and intermittent streams; however, the species also occurs in the main channel of the patoka river at the type locality and further downstream near the mouth of the patoka river near the confluence with the wabash river. this extensively channelized site, characterized by riffle habitat, is uncharacteristic of the lower patoka river. the species has not been collected from areas that are heavily silted, which is especially common in channelized stream reaches. orconectes indianensis is a tertiary burrower (hobbs, ) constructing simple chambers under rocks during the warmer, dry months of the year. reproduction.—page ( ) recorded the presence of ovigerous females during april and may. we observed first form males and females (wetzel, ) during only a short reproductive season in late fall and early spring. two ovigerous females carried and eggs total and five females with attached instars carried to individuals (�x ¼ . ). conservation factors threats.—stream modifications, especially channeliza- tion; increased nutrient and sediment loads; strip mining activities and their consequent influence on ph changes threaten this species. the north and middle forks of the saline river are extremely polluted as a result of water quality degradation from strip-mining, oil-field pollution, siltation, and desiccation (smith, ). rhoades ( ) suggested that the patoka river was the stronghold for the species. currently, coal mining, stream channel clearing, channelization, and poor water quality threaten this water- shed. in the patoka river drainage, eight of the eleven sites sampled by page ( ) failed to provide continued fig. . orconectes indianensis hay in rock habitat. simon and thoma: conservation of orconectes indianensis fig. . ) present and historical distribution of orconectes indianensis hay in illinois and indiana. large open circles ¼ and earlier; medium open circles¼ - ; solid circles¼ - . ) abundance of orconectes indianensis hay in illinois and indiana. increasing dot size represents increasing relative abundance captured at each site. small solid dots ¼ - crayfish; medium hollow dots¼ - crayfish; medium solid dots¼ - crayfish; large hollow dots ¼ - crayfish; and large solid dots ¼ - crayfish. journal of crustacean biology, vol. , no. , evidence of o. indianensis during our - surveys. two of the eight sites were badly polluted. efforts by the patoka national wildlife refuge to reduce water quality impacts of acid mine drainage and oil brine along the perimeter of the refuge should increase available habitat. orconectes rusticus (girard, ) invasion of streams occupied by o. indianensis will likely result in drastic reductions of o. indianensis and confinement of populations to a few small, low nutrient headwater streams in the basin. conservation action.—orconectes indianensis is presently not considered federal endangered and is not state listed by indiana (simon, ). page ( ) reported that a total of historic localities were sampled in both indiana and illinois in . six of the sites no longer have documented populations, three have declining populations, and have stable populations. sites sampled during - showed that only two of the nine sites previously possessing indiana crayfish remained stable in indiana. the remaining sites no longer support populations (simon and thoma, ). this loss of historic populations is a troubling trend for the future of the species. orconectes indianensis has a narrow distribution and large areas of its range have been impacted in ways that have greatly reduced its abundance. further non- point impacts and habitat modifications will further reduce the abundance of o. indianensis and the establish- ment of the rusty crayfish (o. rusticus) could cause the species to slip into the endangered range. continued moni- toring of the species is required to maintain knowledge of its status. conservation recommendations.—streams with suitable water quality would benefit from the placement of rock- rubble substrates in the stream. the addition of rock-rubble substrates will greatly enhance the potential for establish- ment and expansion of o. indianensis populations. pre- vention of non-point run off by employing wooded and grassed buffer strips on stream and ditch edges will also enhance o. indianensis habitat. further efforts to secure additional public lands along the edges of hoosier national forest, the u.s. fish and wildlife service patoka river national wildlife refuge, and land owned by the army corps of engineers would benefit the long-term stability of this species. remarks.—orconectes indianensis was found to be more widely distributed and more abundant than initially con- sidered prior to conducting this survey (fig. . ). new populations found during surveys conducted during - in the headwaters of the patoka river drainage include numerous sites in hoosier national forest, patoka national wildlife refuge, and on public land owned by the army corps of engineers. it does not meet criteria to be designated federal endangered. threats posed by potential introduction of rusty crayfish to streams inhabited by o. indianensis are considered to be extreme. similar to populations of orconectes propinquus or orconectes sanbornii (faxon, ) observed in ohio (thoma and jezerinac, ), o. indianensis shows a pro- pensity to prosper in small first order streams with low nutrient levels. such areas in ohio have proven to be important refuges for the two species as o. rusticus does not prosper in waters low in nutrient that lack aquatic plant growth such as filamentous algae (thoma and jezerinac, ). conclusions the status of orconectes indianensis is more secure than previously thought in indiana and remains stable in illinois (page, ; page and mottesi, ; simon and thoma, ). this species has remained stable in illinois despite widespread degradation in the saline and patoka river drainages. we noted absence of the species in sites where populations were previously reported. we found popula- tions at additional new sites in the headwaters of the patoka river. habitat enhancement for o. indianensis can be facilitated by increasing rock-rubble coverage of stream bottoms and reducing the coverage of silt and sand. establishment of wooded riparian cover would prove to be beneficial to this species. establishment of o. rusticus populations in stream basins with o. indianensis should be immediately and completely eliminated. table . correlations of environmental variables with orconectes indianensis abundance. all variables are significant at p , . , n ¼ . correlation positively correlated variables riffle score . substrate score . qhei . cover score . overhanging vegetation . number of substrate types . % boulder/cobble . gravel . limestone . no embededness . moderate sinuosity . no channelization . riffle depth - cm . riffle embeddedness . gradient . pool score . ph . stream development good . channel score . riffles stable . cobble . pool depth . cm . temperature . boulder cobble cover . narrow riparian . negatively correlated variables specific conductivity � . salinity � . recent channelization � . poor stream development � . sparse cover � . silt abundance � . cover lacking � . bank erosion heavy � . no riffles � . silt heavy � . extensive embeddedness � . low streambed stability � . no sinuosity � . riffles unstable � . simon and thoma: conservation of orconectes indianensis acknowledgements the authors thank foster purrington for field and laboratory assistance in the collection and processing of specimens from this study. this study was funded through the u.s. fish and wildlife service contaminants program through grant number - - n to tps. the opinions expressed do not necessarily represent those of the u.s. fish and wildlife service although this study may have been funded entirely or in part by that agency. references brown, p. l. . the biology of the crayfishes of central and south- eastern illinois. doctoral dissertation, university of illinois, urbana- champaign. hay, w. p. . the crayfishes of the state of indiana. th annual report of the department of geology and natural resources of indiana : - . hobbs, h. h. jr. . the crayfishes of georgia. smithsonian contri- butions to zoology, : viii þ pp, figs - . ———. . an illustrated checklist of the american crayfishes (decapoda: astacidae, cambaridae, and parastacidae). smithsonian contributions to zoology . pp. iucn. . iucn red list categories and criteria: version . . iucn species survival commission. iucn, gland, switzerland and cam- bridge, uk. ii þ pp. page, l. m. . the crayfishes and shrimps (decapoda) of illinois. illinois natural history survey bulletin : - . ———. . the distribution and status of the indiana crayfish, orconectes indianensis. illinois natural history survey, champaign. technical report ( ). ———, and g. b. mottesi. . the distribution and status of the indiana crayfish, orconectes indianensis, with comments of the crayfishes of indiana. proceedings of the indiana academy of science : - . rhoades, r. . the evolution of crayfishes of the genus orconectes section limosus (crustacea: decapoda). the ohio journal of science : - . rietz, n. m. . ecological relations of the crawfishes of illinois. bachelor’s thesis, university of illinois, urbana-champaign. simon, t. p. . checklist of the crayfish and freshwater shrimp (decapoda) of indiana. proceedings of the indiana academy of science : - . ———, r. l. dufour, and b. e. fisher. . changes in the biological integrity of fish communities in the patoka river drainage as a result of anthropogenic disturbance from to , pp. - . in, changes in the biology of large rivers. historical changes in large river fish assemblages of north america. r. m. hughes, j. rinne, and r. calamusso (eds.). american fisheries society, bethesda, maryland. ———, s. a. sobiech, t. h. cervone, and n. e. morales. . histor- ical and present distribution of fishes in the patoka river basin: pike, gibson, and dubois counties, indiana. proceedings of the indiana academy of science : - . ———, and r. f. thoma. . distribution patterns of freshwater shrimp and crayfish (decapoda: cambaridae) in the patoka river basin of indiana. proceedings of the indiana academy of science : - . smith, p. w. . illinois streams: a classification based on their fishes and an analysis of factors responsible for disappearance of native species. illinois natural history survey biological notes . taylor, c. a., m. l. warren jr., j. f. fitzpatrick, h. h. hobbs iii, r. f. jezerinac, w. l. pflieger, and h. w. robison. . conservation status of crayfishes of the united states and canada. fisheries ( ): - . wetzel, j. e. . form alteration of adult female crayfishes of the genus orconectes (decapoda: cambaridae). american midland naturalist : - . received: august . accepted: january . journal of crustacean biology, vol. , no. , prognostic variables associated with improved outcomes in patients with stage iii nsclc treated with chemoradiation followed by consolidation pembrolizumab: a subset analysis of a phase ii study from the hoosier cancer research network lun - m an us cr ip t ac ce pt ed accepted manuscript prognostic variables associated with improved outcomes in patients with stage iii nsclc treated with chemoradiation followed by consolidation pembrolizumab: a subset analysis of a phase ii study from the hoosier cancer research network lun - bilal anouti , sandra althouse , greg durm , nasser hanna * indiana university department of internal medicine, w michigan st, indianapolis, in , usa; indiana university department of biostatistics, w. th street hits , indianapolis, in , usa; indiana university simon cancer center, barnhill dr., indianapolis, in , usa *corresponding author will be dr. nasser hanna (nhanna@iu.edu) ___________________________________________________________________ this is the author's manuscript of the article published in final edited form as: anouti, b., althouse, s., durm, g., & hanna, n. ( ). prognostic variables associated with improved outcomes in patients with stage iii nsclc treated with chemoradiation followed by consolidation pembrolizumab: a subset analysis of a phase ii study from the hoosier cancer research network lun - . clinical lung cancer, ( ). https://doi.org/ . /j.cllc. . . https://doi.org/ . /j.cllc. . . m an us cr ip t ac ce pt ed accepted manuscript abstract: introduction hcrn lun - is a phase ii trial of consolidation pembrolizumab following concurrent chemoradiation for the treatment of patients with stage iii nsclc. time to metastatic disease, progression free survival and overall survival appear superior to historical controls of chemoradiation alone. unfortunately, not all patients benefit from consolidation immunotherapy. we performed a univariate analysis evaluating variables associated with pfs, metastatic disease, and os. methods we conducted a retrospective analysis from patients enrolled on hcrn lun - . data collected included age, sex, stage, smoking status, pd-l status, >g vs <g adverse event, <g vs. >g pneumonitis, duration of pembrolizumab (< vs. > cycles), chemotherapy regimen, ps vs , time to start pembrolizumab ( - vs. - weeks from radiation), v (< % vs. > %). univariable cox regression was performed to determine the variables associated with endpoints: tmdd; pfs; and os. results from april to december , patients were enrolled and were included in the efficacy analysis ( patient was ineligible). for tmdd, improved outcomes may be associated (p< . ) with stage iiia, non-squamous cell, > cycles of pembrolizumab, and v < %. for pfs, improved outcomes (p< . ) may be seen for > cycles of pembrolizumab, and v < %. for os, improved outcomes (p< . ) may be seen for non-squamous histology, > cycles of pembrolizumab. conclusion non-squamous nsclc, longer duration of pembrolizumab, and v < % may be associated with prolonged time to metastatic disease or death, pfs, and os for patients with stage iii nsclc treated with chemoradiation followed by pembrolizumab. keywords: immunotherapy, non-small cell lung cancer, adjuvant therapy, prognosis. downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript introduction: over the last years, the year survival rate for patients with stage iiia and iiib nsclc has been - % and - %, respectively . the standard of care for these patients during this period remained concurrent chemotherapy and radiation therapy - . this standard has recently changed, based upon the results from the randomized, phase iii pacific trial. in this study, patients with stage iii nsclc were treated with a variety of chemotherapy regimens concomitantly with radiation therapy and subsequently randomized in a : fashion to receive either durvalumab, a pd-l inhibitor, or placebo for up to year. median progression-free survival for those receiving durvalumab was . compared with . months for placebo. overall survival was also improved for those receiving durvalumab with a -month overall survival rate of . % compared with . % in the placebo group . a phase ii study from the hoosier cancer research network (hcrn lun - ) evaluated the role of consolidation pembrolizumab after chemoradiation in patients with stage iii nsclc. this study recruited patient from sites in the u.s. results from this study indicated efficacy outcomes [progression free-survival (pfs), time to metastatic disease or death (tmdd), and overall survival (os)] similar to those achieved with durvalumab in the pacific trial with median tmdd of . months and a median pfs of months in a similar study population. prognostic variables associated with improved outcomes in patients with stage iii nsclc treated with chemoradiation are well characterized. performance status - , absence of weight loss in the months preceding the diagnosis of lung cancer, stage iiia (compared with iiib) disease, and volume of lung receiving at least gy of radiation (v ) < % are associated with improved outcomes - . however, prognostic variables associated with improved efficacy from consolidation immunotherapy after chemoradiation therapy have not been defined. the anti-cancer mechanisms of checkpoint inhibitors is substantially different than conventional chemotherapy. furthermore, checkpoint inhibitors are associated with a vastly different side effect profile compared with conventional chemotherapy. we, therefore, performed this retrospective subset analysis to evaluate potential prognostic variables associated with improvements in tmdd, pfs, and os in patients with stage iii nsclc treated with chemoradiation followed by pembrolizumab from the hcrn lun - clinical trial. downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript methods: study population we conducted a retrospective analysis from patients enrolled on hcrn lun - . this was a single arm, phase ii trial of concurrent chemoradiation followed by consolidation pembrolizumab mg iv every weeks until disease progression, unacceptable toxicity, or up to year of therapy was completed. patients were permitted to receive carboplatin plus paclitaxel, cisplatin plus etoposide, or cisplatin plus pemetrexed (non-squamous cell only) given concurrently with . - . gy xrt. patients were also permitted up to cycles of consolidation chemotherapy. patients with non-progression of disease were then enrolled onto the clinical trial. eligible patients had a ps of - , stage iiia or iiib squamous or non-squamous nsclc. patients with contraindications to checkpoint inhibitors were excluded, including a history of non-infectious pneumonitis or interstitial lung disease. patients were not excluded based upon pulmonary function, v or weight loss. the primary objective of the overall study was to determine if consolidation pembrolizumab following concurrent chemoradiation improves time to death or metastatic disease in inoperable stage iii nsclc patients. the efficacy and safety results of the overall study population have been reported . the primary objective of this subset analysis was to determine prognostic variables associated with tmdd, pfs and os. prognostic variables prognostic variables considered for this analysis included age, sex, stage of nsclc (iiia vs. iiib), histology (squamous vs. non-squamous), smoking status (current vs. former vs. never smoker), pd-l status (> % vs. < %), any adverse event grade ((≤ g vs. ≥ g ), pneumonitis (≤ g vs. ≥ g ), duration of treatment with pembrolizumab (< vs. ≥ cycles), chemotherapy regimen (carboplatin/paclitaxel vs. cisplatin/etoposide), ecog performance status ( vs. ), interval between completion of radiation to start of pembrolizumab ( - weeks vs. - weeks) and volume of lung receiving at least gy of radiation (v ) (≤ % vs. > %). pd-l testing was performed using modified proportion score (mps). this scoring system includes both pd-l positive mononuclear inflammatory cells (mic), including macrophages and lymphocytes, and tumor cells. it also evaluates mics within the tumor-associated stroma, though this is not factored into the percent mps score. downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript statistical analysis data analysis was conducted using sas software version . (sas institute inc., cary, nc). a p-value < . denoted statistical significance for all tests. basic comparisons were made using chi-square tests. univariable cox regression was performed to determine the variables associated with endpoints: tmdd, pfs and os. variables found to have a p < . on univariable cox regression analysis were included in a multivariable cox regression model. due to a small sample size (n= ), we chose p < . as the cutoff in the univariable regression for inclusion into the multivariable regression model, to allow us to still develop hypothesis- generating prognostic variables to be studied in larger trials in the future.. results: from april to december , patients were enrolled and were included in the efficacy analysis ( patient was ineligible). the median follow-up was . months with a minimum follow-up of months for patients who are alive. patient and disease characteristics are summarized in table . for time to metastatic disease or death (tmdd), improved outcomes (p< . ) were associated with stage iiia (p= . ), non-squamous cell (p= . ), > cycles of pembrolizumab (p< . ), and v ≤ % (p= . ) according to the univariable cox regression model. a multivariable model was run with those variables and stage iiia (hr= . , p= . ) and > cycles of pembrolizumab (hr= . , p< . ) remained significant. on the other hand, smoking status was not associated with a difference in tmdd (p= . ) on univariable cox regression model. pdl- status was also not associated with tmdd (p= . ) (table ). for pfs, improved outcomes (p< . ) may be seen for > cycles of pembrolizumab (p< . ), and v ≤ % (p= . ) according to the univariable cox regression model. a multivariable model was run with those variables and > cycles of pembrolizumab (hr= . , p< . ) remained significant. smoking status (p= . ), pdl- status (p= . ), and stage iiia vs. iiib (p= . ) were not associated with pfs. (table ). for os, improved outcomes (p< . ) may be seen for non-squamous histology (p= . ), and > cycles of pembrolizumab (p< . ) according to the univariable cox regression model. a multivariable model was run with those variables and > cycles of downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript pembrolizumab (hr= . , p< . ) remained significant. smoking status (p= . ), pdl- status (p= . ) and stage of disease (p= . ) were not prognostic for os (table ). table : baseline characteristics of study population (n= ) characteristics statistics age (years), mean (std) ( . ) gender, n (%) male ( ) female ( ) stage, n (%) iiia ( ) iiib ( ) histology of nsclc, n (%) non-squamous ( ) squamous ( ) distant metastasis or death on follow-up*, n (%) present ( ) not present ( ) *one patient did not have an evaluable response for distant metastasis or death table : basic comparisons of outcomes variable tmdd pfs os iiia (n= ) iiib (n= ) % p= . % % p= . % % p= . % non-scca (n= ) scca (n= ) % p= . % % p= . % % p= . % pd-l [-] (n= ) pd-l [+] (n= ) % p= . % % p= . % % p= . % downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript > pembro (n= ) < pembro (n= ) % p= . % % p= . % % p= . % v ≤ % (n= ) v > % (n= ) % p= . % % p= . % % p= . % g < pneumonitis (n= ) g > pneumonitis (n= ) % p= . % % p= . % % p= . % female (n= ) male (n= ) % p= . % % p= . % % p= . % < years old (n= ) ≥ years old (n= ) % p= . % % p= . % % p= . % current smoker (n= ) former smoker (n= ) never smoker (n= ) % p= . % % % p= . % % % p= . % % any ae grade ≤ (n= ) any ae grade ≥ (n= ) % p= . % % p= . % % p= . % carboplatin/paclitaxel (n= ) cisplatin/etoposide (n= ) % p= . % % p= . % % p= . % rad interval - weeks (n= ) rad interval - weeks (n= ) % p= . % % p= . % % p= . % ecog= (n= ) ecog= (n= ) % p= . % % p= . % % p= . % table : univariate analysis for prognostic variables associated with time to metastatic disease or death univariable results multivariable results variable comparison* hazard ratio ( % ci) p-value hazard ratio ( % ci) p-value age < vs. ≥ years old . ( . , . ) . gender female vs male . ( . , . ) . downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript univariable results multivariable results variable comparison* hazard ratio ( % ci) p-value hazard ratio ( % ci) p-value stage iiib vs iiia . ( . , . ) . . ( . , . ) . histology non-squamous vs squamous . ( . , . ) . . ( . , . ) . smoking status current vs former . ( . , . ) . current vs never . ( . , . ) former vs never . ( . , . ) pdl- status negative vs positive . ( . , . ) . any adverse event grade <= vs grade >= . ( . , . ) . pneumonitis grade grade <= vs grade >= . ( . , . ) . duration of treatment with pembrolizumab < cycles vs >= cycles of treatment . ( . , . ) <. . ( . , . ) <. type of chemo carboplatin/paclitaxel vs cisplatin/etoposide . ( . , . ) . ecog vs . ( . , . ) . time between radiation to pembrolizumab - weeks vs - weeks . ( . , . ) . v <= % vs > % . ( . , . ) . . ( . , . ) . *reference category is the second category table : univariate analysis for prognostic variables associated with progression-free survival univariable results multivariable results variable comparison* hazard ratio ( % ci) p-value hazard ratio ( % ci) p-value age < vs. ≥ years old . ( . , . ) . gender female vs male . ( . , . ) . stage iiib vs iiia . ( . , . ) . downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript univariable results multivariable results variable comparison* hazard ratio ( % ci) p-value hazard ratio ( % ci) p-value histology non-squamous vs squamous . ( . , . ) . smoking status current vs former . ( . , . ) . current vs never . ( . , . ) former vs never . ( . , . ) pdl- status negative vs positive . ( . , . ) . any adverse event grade <= vs grade >= . ( . , . ) . pneumonitis grade grade <= vs grade >= . ( . , . ) . duration of treatment with pembrolizumab < cycles vs >= cycles of treatment . ( . , . ) <. . ( . , . ) . type of chemo carboplatin/paclitaxel vs cisplatin/etoposide . ( . , . ) . ecog vs . ( . , . ) . time between radiation to pembrolizumab - weeks vs - weeks . ( . , . ) . v <= % vs > % . ( . , . ) . . ( . , . ) . *reference category is the second category table : univariate analysis for prognostic variables associated with overall survival univariable results multivariable results variable comparison* hazard ratio ( % ci) p-value hazard ratio ( % ci) p-value age < vs. ≥ years old . ( . , . ) . gender female vs male . ( . , . ) . stage iiib vs iiia . ( . , . ) . histology non-squamous vs squamous . ( . , . ) . . ( . , . ) . downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript univariable results multivariable results variable comparison* hazard ratio ( % ci) p-value hazard ratio ( % ci) p-value smoking status current vs former . ( . , . ) . current vs never . ( . , . ) former vs never . ( . , . ) pdl- status negative vs positive . ( . , . ) . any adverse event grade <= vs grade >= . ( . , . ) . pneumonitis grade grade <= vs grade >= . ( . , . ) . duration of treatment with pembrolizumab < cycles vs >= cycles of treatment . ( . , . ) <. . ( . , . ) . type of chemo carboplatin/paclitaxel vs cisplatin/etoposide . ( . , . ) . ecog vs . ( . , . ) . time between radiation to pembrolizumab - weeks vs - weeks . ( . , . ) . v <= % vs > % . ( . , . ) . *reference category is the second categor downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript discussion: prognostic variables associated with better or worse outcomes for patients with stage iii nsclc treated with concurrent chemoradiation are well characterized . however, prognostic variables associated with outcomes are not well defined in this patient population receiving consolidation pd- or pd-l inhibition. the analysis performed from hcrn lun - reported here indicates that receiving or more cycles of consolidation pembrolizumab is associated with improved pfs, tmdd, and os compared with receiving fewer than cycles. the interpretation of this, however, should be undertaken with caution. it is possible (and even likely) that this simply represents a patient population who tolerated therapy well and may have had more favorable tumor biology. it should be noted, though, that duration of pembrolizumab therapy remained prognostic when factoring in other variables, including stage, pd-l score, histology, v , and ps, which can influence efficacy outcomes. the optimal duration of consolidation pd- or pd-l inhibition after chemoradiation is not defined. the pacific trial and the hcrn lun - trials each permitted up to year of consolidation checkpoint inhibition. approximately % of patients on each study were able to receive year of therapy, and the median duration of consolidation therapy received in both studies was approximately months . other known prognostic variables for patients with stage iii nsclc were evaluated in the current study. v < % is associated with less toxicity and better outcomes for patients treated with chemoradiation for stage iii nsclc . based upon the current analysis, it appears that v < % is also associated with improved pfs and tmdd with consolidation pembrolizumab. lower v may simply be a surrogate for lower volume of disease, selecting out patients who may be more curable. furthermore, non-squamous nsclc was also associated with improved survival with consolidation pembrolizumab. subset analyses from the pacific trial suggested that patients with pd-l expression of < % may not have improved survival with consolidation durvalumab compared with placebo, although pfs was improved with durvalumab in that analysis . the hcrn lun - trial utilized a pd-l testing platform that assessed pd-l expression in tumor cells along with other cells in the microenvironment. utilizing this assay, pd-l expression of < % was not associated with inferior survival with consolidation pembrolizumab. it is possible that the downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript utilization of different assays resulted in disparate results between hcrn lun - and the pacific trial. the optimal therapy for patients with pd-l scores of % remains undefined. in addition, a subset analysis from the pacific trial suggested that initiating consolidation durvalumab within weeks of completing chemoradiation was associated with improved outcomes . in contrast, patients enrolled onto hcrn lun - were required to wait a minimum of weeks and a maximum of weeks after completing chemoradiation prior to initiating pembrolizumab. the current study evaluated the differences in outcomes of patients initiating pembrolizumab - weeks after chemoradiation versus those initiating pembrolizumab - weeks after chemoradiation. no difference was detected in efficacy outcomes (pfs, tmdd, os) between these subgroups. it is possible that patients receiving durvalumab within weeks of chemoradiation on the pacific trial represented patients with fewer medical problems, improved tolerance to therapy, and better performance status. furthermore, smoking status, age, gender, chemotherapy regimen, and development of pneumonitis during consolidation immunotherapy do not appear to be associated with outcomes in this patient population. while this was the first attempt to evaluate prognostic variables utilizing immunotherapy after chemoradiation for patients with stage iii nsclc, it is important to recognize the limitations of this analysis. this analysis was retrospective and unplanned. secondly, a p value of < % was utilized as a cut-off for utilization in the multivariate model. third, the sample size of this study is relatively small. fourth, the pd-l assay utilized is not standard. fifth, pfs can be challenging to define in a patient population treated with radiation due to the difficulty of interpreting local progression. nevertheless, despite these significant limitations, this study serves as hypothesis generating to test variables in larger studies of this patient population.. in conclusion, this is the first dedicated report assessing prognostic variables associated with outcomes in patients with stage iii nsclc treated with chemoradiation followed by consolidation pembrolizumab. efforts are underway to build off the success of the pacific and hcrn lun - trials. given the increased risks and costs associated with consolidation immunotherapy in this setting, it is essential to define the optimal duration of immunotherapy and to identify prognostic markers associated with benefit (or detriment) with this strategy. one such effort is underway through the big cancer research consortium (nct ) which downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript will evaluate multiple clinical factors, radiation treatment variables, and molecular biomarkers associated with outcomes in patients treated with consolidation immunotherapy strategies. downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. m an us cr ip t ac ce pt ed accepted manuscript references: . national cancer institute nc. non-small cell lung cancer treatment. available at https://www.cancer.gov/types/lung/hp/non-small-cell-lung-treatment-pdq#section/_ . . . albain ks, crowley jj, iii att, et al. concurrent cisplatin, etoposide, and chest radiotherapy in pathologic stage iiib non–small-cell lung cancer: a southwest oncology group phase ii study, swog . journal of clinical oncology ; : - . . albain ks, rusch vw, crowley jj, et al. concurrent cisplatin/etoposide plus chest radiotherapy followed by surgery for stages iiia (n ) and iiib non-small-cell lung cancer: mature results of southwest oncology group phase ii study . journal of clinical oncology ; : - . . (nccn) nccn. nccn clinical practice guidelines in oncology. available at https://www.nccn.org/professionals/physician_gls/pdf/aml.pdf accessed february . . antonia sj, villegas a, daniel d, et al. durvalumab after chemoradiotherapy in stage iii non– small-cell lung cancer. new england journal of medicine ; : - . . berghmans t, paesmans m, sculier jp. prognostic factors in stage iii non-small cell lung cancer: a review of conventional, metabolic and new biological variables. therapeutic advances in medical oncology ; : - . . urvay se, yucel b, erdis e, et al. prognostic factors in stage iii non-small-cell lung cancer patients. asian pacific journal of cancer prevention : apjcp ; : - . . sloan ja, zhao x, novotny pj, et al. relationship between deficits in overall quality of life and non-small-cell lung cancer survival. j clin oncol ; : - . . durm ga, althouse sk, sadiq aa, et al. phase ii trial of concurrent chemoradiation with consolidation pembrolizumab in patients with unresectable stage iii non-small cell lung cancer: hoosier cancer research network lun - . journal of clinical oncology ; : - . . ramella s, trodella l, mineo tc, et al. adding ipsilateral v and v to conventional dosimetric constraints predicts radiation pneumonitis in stage iiia–b nsclc treated with combined- modality therapy. international journal of radiation oncology* biology* physics ; : - . downloaded for anonymous user (n/a) at indiana university ruth lilly medical library from clinicalkey.com by elsevier on july , . for personal use only. no other uses without permission. copyright © . elsevier inc. all rights reserved. book reviews the most important of six factors in bringing about the decision. he asserts t h a t allied propaganda accentuated many of these factors, but t h a t i t would not have borne f r u i t without receptive soil. unfortunately, except f o r t h c defense of neutral rights, the majority of hoosiers gave little thought to the relation of belligerency and american foreign policy. “their failure was of little consequence in the prosecution of the w a r , f o r they agreed well enough on the immediate target. but what was unessential in w a r might be crucial in making peace.” the farmer’s last frontiey, agriculture, - . by fred a. shannon (volume v of t h e economic history of the united states, f a r r a r & rinehart, inc., new york, c. , pp. xii, . text edition, $ . , t r a d e edition, $ . .) with the appearance of this volume a new co-operative economic history is introduced. the series is to contain nine volumes. the period before will be described i n two general volumes, the years from to will be treated in a volume devoted to agriculture and another to industry, the period from to will also be covered i n this manner, while the present century will be described in three general volumes. if the others meet the high standards set by this one, history will be enriched by a significant and thorough treatment of the economic development of the united states. the absence of a work of this type has been a notable defect of historical literature. this volume is f i r s t of all a comprehensive account of agricultural developments from the beginning of the civil w a r to the end of the century. it also summarizes i n the excellent footnote citations and in the final bibliographical chapter the scholarly literature and much of the source ma- terial concerning the subject. occasionally one fails to find a n important work included, but t h a t is rarely true. it sets f o r t h conclusions and interpretations at variance with older works particularly in relation to the national land policy. after noting t h a t the fundamental basis of the farmers’ difficulties on the last frontier were the differences of soil, climate, and distance, the author described the rapid settle- ment which was often influenced by artificial stimulants max p. allen. indiana imagazine o f history supplied by t h e railroads, states, and land companies. the corporations, railroads, cattle and lumber barons, and land companies were the chief beneficiaries of t h e land laws ac- cording to the author’s analysis. his judgments a r e reflected in a group of quotations. “in its operations the homestead act could hardly have defeated the hopes of the enthusiasts of - more completely if the makers had actually drafted i t with t h a t purpose uppermost i n mind [p. .” “it tempted settlers out to the arid stretches where a quar- t e r section was barely enough f o r t h e grazing of two or three cattle [p. .” “the [railroad] companies were not only given their railroads-they were given a bonus [usually land] to accept them [p. .” “some $ , , were paid to indian tribes to quiet their claims, and then t h e land was turned over t o speculators a t less t h a n cost [p. .” the agricultural reorganization of the south a f t e r with its labor system of tenant f a r m e r s and s h a r e croppers and its credit system of crop mortgages is also critically examined. “the outright confiscation of large p a r t s of es- tates created by slave labor, to make f a r m s f o r freedmen, would in the long r u n have created more prosperity f o r t h e section t h a n the growth of land monopolization t h a t took place instead, and t h e simultaneous establishment of a sys- tem of quasi serfdom t h a t left the toilers without ambition and t h e landlord with ruined soils and finances [pp. - ].” the classes of the south where the crop-lien system predom- inated were “the landlord-merchant-banker-capitalist group, numbering approximately a sixth of t h e total population and having all t h e political power; [and] . . . t h e bulk of field workers, living from enfeebled hand t o empty mouth [p. .” even in the north central states “a remarkable growth in tenancy” occurred which was not checked by t h e mechan- ization of the farms. the expansion of prairie agriculture in t h e great plains led to the plowing of t h e g r a s s lands t h a t should have been reserved f o r grazing. both t h e “little fellow” and the “bonanza f a r m e r ” often failed. the live- stock frontier ,was another story of injustice a n d special privilege, and the cowboy was f a r less romantic t h a n usually pictured. problems of the west a r e discussed in the light of the agrarian revolt and t h e co-operative movement which a r e also described sympathetically. eastern adjustments, the declining position of the f a r m e r s in the nation, and his book reviews social life a r e not overlooked in completing the story. perhaps some views of t h e author have been over- emphasized, but one can scarcely avoid t h e feeling t h a t he has offered a few judgments t h a t a r e somewhat extreme in his treatment of the great plains and the south. his analysis is skillful and scholarly and the total picture may not be overdrawn. the illustrations and charts a r e excel- lent. the binding of the textbook edition is not attractive and is so light in color t h a t it will soil readily and become less attractive. although much of the volume is not con- cerned with t h e middle west, i t presents the background f o r understanding the agricultural development of this region. w a r , peace, and nonresistance. by guy franklin hersh- berger. (the herald press, scottdale, pennsylvania, , pp. xv, . $ . .) the history of nonresistance and t h e scriptural and doctrinal teachings upon which it is founded f o r m the sub- ject of this volume. i t s preparation was undertaken some years ago a s a commission from t h e peace problems commit- tee of the mennonite church. its purposes were t o clarify the position of the church from its beginning to the present war and to strengthen the members in maintaining t h a t position a s individuals. the author i s professor of history and sociology a t goshen college, a mennonite college at goshen, indiana. the work is significant to the historian as a n historical account of the mennonites’ reaction against w a r from t h e sixteenth century to the present, and a s a resume of other important pacifist attitudes, but more especially as a means of understanding the mennonites and their unique position in indiana and in other states and nations. some historical material is included in the f i r s t f o u r chapters which give the doctrinal basis of nonresistance. these chapters review “war in history,” “peace a n d w a r in the old testament,” “nonresistance in the new testament,” and peace and war and the church. the mennonites trace their origin to the anabaptists in switzerland in . f r o m here they spread to holland, germany, france, and russia, often fleeing from persecution. those most interested in preserving the nonresident way of life came t o america, john d. barnhart. wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ vol. , - , april clinical cancer research the abbreviations used are: -fu, -fluorouracil; ncctg, north central cancer treatment group: kps. karnofsky performance status. prospective randomized trial of -fluorouracil versus -fluorouracil plus levamisole in the treatment of metastatic colorectal cancer: a hoosier oncology group trial’ malik t. bandealy, ren#{ }gonin, patrick j. loehrer, frank monaco, and lawrence h. einhorn divisions of hematology/oncology [m. t. b., p. j. l., l. h. e.] and biostatistics [r. g.], indiana university, and hoosier oncology group [f. m.], indianapolis, indiana - abstract the purpose of this study was to compare the objective response rate, duration of remission, and survival of -flu- orouracil ( -fu) versus those of -fu plus levamisole in metastatic colorectal cancer using the same dose and sched- ule of these agents as in the north central cancer treat- ment group and intergroup studies of adjuvant therapy. patients with no prior history of chemotherapy for meta- static disease were entered on this hoosier oncology group randomized phase iii trial. patients were stratified by karnofsky performance status and presence or absence of liver metastases. they were randomized to receive mg/rn -fu i.v. for days followed by mg/kg i.v. weekly (arm ) or the same dose of -fu plus levamisole mg p.o. every h for days every weeks (arm ). the duration of treatment for both arms was weeks. from april to march , patients were entered. one hundred eighty-two patients, in each arm, were fully evaluable. the response rates were % on arm and % on arm . the median duration of response was weeks on both arms. the median survival was weeks on arm and weeks on arm (p = . ). this study failed to show any improvement in survival, response, or duration of remission with the addition of levamisole to -fu in the treatment of metastatic colorectal cancer. received / / ; revised / / ; accepted / / . the costs of publication of this article were defrayed in part by the payment of page charges. this article must therefore be hereby marked advertisement in accordance with u.s.c. section solely to indicate this fact. i this work was supported in part by national cancer institute grant r ca - and the walther cancer institute and was presented in part at the annual meeting of the american society of clinical oncology, held may - , in philadelphia, pa. to whom requests for reprints should be addressed, at hoosier on- cobogy group, north meridian street, indianapolis, in - . introduction colorectal carcinoma is the second leading cause of cancer death in the united states. in , it was estimated that , new cases of colorectal cancer would be diagnosed and that , men and women would die from this disease ( ). heidelberger and colleagues ( ) at the university of wis- consin first reported clinical trials with -fu in . since then, -fu has been the mainstay of therapy for metastatic coborectal cancer. however, the results with single agent -fu reveal only a - % objective response rate, with no impact on survival. attempts to improve upon this have lead to devel- opment of various regimens using biochemical modulation of -fu, based upon laboratory and preclinical data. the most popular of these has been the combination of -fu and leucov- orin, based upon a sound preclinical pharmacological rationale. there have been various schedules of -fu in combination with leucovorin that have demonstrated encouraging results in phase ii trials ( ). subsequently, these have led to phase iii studies in metastatic coborectal cancer evaluating -fu (as the control arm) compared to -fu plus leucovorin. there has been a lack of congruence in the results from these disparate phase iii studies ( ). review of the published randomized studies comparing -fu with -fu plus leucovorin reveals that the consistently higher response rate of the combination regimen has failed to translate into significant survival advantage ( - ). these studies all used -fu given by iv. push. only three studies have shown marginal survival advantage for the combi- nation over -fu alone ( , , ). in fact, in one study, the benefit was significant only for patients with nonmeasurable and, hence, probably low volume disease ( ). even a meta- analysis failed to show any significant improvement in survival with -fu plus leucovorin ( ). furthermore, combining -fu with other agents, such as cisplatin, methotrexate, pala, and thymidine, has not improved the survival statistics in metastatic coborectal cancer ( - ). levamisole is a pure chemical that was first used as an antihelminthic drug and has documented immunomodulatory effects. verhacgen et a!. ( ) initially evaluated levamisole versus placebo in a very small patient population in surgically resected colon cancer and demonstrated a statistically signifi- cant survival advantage. the western cancer study group and the european organization for research and treatment of can- cer compared levamisole to placebo in the adjuvant setting and found no significant survival benefit ( , ). on april , . © american association for cancer research.clincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ -fu +/- levamisole in metastatic colorectal cancer the major interest in levamisole as adjuvant therapy of duke’s b and c colon cancer is because of a positive ncctg study ( ) and a larger confirmatory intergroup study ( ), both demonstrating a significant advantage for -fu plus levamisole over levamisole alone. however, this dose and schedule of -fu plus levamisole has never been adequately tested in metastatic disease. surprisingly, there have been only three published phase iii studies of -fu plus levamisole in metastatic coborec- tal cancer. borden et al. ( ) at the university of wisconsin published a positive study demonstrating a survival advantage for -fu plus levamisole, whereas the ncctg study was neg- ative ( ). however, neither of these studies these used the same dosage and schedule as in the successful adjuvant trials. the study by the eastern cooperative oncology group, published only in abstract form thus far, was also negative ( ). the dose and schedule of -fu and levamisole in this trial were also different from those used in the adjuvant studies. there are few, if any, examples in oncology in which a regimen that provides a significant survival advantage in the adjuvant setting is not also effective in metastatic disease. therefore, the hoosier oncology group embarked upon this study evaluating -fu plus levamisole in previously untreated patients with metastatic colorectal cancer, using dosage and schedule of these agents that were identical to those in the adjuvant trials. patients and methods all patients had histologically proven adenocarcinoma of the colon or rectum and had not received prior chemotherapy for metastatic disease. patients may have had adjuvant therapy but not within the months prior to registration. patients with measurable as well as nonmeasurable disease were included. measurable disease was defined as being bidimensionally meas- urable on physical or roentgenographic examination. malignant hepatomegaly could be used as measurable disease if the liver edge was palpable at beast cm below the costal margin at the right midclavicubar line or xiphoid on quiet respiration. the liver span should have been greater than cm. all patients had a kps of at least %, as well as adequate bone marrow reserve (absolute granubocyte count of , cells/mm and platelet count of , i.l). patients were excluded if they had a history of prior malignancy, except squamous or basal cell carcinoma of the skin or carcinoma in situ of the cervix, unless they had been disease-free for at least years. pretreatment studies included complete history and physi- cal examination, complete blood count including wbc differ- ential and platelet counts, -channel sequential multiple anal- ysis, carcinoembryonic antigen measurement, chest x-ray, and abdominal computerized tomographic scan. during treatment, patients were evaluated by history and physical examination every weeks, and radiographs were repeated every weeks for tumor measurement. in addition, a complete blood count includ- ing wbc differential and platelet count was obtained every week. all patients provided written informed consent. eligible patients were stratified by kps ( - versus - ) and presence or absence of liver metastases. patients were then randomized by telephone through the hoosier oncology group office to receive -fu alone or -fu plus levamisole. -fu was administered at a dose of mg/m by iv. push on days - ; thereafter, beginning on day , it was given weekly at mg/kg by iv. push. this was identical for both arms. patients randomized to the combination arm received mg of levamisole p.o., every h for days, every weeks. the total duration of therapy was weeks. there was no dose escalation of -fu, and cross-over to -fu plus levamisole was not allowed. all patients with prior pelvic irradiation received a % reduction in the -fu dose initially. -fu was given as sched- uled if the wbc count was greater than cells/mm , and platelet count was greater than , /gil. if counts were below these levels, treatment was held until these levels were reached, and then it was initiated with % reduction in the -fu dose. if -fu was delayed, levamisole was concomitantly delayed, but its dose was not reduced. in cases of severe diarrhea or stomatitis lasting more than days, granubocytopenic fever, or thrombocytopenia requiring platelet transfusion, the dose of -fu was reduced by % with no change in the bevamisole dose. a complete response was defined as complete disappear- ance of all objective evidence of disease. a partial response was defined as a decrease of % or more in the sum of products of diameters of measurable disease. when malignant hepatomeg- aly was used for tumor measurement, at least a % decrease in the sum of the measurements of the liver edge below the costal margin at the right midclavicular line and the xiphoid process was required for a pr. a decrease of less than % in the sum of product of diameters of measurable disease was considered stable disease provided no new lesions had appeared, and an increase of % or more or appearance of new lesions defined progression. survival was measured from the date entered on study until death. remission duration was measured from the date of doe- umented response until relapse. patients who relapsed were treated at the investigator’s discretion. statistical considerations. descriptive statistics are re- ported as frequencies, medians, and ranges. grade / toxicities were compared between the two arms ( -hi versus -fu + levamisole) using fisher’s exact test ( ). survival was defined from date on study until date of death or date last known alive. follow-up was the survival of all patients alive. the survival curve for each treatment arm was constructed using the kaplan- meier product limit method ( ). survival curves between the two treatment arms were compared with the log-rank test of survival ( ). the association of treatment arm and survival, adjusting for the stratification factors, liver metastases and kps, was assessed by the wald x of cox’s proportional hazards regression ( ). all tests were two-sided. the intention-to-treat principle was used throughout. results from april through march , patients with metastatic adenocarcinoma of the colon or rectum were entered on this trial. of the patients randomized to -fu, patients never received treatment, was lost to follow-up, and had on april , . © american association for cancer research.clincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ table patient characteristics ap . . bp . . c p treatment-related only. a pd, progressive disease; psr, poor subjective response (� % fall in kps, % weight loss, and increased pain without radiographic documentation of progression). b not rebated to disease or therapy. clinical cancer research -hi -hi + levamisole no. of patients registered sex (male/female) / / median age (yr) range (yr) - - kps - - prior pelvic ri”' metastatic sites liver lung other a rt, radiotherapy. table treatment compliance -fu -nj + levamisole (n ) (n� ) completed treatment reasons for discontinuation pd” psr toxicity noncompliance death” inadequate response data. one hundred patients were random- ized to -hi plus levamisole, of whom never received treat- ment and s had inadequate response data. in all, patients were fully evaluable for survival and response, and another patients were evaluable for survival only. patient characteristics are shown in table . patients on the two arms were well matched for age, kps, presence or absence of liver metastases, and prior pelvic radiotherapy. thirty-two patients on the -fu arm and on the -fu plus levamisole arm completed weeks of treatment. reasons for discontinuation of therapy are listed in table . the most frequent reason for early termination of chemotherapy was secondary to tumor progression during therapy. toxicity. the toxicity data are shown in table . this table reflects the worst grade of toxicity experienced by each patient at any time while in the study. the higher incidence of grade / diarrhea in the -fu plus levamisole arm was statis- tically significant (p = . ), whereas the difference in the incidence of grade / granulocytopenia between the two arms did not reach statistical significance (p = . ). there were no treatment-related deaths on the -fu arm, whereas there were four on the combination arm. response and survival. there were objective re- sponses in fully evaluable patients. there was one complete response on each arm. overall, the response rate was % ( % ci = - %) on the -fu arm and % ( % ci = - %) on the -ri plus levamisole arm. the median duration of response table toxicity (grade / ) percentage of patients -fu -hi + levamisole anemia granulocytopenia” thrombocytopenia nausea s s vomiting diarrhea” stomatitis infection fever neurological death’ on both arms was weeks. thirty-eight patients on the -hi arm and patients on the combination arm had stable disease. the median time to progression was weeks on the -hi arm and weeks on the -hi plus levamisole arm (p - . ). median survival was weeks on the -hi arm and weeks on the -hi plus levamisole arm (fig. ). survival times on the two arms were not significantly different by the log-rank test (p = . ). in the cox proportional hazards regression, the wald x was . (p = . ; degree of freedom) for treatment arm, adjusting for the stratification factors, liver me- tastases and kps. at the time of analysis of this data, patients on the -hi arm and on the combination arm were alive. discussion one of the basic principles of medical oncology is that the use of a regimen to improve the surgical cure rate in a malig- nancy is preceded by documented efficacy in advanced disease. however, the choice of -hi plus levamisole in the adjuvant ncctg trial was quite empirical ( ). this dose and schedule of these agents had never been adequately tested in metastatic coborectal cancer. the university of wisconsin randomized study of -hi versus -hi plus levamisole showed a statistically significant advantage for the combination ( ). however, the dosage and schedule of these agents was different from that used in the successful adjuvant trials. furthermore, only of patients on this study had measurable disease. adjustment by a cox regression model for the influence of measurable disease re- duced the significance of treatment difference. the ncctg study in metastatic coborectal cancer was a -arm trial testing -hi against various combinations with -fu and included a -hi plus levamisole arm ( ). again, the dosage and schedule were different from those used in their subsequent adjuvant study. this study failed to demonstrate superiority of any of the combination arms over -hi alone. remarkably, -hi in this study had a response rate of %, which is unusually high for -hi as a single agent in metastatic coborectal cancer. the eastern cooperative oncology group has reported a trial of -hi versus -hi plus levamisole in meta- static colorectal cancer in patients with nonmeasurabbe disease on april , . © american association for cancer research.clincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ log rank test: p = . median -fu +/- levamisole in metastatic colorectal cancer a i. . . . . . survival in weeks arm cnsr death tot - -fu -fu+levamisole fig. i survival. and found no benefit for the combination ( ). however, this study has been reported in abstract form only, and the doses are different from those used in the ncctg ( ) and intergroup ( ) studies. therefore, in the adjuvant setting, the success of the -hi plus levamisole regimen in the ncctg trial ( ) was quite surprising and bed to some skepticism. however, the results of the larger intergroup study ( ) established -hi plus levamisole as standard adjuvant therapy in completely resected stage iii colorectal cancer. it is unusual for a regimen to prolong survival in the adjuvant setting and have no substantive activity in metastatic disease. this trial included patients with metastatic colorectal cancer, most of whom had good performance status. both of the patient groups were well matched. the combination of -fu and levamisole failed to demonstrate any superiority over -fu alone in terms of response rate, duration of remission, or sur- vival. another interesting aspect of this trial is the toxicity of -hi plus levamisole. in prior studies in coborectal cancer, levamisole has not been shown to have significant hematolog- ical toxicity, and only minor nonhematological toxicities such as mild nausea, stomatitis, and diarrhea have been reported. treat- ment-related deaths were rare in these studies. even trials that have tested the -hi plus levamisole combination have not reported any increase in the usual -hi toxicity with the com- bination. longr#{ }eet al. ( ), in their studies of adjuvant therapy in duke’s b and c coborectal cancer, found a higher incidence of grade iii and iv. granulocytopenia when levamisole was added to -hi. in our study, the incidence of grade / diarrhea was significantly higher in the -fu plus levamisole arm, and there was a trend toward higher incidence of grade / granu- locytopenia as well. treatment compliance was, however, the same as for the -hi arm. this study indirectly raises questions about the role of levamisole in adjuvant therapy of colon cancer. there have been speculations that levamisole has no role at all and that the successful adjuvant studies demonstrate the effectiveness of more intense administration of -fu in properly selected pa- tients ( ). the ncctg and intergroup studies of adjuvant therapy compared -hi plus levamisole to levamisole alone and did not include a -hi alone arm ( , ). more recently, the national surgical adjuvant breast and bowel project reported the results of their adjuvant trial comparing -hi plus leucov- orin versus -hi plus levamisole versus -hi plus leucovorin plus levamisole ( ). the -hi plus leucovorin arm was ad- ministered in -day cycles for six courses, whereas the levami- sole containing arms were administered for a period of year. there was a statistically significant advantage for -fu plus leucovorin over -hi plus levamisole in terms of disease-free (p < . ) and overall (p < . ) survival. the addition of levamisole did not improve the results of -hi plus leucovorin. two other trials evaluated -hi plus levamisole. the ncctg evaluated -fu/levamisole ( months versus months) versus -fu/leucovorin/levamisole ( months versus months) and demonstrated that months of -hi/levamisole was inferior to months of -hilleucovorinllevamisole ( ). the eastern cooperative oncology group evaluated several different regimens, including high versus low dose leucovorin plus -hi and -fu/levamisole and -hillevamisolelleucovo- rim. six months of -hi + leucovorin was equivalent, if not superior to, months of -hi + levamisole, and the addition of levamisole to -hi + leucovorin did not provide any addi- tional benefit ( ). in conclusion, this cooperative group failed to demonstrate any advantage to the addition of levamisole to -hi in the treatment of metastatic colorectal cancer. the role, if any, of levamisole in duke’s c colorectal cancer is further questioned by this negative study in metastatic disease. appendix study participants: arnett clinic, lafayette, in; ball memorial hospital, muncie, in; community hospitals, indianapolis, in; consultants in blood disorders and cancer, louisville, ky; lewis domke, m.d., merrillville, in; juan correa, m.d., terre haute, in; ray drasga, m.d., merrillville, in; elkhart clinic, elkhart, in; edward fox, m.d., evansville, in; fort wayne hematology/oncology, fort wayne, in; galesburg clinic, galesburg, il; georgia hematology/oncology, atlanta, ga; alan grosbach, m.d., shreveport, la; indiana university, indianapolis, in; johnson memorial hospital, franklin, in; thomas lutz, m.d., evansville, in; memorial hospital, south bend, in; methodist hospital, indianapolis, in; st. vincent’s hospital, indianapolis, in; and washington university, st. louis, mo. references . parker, s. l., tong, t., bolden, s., and wingo, p. a. cancer statistics, . ca cancer j. clin., : - , . on april , . © american association for cancer research.clincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ clinical cancer research . heidelberger, c., chaudhari, n. k., danneberg, p., et al. fluorinated pyrimidine. a new class of tumor-inhibitory compounds. nature (lond.), : - , . . grem, j. l., hoth, d. f., hamilton, j. m., et al. overview of current status and future direction of clinical trials with -fluorouracil in com- bination with folinic acid. cancer treat. rep., : - , . . einhorn, l. h. improvements in -hi chemotherapy? j. clin. oncol., : - , . . nobile, m. t., rosso, r., sertoli, m. r., et al. randomized compar- ison of weekly bolus -fluorouracil with or without leucovorin in metastatic coborectal carcinoma. eur. j. cancer, a: - , . . petrelli, n., douglass, h. ., herrera, l., et a!. the modulation of fluorouracil with leucovorin in metastatic coborectal carcinoma: a pro- spective randomized phase iii trial. j. clin. oncol., : - , . . petrelli, n., herrera, l., rustum, y., burke, p., creaven, p., stulc, j., emrich, l. j., and mittelman, a. a prospective randomized trial of -fluorouracil versus -fluorouracil and high-dose leucovorin versus -fluorouracib and methotrexate in previously untreated patients with advanced colorectal carcinoma. j. clin. oncol., : - , . . valone, f. h., friedman, m. a., wittlinger, p. s., drakes, t., eisenberg, p. d., malec, m., hannigan, j. f., and brown, b. w., jr. treatment of patients with advanced colorectal carcinomas with flu- orouracil alone, high dose leucovorin plus fluorouracil, or sequential methotrexate, fluorouracil, and leucovorin: a randomized trial of the northern california oncology group. j. clin. oncol., : - , . . di costanzo, f., bartolucci, r., calabresi, f., sofra, m., marzola, m., belsanti, v., boni, c., and bacchi, m., fluorouracil-alone versus high-dose folinic acid and fluorouracil in advanced coborectal cancer: a randomized trial of the italian oncology group for clinical research (goirc). ann. oncol., : - , . . labianca, r., pancera. g., aitini, e., et al. folinic acid + - fluorouracil versus equidose -hi in advanced colorectal cancer. phase ill study of “giscad” (italian group for the study of digestive tract cancer). ann. oncol., : - , . . poon, m. a., o’connell, m. j., moertel, c. g., et al. biochemical modulation of fluorouracil: evidence of significant improvement of survival and quality of life in patients with advanced coborectal carci- noma. j. clin. oncol., : - , . . doroshow, j. h., multhauf, p., leong, l., et al. prospective ran- domized comparison of fluorouracil versus fluorouracil and high-dose continuous infusion leucovorin calcium for the treatment of advanced measurable coborectal cancer in patients previously unexposed to chem- otherapy. j. clin. oncol., : - , . . erlichman, c., fine, s., wong, a., and elhakim, t. a randomized trial of fluorouracil and folinic acid in patients with metastatic colorectal carcinoma. j. clin. oncol., : - , . . martoni, a., cricca, a., guaraldi, m., covizzi, m., farris, a., and pannuti, f. weekly regimen of -hi versus -hi + intermediate dose folimc acid in the treatment of advanced coborectal cancer. anticancer res., : - , . . petrioli, r., lorenzi, m., aquino, a., et al. treatment of advanced colorectal cancer with high-dose intensity folinic acid and -fluorouracil plus supportive care. eur. j. cancer, a: - , . . steinke, b., gunther, e., hirschmann, w. d., sondern, w., konic- zek, k. h., wander, h. e., natt, f., wagner, t., hinrichs, h. f., and werdier, d. fluorouracil versus folinic acid/fluorouracil in advanced colorectal cancer: preliminary results of a randomized trial. semin. oncol., : - , . . the advanced coborectal cancer meta-analysis project. modula- tion of fluorouracil by leucovorin in patients with advanced colorectal cancer: evidence in terms of response rate. j. clin. oncol., : - , . . loehrer, p. j., turner, s., kubilis, p., hui, s., correa, j., ansari, r., stephens, d., woodbum, r., and meyers, s. a prospective randomized trial of fluorouracil versus fluorouracil plus cisplatin in the treatment of metastatic coborectal cancer: a hoosier oncology group trial. j. clin. oncob., : - , . . labianca, r., pancera, g., cesana, b., et al. cisplatin + -fluorou- racil versus -fluorouracil alone in advanced colorectal cancer: a ran- domized study. eur. j. cancer clin. oncol., : - , . . buroker, . r., moertel, c. g., fleming, t. r., et al. a controlled evaluation of recent approaches to biochemical modulation or enhance- ment of -fluorouracil therapy in colorectal carcinoma. j. clin. oncob., : - , . . verhaegen, h., decree, j., decook, w., et al. levamisole therapy in patients with coborectal cancer. in: w. d. terry and s. a. rosenberg (eds.), immunotherapy of human cancer, pp. - . new york: elsevier north holland, . . chlebowski, r. t., nystrom, s., reynolds, r., et al. long-term survival following levamisole or placebo adjuvant treatment of coborec- tab cancer: a western cancer study group trial. oncology, : - , . . arnaud, j. p., buyse, m., nordlinger, b., martin, f., pector, j. c., zeitoun, p., adboff, a., and duez, n. adjuvant therapy of poor prog- nosis colon cancer with levamisole: results of an eortc double-blind randomized clinical trial. br. j. surg., : - , . . laurie, j. h., moertel, c. g., fleming, t. r., et al. surgical adjuvant therapy of large-bowel carcinoma: an evaluation of levamisole and the combination of levamisole and fluorouracil. j. clin. oncol., : - , . . moertel, c. g., fleming, t. r., macdonald, j. s., et al. fluorouracil plus levamisole as effective adjuvant therapy after resection of stage iii colon carcinoma: a final report. ann. intern. med., : - , . . borden, e. c., davis, t. e., crowley, j. j., et al. interim analysis of a trial of levamisole and -fluorouracil in metastatic colorectal carci- noma. in: w. d. terry and s. a. rosenberg (eds.), immunotherapy of human cancer, pp. - . new york: elsevier north holland, . . mansour, e. g., cnaan, a., davis, t., et al. combined modality therapy following resection of coborectal carcinoma in patients with non-measurable intra-abdominal metastases. an ecog study . proc. am. soc. clin. oncol. annu. meet., : , . . zar, j. h. biostatistical analysis ( nd ed), pp. - , - . englewood cliffs, nj: prentice-hall, . . kaplan, e. l., and meier, p. nonparametric estimation from incom- plete observations. j. am. stat. assoc., : - , . . armitage, p., and berry, g. statistical methods in medical re- search ( nd ed), pp. - . oxford: blackwell scientific publica- tions, . . cox, d. r. regression models and life-tables. j. r. stat. soc. b, : - , . . longr#{ }e, l., focan, c., bury, j., beauduin, m., brohee, d., du- vivier, a., lecomte, m., markiewicz, s., vindevoghel, a., and weerts, j. levamisole adds granulocyte toxicity to -hi based chemotherapies in adjuvant treatment of dukes b-c coborectal cancer. a preliminary report. anticancer res., : - , . . mayer, r. j. does adjuvant therapy work in colon cancer? n. engl. j. med., : - , . . wolmark, n., rockett, h., mamounas, e. p., et al. the relative efficacy of s-as + leucovorin, -hi + levamisole, and -hi + leucovorin + levamisole in patients with duke’s b and c carcinoma of the colon: first report of nsabp c- . proc. am. soc. clin. oncol. annu. meet., : , . . o’connell, m. j., laurie, j. a., shepherd, l., et al. a prospective evaluation of chemotherapy duration and regimen as surgical adjuvant treatment for high risk colon cancer: a collaborative trial of the north central cancer treatment group and the national cancer institute of canada clinical trials group. proc. am. soc. clin. oncol. annu. meet., : , . . haller, d. g., catalona, p. j., macdonald, j. s., and mayer, r. j. fluorouracil, leucovorin, and levamisole adjuvant therapy for colon cancer: four year results of int- . proc. am. soc. clin. oncol. annu. meet., : , . on april , . © american association for cancer research.clincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ ; : - . clin cancer res m t bandealy, r gonin, p j loehrer, et al. colorectal cancer: a hoosier oncology group trial. -fluorouracil plus levamisole in the treatment of metastatic prospective randomized trial of -fluorouracil versus updated version http://clincancerres.aacrjournals.org/content/ / / access the most recent version of this article at: e-mail alerts related to this article or journal.sign up to receive free email-alerts subscriptions reprints and .pubs@aacr.orgdepartment at to order reprints of this article or to subscribe to the journal, contact the aacr publications permissions rightslink site. click on "request permissions" which will take you to the copyright clearance center's (ccc) .http://clincancerres.aacrjournals.org/content/ / / to request permission to re-use all or part of this article, use this link on april , . © american association for cancer research.clincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/content/ / / http://clincancerres.aacrjournals.org/cgi/alerts mailto:pubs@aacr.org http://clincancerres.aacrjournals.org/content/ / / http://clincancerres.aacrjournals.org/ tire model with temperature effects for formula sae vehicle delft university of technology tire model with temperature effects for formula sae vehicle shyrokau, barys; harsh, diwakar doi . /app publication date document version final published version published in applied sciences citation (apa) shyrokau, b., & harsh, d. ( ). tire model with temperature effects for formula sae vehicle. applied sciences, ( ), [ ]. https://doi.org/ . /app important note to cite this publication, please use the final published version (if applicable). please check the document version above. copyright other than for strictly personal use, it is not permitted to download, forward or distribute the text or part of it, without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license such as creative commons. takedown policy please contact us and provide details if you believe this document breaches copyrights. we will remove access to the work immediately and investigate your claim. this work is downloaded from delft university of technology. for technical reasons the number of authors shown on this cover page is limited to a maximum of . https://doi.org/ . /app https://doi.org/ . /app applied sciences article tire model with temperature effects for formula sae vehicle diwakar harsh and barys shyrokau ,* rimac automobili d.o.o., sveta nedelja, croatia; diwakar.harsh@rimac-automobili.com department of cognitive robotics, delft university of technology, cd delft, the netherlands * correspondence: b.shyrokau@tudelft.nl received: october ; accepted: december ; published: december ���������� ������� abstract: formula society of automotive engineers (sae) (fsae) is a student design competition organized by sae international (previously known as the society of automotive engineers, sae). commonly, the student team performs a lap simulation as a point mass, bicycle or planar model of vehicle dynamics allow for the design of a top-level concept of the fsae vehicle. however, to design different fsae components, a full vehicle simulation is required including a comprehensive tire model. in the proposed study, the different tires of a fsae vehicle were tested at a track to parametrize the tire based on the empirical approach commonly known as the magic formula. a thermal tire model was proposed to describe the tread, carcass, and inflation gas temperatures. the magic formula was modified to incorporate the temperature effect on the force capability of a fsae tire to achieve higher accuracy in the simulation environment. considering the model validation, the several maneuvers, typical for fsae competitions, were performed. a skidpad and full lap maneuvers were chosen to simulate steady-state and transient behavior of the fsae vehicle. the full vehicle simulation results demonstrated a high correlation to the measurement data for steady-state maneuvers and limited accuracy in highly dynamic driving. in addition, the results show that neglecting temperature in the tire model results in higher root mean square error (rmse) of lateral acceleration and yaw rate. keywords: tire model; tire temperature; fsae vehicle . introduction formula sae (fsae) also known as formula student (fs) is a competition in which students design a single seat formula race car to compete against other fs teams from all over the world [ ]. the competition motivates the students for innovative solutions to demonstrate their engineering talent and obtain new skills. it also allows the students to apply theoretical knowledge into practice in a dynamic and competitive environment. since students have joined formula student team delft [ ] on an annual basis to participate in the fsae competition, earlier in the combustion class and electric class from . aiming to improve the fsae vehicle performance, the team collaborated with apollo vredestein b.v. (enschede, the netherlands) to develop original tires. the fsae vehicle ( version) from fs team delft equipped with such tires is shown in figure . appl. sci. , , ; doi: . /app www.mdpi.com/journal/applsci http://www.mdpi.com/journal/applsci http://www.mdpi.com https://orcid.org/ - - - http://www.mdpi.com/ - / / / ?type=check_update&version= http://dx.doi.org/ . /app http://www.mdpi.com/journal/applsci appl. sci. , , of appl. sci. , , x for peer review of figure . dut formula society of automotive engineers (fsae) vehicle at the formula student germany competition [ ]. to achieve the best fsae vehicle performance, the team developed various tools to evaluate vehicle dynamics and to predict its behavior. however, predictions are based on estimated friction coefficient, neglecting the thermal effect on the tire performance. frequently, lap simulation is based on a point mass or simplified vehicle models allowing for the development of the top-level concept of the new vehicle. it can quantify vehicle parameters such as the height of the center of gravity and lift coefficient. however, to design various vehicle components and, specifically, control algorithms, a full vehicle simulation including a comprehensive tire model is required. for an accurate modeling of tire forces and moments, a physical, semi- or empirical tire model can be used [ ]. the physical approach provides more insights regarding tire behavior and better represents the whole operation range; however, it lacks accuracy, specifically considering camber influence, conicity, plysteer, and other phenomena [ , ]. the empirical approach provides a higher modeling accuracy in the predefined tested range but requires extensive special tests. according to [ , ] the tire temperature, especially for racing application [ ], has a significant effect on the force capability of the tire. several advanced tire models using the brush element approach incorporated with thermal effects have been proposed [ – ]; however, their parametrization requires intensive test sessions, commonly unfeasible for fs teams. for a fsae vehicle, less complex models incorporating temperature effect are proposed using the physical approach based on the brush model [ , ]. they demonstrate a good correlation with the experimental tire data collected using the indoor flat track tire test machine [ ]. also, the proposed physical-based models were compared to the original magic formula demonstrating close accuracy; however, original magic formula does not include the effect of temperature on the force capability of the tire. thus, the goal of the proposed study was to develop an empirical tire model incorporating thermal effects and to evaluate its performance compared to the experimental measurements. the main contribution of the study is the improvement of the accuracy of the empirical tire model using the proposed thermal model and modification of the magic formula to incorporate temperature effects. the paper is structured as follows. section describes the experimental setup, test program, and experimental results. a thermal tire model is proposed in section discussing the basic thermal equations and comparing them to well-established thermal models. section presents the widely used magic formula with the extension related to the temperature effect. the complete vehicle model including thermal and modified magic formula is presented in section ; it includes the whole model simulation for steady-state and transient maneuvers compared to real tests and simulation without the temperature effect. the paper concludes with a discussion of the work and the recommendations for future research presented in section . . experimental setup and results . . setup and test program figure . dut formula society of automotive engineers (fsae) vehicle at the formula student germany competition [ ]. to achieve the best fsae vehicle performance, the team developed various tools to evaluate vehicle dynamics and to predict its behavior. however, predictions are based on estimated friction coefficient, neglecting the thermal effect on the tire performance. frequently, lap simulation is based on a point mass or simplified vehicle models allowing for the development of the top-level concept of the new vehicle. it can quantify vehicle parameters such as the height of the center of gravity and lift coefficient. however, to design various vehicle components and, specifically, control algorithms, a full vehicle simulation including a comprehensive tire model is required. for an accurate modeling of tire forces and moments, a physical, semi- or empirical tire model can be used [ ]. the physical approach provides more insights regarding tire behavior and better represents the whole operation range; however, it lacks accuracy, specifically considering camber influence, conicity, plysteer, and other phenomena [ , ]. the empirical approach provides a higher modeling accuracy in the predefined tested range but requires extensive special tests. according to [ , ] the tire temperature, especially for racing application [ ], has a significant effect on the force capability of the tire. several advanced tire models using the brush element approach incorporated with thermal effects have been proposed [ – ]; however, their parametrization requires intensive test sessions, commonly unfeasible for fs teams. for a fsae vehicle, less complex models incorporating temperature effect are proposed using the physical approach based on the brush model [ , ]. they demonstrate a good correlation with the experimental tire data collected using the indoor flat track tire test machine [ ]. also, the proposed physical-based models were compared to the original magic formula demonstrating close accuracy; however, original magic formula does not include the effect of temperature on the force capability of the tire. thus, the goal of the proposed study was to develop an empirical tire model incorporating thermal effects and to evaluate its performance compared to the experimental measurements. the main contribution of the study is the improvement of the accuracy of the empirical tire model using the proposed thermal model and modification of the magic formula to incorporate temperature effects. the paper is structured as follows. section describes the experimental setup, test program, and experimental results. a thermal tire model is proposed in section discussing the basic thermal equations and comparing them to well-established thermal models. section presents the widely used magic formula with the extension related to the temperature effect. the complete vehicle model including thermal and modified magic formula is presented in section ; it includes the whole model simulation for steady-state and transient maneuvers compared to real tests and simulation without the temperature effect. the paper concludes with a discussion of the work and the recommendations for future research presented in section . appl. sci. , , of . experimental setup and results . . setup and test program the experimental tests were performed at dynamic test center ag [ ] located in vaufellin, switzerland. the test program is shown in table covering uncombined slip testing using sweep tests of slip angle α and wheel slip κ under various normal loads fz for three tires: apollo single compound, apollo double compound, and hoosier. table . tire test program. measurement type α (rad) κ (–) fz (n) pure cornering sweep / pure acceleration/braking sweep the measurements were performed outdoor using a test truck (figure ). to measure tire forces and moments, a kistler wheel force transducer [ ] was used. the measurement accuracy of the wheel force transducer is % of the full scale. for the investigated load range, it resulted in approx. % absolute measurement error including crosstalk. the normal load during the test was controlled using hydraulic suspension. the target normal load was obtained by the adjustment of hydraulic pressure. the wheel was driven through the axle unit from the motor for acceleration tests and using a reverse gear for braking. another actuator was mounted on the tie-rod in order to generate the tire slip angle. appl. sci. , , x for peer review of the experimental tests were performed at dynamic test center ag [ ] located in vaufellin, switzerland. the test program is shown in table covering uncombined slip testing using sweep tests of slip angle α and wheel slip κ under various normal loads fz for three tires: apollo single compound, apollo double compound, and hoosier. table . tire test program. measurement type α (rad) κ (–) fz (n) pure cornering sweep / pure acceleration/braking sweep the measurements were performed outdoor using a test truck (figure ). to measure tire forces and moments, a kistler wheel force transducer [ ] was used. the measurement accuracy of the wheel force transducer is % of the full scale. for the investigated load range, it resulted in approx. % absolute measurement error including crosstalk. the normal load during the test was controlled using hydraulic suspension. the target normal load was obtained by the adjustment of hydraulic pressure. the wheel was driven through the axle unit from the motor for acceleration tests and using a reverse gear for braking. another actuator was mounted on the tie-rod in order to generate the tire slip angle. figure . test setup. . . pure cornering results tire lateral performance was assessed by performing sweep tests at a constant road wheel steering rate of °/s. the tests were conducted under various normal loads fz of and n correspondingly. during the tests, the vertical force was fluctuated due to road modulation and other irregularities, which as a result affected the lateral force fy measurement. to obtain the lateral force, this influence was compensated using the normalized ratio between lateral and normal forces and then multiplying by mean vertical load. figure a shows the effect of vertical load on the utilized friction coefficient µ. the results are aligned to the load sensitivity phenomena that utilized friction coefficient reduces as normal load increases. the test results of hoosier and apollo tires (double compound) are compared in figure b. the peak utilized friction for the hoosier tire was around % higher than the apollo one; however, the average utilized friction was in the same range. figure . test setup. . . pure cornering results tire lateral performance was assessed by performing sweep tests at a constant road wheel steering rate of ◦/s. the tests were conducted under various normal loads fz of and n correspondingly. during the tests, the vertical force was fluctuated due to road modulation and other irregularities, which as a result affected the lateral force fy measurement. to obtain the lateral force, this influence was compensated using the normalized ratio between lateral and normal forces and then multiplying by mean vertical load. figure a shows the effect of vertical load on the utilized friction coefficient µ. the results are aligned to the load sensitivity phenomena that utilized friction coefficient reduces as normal load increases. the test results of hoosier and apollo tires (double compound) are compared in figure b. the peak utilized friction for the hoosier tire was around % higher than the apollo one; however, the average utilized friction was in the same range. appl. sci. , , of appl. sci. , , x for peer review of (a) (b) figure . normalized lateral friction µ vs. slip angle α: (a) different normal loads fz (apollo); (b) different tire types (fz = n). . . pure acceleration and braking results for the longitudinal force measurement, the wheel was lifted, accelerated, and then pressed onto the ground. the brake torque was applied to realize sweep tests of wheel slip while the forward velocity of the truck was kept constant. the effect of normal load on utilized friction is much less compared to lateral friction (figure a). the performance of the apollo tire was similar compared to the hoosier one (figure b). the difference was found to be % which can be related to the ambient and track conditions. (a) (b) figure . normalized longitudinal friction µ vs. wheel slip κ: (a) different normal loads fz (apollo); (b) different tire types (fz = n). . . results related to temperature effect both the force capability and the tire lifetime depend on tire temperature. since the tire carcass is an elastic element, the temperature change will result in the modulus of elasticity of the rubber and therefore influence the cornering stiffness [ ]. the effect of temperature on the normalized longitudinal and lateral friction is shown in figure . it can be observed that the temperature affects the stiffness (curve slope) and the peak friction. figure . normalized lateral friction µ vs. slip angle α: (a) different normal loads fz (apollo); (b) different tire types (fz = n). . . pure acceleration and braking results for the longitudinal force measurement, the wheel was lifted, accelerated, and then pressed onto the ground. the brake torque was applied to realize sweep tests of wheel slip while the forward velocity of the truck was kept constant. the effect of normal load on utilized friction is much less compared to lateral friction (figure a). the performance of the apollo tire was similar compared to the hoosier one (figure b). the difference was found to be % which can be related to the ambient and track conditions. appl. sci. , , x for peer review of (a) (b) figure . normalized lateral friction µ vs. slip angle α: (a) different normal loads fz (apollo); (b) different tire types (fz = n). . . pure acceleration and braking results for the longitudinal force measurement, the wheel was lifted, accelerated, and then pressed onto the ground. the brake torque was applied to realize sweep tests of wheel slip while the forward velocity of the truck was kept constant. the effect of normal load on utilized friction is much less compared to lateral friction (figure a). the performance of the apollo tire was similar compared to the hoosier one (figure b). the difference was found to be % which can be related to the ambient and track conditions. (a) (b) figure . normalized longitudinal friction µ vs. wheel slip κ: (a) different normal loads fz (apollo); (b) different tire types (fz = n). . . results related to temperature effect both the force capability and the tire lifetime depend on tire temperature. since the tire carcass is an elastic element, the temperature change will result in the modulus of elasticity of the rubber and therefore influence the cornering stiffness [ ]. the effect of temperature on the normalized longitudinal and lateral friction is shown in figure . it can be observed that the temperature affects the stiffness (curve slope) and the peak friction. figure . normalized longitudinal friction µ vs. wheel slip κ: (a) different normal loads fz (apollo); (b) different tire types (fz = n). . . results related to temperature effect both the force capability and the tire lifetime depend on tire temperature. since the tire carcass is an elastic element, the temperature change will result in the modulus of elasticity of the rubber and therefore influence the cornering stiffness [ ]. the effect of temperature on the normalized longitudinal and lateral friction is shown in figure . it can be observed that the temperature affects the stiffness (curve slope) and the peak friction. appl. sci. , , of appl. sci. , , x for peer review of (a) (b) figure . temperature effects on tire forces for apollo tire (fz = n): (a) normalized longitudinal friction µ vs. wheel slip κ; (b) normalized lateral friction µ vs. slip angle α. . thermal model . . proposed thermal model a lumped parameter approach is used according to [ ] and the thermal model consists of three bodies such as the tread with temperature ttread, the carcass with temperature tcarcass, and the inflation gas with temperature tgas. the road surface with ambient air provides the boundary conditions with fixed temperatures troad and tamb. in contrast to the study [ ], a single value of tread temperature was calculated without the need to define a complex temperature distribution across the contact patch area. the temperature change of tread, carcass, and gas is described as: sliding tread road carcass tread tread amb tread tread tread damp carcass tread carcass amb carcass gas carcass carcass carcass carcass gas gas gas gas q q q q t s m q q q q t s m q t s m → → → → → → → − + − = − − − = =    ( ) where qsliding is the heat flow due to sliding; tread roadq → is the heat flow between tread and road; carcass treadq → is the heat flow between carcass and tread; tread ambq → is the heat flow between tread and ambient air; qdamp is the heat flow due to carcass deflection; carcass ambq → is the heat flow between carcass and ambient air; carcass gasq → is the heat flow between carcass and inflation gas; stread is the specific heat capacity of carcass; scarcass is the specific heat capacity of carcass; sgas is the specific heat capacity of inflation gas; mtread is the tread mass; mcarcass is the carcass mass; mgas is the inflation gas mass. according to [ ], two heat generation processes should be considered: • due to carcass deflection ( )damp x x y y z z xq e f e f e f v= + + ( ) where ex, ey, and ez are the carcass longitudinal, lateral, and vertical force efficiency factors; fx, fy, and fz are the longitudinal, lateral, and normal forces; vx is the longitudinal velocity. • due to sliding friction in the contact patch sliding d z sq f vμ= ( ) figure . temperature effects on tire forces for apollo tire (fz = n): (a) normalized longitudinal friction µ vs. wheel slip κ; (b) normalized lateral friction µ vs. slip angle α. . thermal model . . proposed thermal model a lumped parameter approach is used according to [ ] and the thermal model consists of three bodies such as the tread with temperature ttread, the carcass with temperature tcarcass, and the inflation gas with temperature tgas. the road surface with ambient air provides the boundary conditions with fixed temperatures troad and tamb. in contrast to the study [ ], a single value of tread temperature was calculated without the need to define a complex temperature distribution across the contact patch area. the temperature change of tread, carcass, and gas is described as: . ttread = qsliding−qtread→road+qcarcass→tread−qtread→amb streadmtread. tcarcass = qdamp−qcarcass→tread−qcarcass→amb−qcarcass→gas scarcassmcarcass. tgas = qcarcass→gas sgasmgas ( ) where qsliding is the heat flow due to sliding; qtread→road is the heat flow between tread and road; qcarcass→tread is the heat flow between carcass and tread; qtread→amb is the heat flow between tread and ambient air; qdamp is the heat flow due to carcass deflection; qcarcass→amb is the heat flow between carcass and ambient air; qcarcass→gas is the heat flow between carcass and inflation gas; stread is the specific heat capacity of carcass; scarcass is the specific heat capacity of carcass; sgas is the specific heat capacity of inflation gas; mtread is the tread mass; mcarcass is the carcass mass; mgas is the inflation gas mass. according to [ ], two heat generation processes should be considered: • due to carcass deflection qdamp = ( ex|fx|+ ey ∣∣∣fy∣∣∣ + ez|fz|)vx ( ) where ex, ey, and ez are the carcass longitudinal, lateral, and vertical force efficiency factors; fx, fy, and fz are the longitudinal, lateral, and normal forces; vx is the longitudinal velocity. • due to sliding friction in the contact patch qsliding = µdfzvs ( ) where µd is the dynamic friction coefficient; vs is the sliding velocity. to define the shift along the µ axis with compound temperature, the dynamic friction model [ ] was modified. based on the friction model [ ], equation ( ) is proposed incorporating the shift along appl. sci. , , of the µ axis due to temperature using the parameters µ and h as temperature dependent. the idea was taken from research [ ] where the parameters of the friction models from [ ] and [ ] were made temperature-dependent. the final equation for the dynamic friction model is described: µd(t) = µbase + [ µpeak(t)−µbase ] e−(h(t)(log ( vs vmax )−kshi f t(ttread−tre f ))) ( ) the temperature-dependent peak friction coefficient µpeak is defined: µpeak(t) = a t + a t + a ( ) where a , a , and a are the tuning parameters for a second order polynomial function. dimensionless parameter h is related to the width of the speed range in which the friction coefficient varies significantly [ ]. the following temperature-dependent adjustment is made similar to the adaptation of the compound shear modulus in [ ]: h(t) = b eb ttread eb tre f ( ) where b and b are the tuning parameters. the proposed modifications cover both sliding and non-sliding friction components. temperature-dependent peak friction coefficient µpeak corresponds to the peak friction properties of the tire compound. the parameter h affects the curve slope and results in the change of the shear modulus of the tire compound. the peak friction coefficient µpeak, the lower limit of friction coefficient µbase, and the parameter h are identified from measurement data. the simulation and experimental results for the obtained normalized lateral friction coefficient are shown in figure . the model shows similar qualitative behavior observed in the experimental data at the slip angle above ◦. appl. sci. , , x for peer review of where µd is the dynamic friction coefficient; vs is the sliding velocity. to define the shift along the µ axis with compound temperature, the dynamic friction model [ ] was modified. based on the friction model [ ], equation ( ) is proposed incorporating the shift along the µ axis due to temperature using the parameters µ and h as temperature dependent. the idea was taken from research [ ] where the parameters of the friction models from [ ] and [ ] were made temperature-dependent. the final equation for the dynamic friction model is described: ( ) ( ) ( ) ( ) max log s shift tread ref v h t k t t v d base peak baset t eμ μ μ μ      − − −         = + −  ( ) the temperature-dependent peak friction coefficient µpeak is defined: ( ) peak t at a t aμ = + + ( ) where a , a , and a are the tuning parameters for a second order polynomial function. dimensionless parameter h is related to the width of the speed range in which the friction coefficient varies significantly [ ]. the following temperature-dependent adjustment is made similar to the adaptation of the compound shear modulus in [ ]: ( ) tread ref b t b t e h t b e = ( ) where b and b are the tuning parameters. the proposed modifications cover both sliding and non-sliding friction components. temperature-dependent peak friction coefficient µpeak corresponds to the peak friction properties of the tire compound. the parameter h affects the curve slope and results in the change of the shear modulus of the tire compound. the peak friction coefficient µpeak, the lower limit of friction coefficient µbase, and the parameter h are identified from measurement data. the simulation and experimental results for the obtained normalized lateral friction coefficient are shown in figure . the model shows similar qualitative behavior observed in the experimental data at the slip angle above °. figure . comparison between measurement data and modified friction model. four types of heat flows are considered: i) heat flow between tread and ambient air; ii) heat flow between carcass and ambient air; iii) heat flow between tread and carcass; iv) heat flow between carcass and inflation gas. heat flow with the ambient air tread ambq → and carcass ambq → is defined as: figure . comparison between measurement data and modified friction model. four types of heat flows are considered: (i) heat flow between tread and ambient air; (ii) heat flow between carcass and ambient air; (iii) heat flow between tread and carcass; (iv) heat flow between carcass and inflation gas. appl. sci. , , of heat flow with the ambient air qtread→amb and qcarcass→amb is defined as: qtread→amb = htread−amb(ttread − tamb) qcarcass→amb = hcarcass−amb ( tcarcass − tgas ) ( ) the heat flow coefficient hcarcass−amb is constant and the heat flow coefficient htread-amb is assumed to be a function of the longitudinal vehicle speed vx and taken as a linear function [ ]: htread−amb = vx + ( ) heat flow between the tread and carcass qcarcass→tread is calculated as: qcarcass→tread = hcarcass−tread(tcarcass − ttread) ( ) heat flow between the carcass and inflation gas qcarcass→gas is defined as: qcarcass→gas = hcarcass−gas ( tcarcass − tgas ) ( ) using heat flow coefficient htread−road, heat flow between tread and road qtread→road can be found as: qtread→road = htread−roadacp(ttread − troad) ( ) the contact patch area, acp, is calculated assuming a constant width of the contact patch b and the variable contact patch length a which was adjusted compared to [ ] to better match the data of the fsae tires: acp = ab = . p − . bar ( fz ) . b ( ) since the volume of the gas is constant, the pressure is directly proportional to the kelvin temperature and thus the inflation pressure pbar can be calculated as: pbar = pcold tgas + tamb + ( ) where pcold is the pressure of the tire at ambient temperature. . . comparison with the established models the sorniotti model [ ] describes an empirical model to estimate tire temperature as the function of the actual working conditions of the component. to evaluate the temperature effect on tire forces, a combination of the estimated temperature with a tire brush model [ ] was used. the model was empirically tuned using experimental data to demonstrate the variation of tire performance as temperature function. the thermal model considers distinct thermal capacities related to the tread and carcass. the tread thermal capacity is related to the heat flux caused by the tire–road forces and carcass thermal capacity is affected by rolling resistance and exchanges heat with the external ambient. other heat fluxes corresponded to the ambient and exchange between the two capacities. the original model [ ] did not consider the heat flow between the road and tread. therefore, the heat flow term ptread,road was introduced to the original model to improve the accuracy according to [ ]. the model is described as: ceq_carcass dtcarcass dt = prolling_resis tan ce + pconduction + pambient,carcass ceq_tread dttread dt = pfx,tire + pfy,tire − pconduction + pambient,tread + ptread,road ( ) appl. sci. , , of power fluxes corresponding to the cooling flux due to the temperature difference between carcass and ambient pambient,carcass, and tire tread and ambient pambient,tread are defined as: pambient,carcass = hcarcass(tambient − tcarcass) pambient,tread = htread(tambient − ttread) ( ) power fluxes related to conduction between tread and carcass pconduction and between tread and road ptread,road are calculated as: pconduction = hconduction(ttread − tcarcass) ptread,road = htread−road(ttread − troad) ( ) compared to the sorniotti model, the proposed thermal model due to a higher differential order should capture more dynamics related to heat flow and potentially produce better results. the kelly and sharp model [ ] states that in racing applications, the temperature of the tread significantly affects both the tire stiffness and the contact patch friction. it should be noted that the effect on the friction is higher. since the rubber viscoelastic properties depend on temperature, the maximum performance on the racetrack is only available in a specific temperature range. the tire model is also based on the brush model. using the bristle stiffness cp the adhesion part can be presented: cp = wcpgtread htread ( ) where wcp is the contact patch width; htread is the tread height. the shear modulus of the tread gtread is defined as: gtread = gta − glimit e−kgtga e−kgttread + glimit, where kg = log(gta − glimit)− log(gtb − glimit) tgb − tga ( ) where gta is the reference shear modulus at temperature a; gtb is the reference shear modulus at temperature b; glimit is the limit shear modulus value at a high temperature; tga is the reference temperature a; tgb is the reference temperature b. the sliding part of the contact patch is described using the dynamic friction coefficient. the master curve of the friction coefficient µmc for the tread compound is assumed to have a gaussian shape and it describes a friction curve with a gaussian shape on a log frequency axis: µmc = µbase + [ µpeak −µbase ] e−(kshape(log vs−kshi f t(ttread−tre f ))) ( ) where µbase and µpeak are the lower and upper limits of the dynamic friction coefficient; kshape is the master curve shape factor; vs is the sliding velocity; kshift is the master curve temperature shift factor; ttread is the tread temperature; tref is the master curve reference temperature. in addition, the model takes into account the influence of contact patch pressure on friction coefficient. it is assumed that friction coefficients are reduced linearly with contact patch pressure. the static and dynamic friction coefficients are defined as: µ = kcppµre f µd = kcppµmc , where kcpp = − kpcpp pcp kre f cpp ( ) where pcp is the contact patch pressure; kcpp is the parameter of friction reduction rate; kpcpp is the friction roll-off factor with contact patch pressure; krefcpp is the reference contact patch pressure; µ is the static friction coefficient; µd is the dynamic friction coefficient; µref is the reference static friction coefficient. appl. sci. , , of compared to the kelly and sharp model, the peak friction coefficient µpeak is temperature-dependent in the proposed model, which could result in better accuracy. . . comparison results the simulation results obtained from the considered thermal models were compared with measurement data for various maneuvers. the initial values of coefficients were taken from [ , ] and further tuned using constrained parameter optimization. to tune parameters, a non-linear least square method was applied to obtain a suitable approximation of temperature effect and was based on the error from the measurement defined as: emeas = √√∑ (tmodel − tmeasurement) ∑ t model ( ) it was checked that the fitting parameters provide a good estimation for different maneuvers. for example, the parameters were first optimized for lateral tests and later checked again for acceleration/braking tests. then an average was taken for the tuning parameters to cover all the experimental tests and represent the whole driving envelope. the comparisons between the measured temperature, the temperature obtained from the kelly and sharp model, and the proposed thermal model are shown in figure . appl. sci. , , x for peer review of the simulation results obtained from the considered thermal models were compared with measurement data for various maneuvers. the initial values of coefficients were taken from [ , ] and further tuned using constrained parameter optimization. to tune parameters, a non-linear least square method was applied to obtain a suitable approximation of temperature effect and was based on the error from the measurement defined as: ( ) model measurement model meas t t e t − =   ( ) it was checked that the fitting parameters provide a good estimation for different maneuvers. for example, the parameters were first optimized for lateral tests and later checked again for acceleration/braking tests. then an average was taken for the tuning parameters to cover all the experimental tests and represent the whole driving envelope. the comparisons between the measured temperature, the temperature obtained from the kelly and sharp model, and the proposed thermal model are shown in figure . the accumulated error for the compared thermal models is summarized in table . table . accumulated errors for compared thermal models. sorniotti model kelly and sharp model proposed model pure acceleration . % . % . % pure braking . % % . % pure cornering . % . % . % the results of the proposed thermal model and the compared models are close to the measurement data for all conducted tests. the proposed thermal model provides the smallest error for pure cornering. for pure braking the sorniotti model shows the best performance and for pure acceleration the smallest error can be achieved using the kelly and sharp model. however, the conclusions regarding adaptability of the models to different normal loads cannot be extracted from table . since the kelly and sharp model considers the contact patch area depending on the normal load and air pressure, the model can adapt to different normal loads and the maximum error is limited to ~ %. therefore, the models were also checked using measurement data for different load conditions. it was found that although the sorniotti model has a good fit to the measured data, the error increases when the load condition is changed (figure ). (a) (b) figure . temperature change in pure braking test: (a) results of the kelly and sharp model; (b) results of the proposed model. figure . temperature change in pure braking test: (a) results of the kelly and sharp model; (b) results of the proposed model. the accumulated error for the compared thermal models is summarized in table . table . accumulated errors for compared thermal models. sorniotti model kelly and sharp model proposed model pure acceleration . % . % . % pure braking . % % . % pure cornering . % . % . % the results of the proposed thermal model and the compared models are close to the measurement data for all conducted tests. the proposed thermal model provides the smallest error for pure cornering. for pure braking the sorniotti model shows the best performance and for pure acceleration the smallest error can be achieved using the kelly and sharp model. however, the conclusions regarding adaptability of the models to different normal loads cannot be extracted from table . since the kelly and sharp model considers the contact patch area depending on the normal load and air pressure, the appl. sci. , , of model can adapt to different normal loads and the maximum error is limited to ~ %. therefore, the models were also checked using measurement data for different load conditions. it was found that although the sorniotti model has a good fit to the measured data, the error increases when the load condition is changed (figure ).appl. sci. , , x for peer review of (a) (b) figure . temperature change in pure braking test using sorniotti model: (a) normal load fz of n; (b) normal load fz of n. . modified magic formula incorporating temperature effect . . baseline parameter estimation temperature plays a crucial role in tire performance. a large number of factors, such as the normal load, ambient conditions, etc., affect the temperature of the tread. since the tread temperature profile was different on the two tires due to normal force fluctuation (two tires were tested simultaneously), it was necessary to align the measurements in such a way that the tires were in a similar temperature range. once the measurements were processed according to similar temperature ranges, the measurement data was converted according to tydex format [ ]. this format enables the usage of mftool software provided by tass international (former tno automotive, helmond, netherlands) [ ]. the mftool software is a commercial software to determine the coefficients of the empirical model, well known as the magic formula, from the pre-processed measurement data. it also allows to compensate for the variation in normal load and fits the curve for a nominal load. in the content of this study, mftool was used to generate a tire property file for the extended magic formula (appendix a). . . basic magic formula the magic formula is an empirical model commonly used to simulate steady-state tire forces and moments. it is based on a special function instead of look-up tables or other various polynomial functions. by adjusting the function coefficients, the same special function can be used to describe longitudinal and lateral forces (sine function), and self-aligning moment (cosine function). the basic magic formula has the following form [ ]: ( ) ( )( ) ( ) ( ) sin arctan arctan v h y x d c bx e bx bx y x y x s x x s = − − = + = + ( ) where y(x) may represent longitudinal, lateral forces or self-aligning moment; x denotes slip angle α or longitudinal wheel slip κ; b is stiffness factor; c is shape factor; d is peak factor; e is curvature factor; sh is horizontal shift; sv is vertical shift. the peak value is described by the peak factor d. the derivative at the origin provides bcd as a result and defines the longitudinal or cornering stiffness of the tire. the curve shape can be adjusted by both factors c and e, where c, called as shape factor, controls the “stretching” in the x direction and e, called the curvature factor, enables a local stretch or compression. the parameters sh and sv figure . temperature change in pure braking test using sorniotti model: (a) normal load fz of n; (b) normal load fz of n. . modified magic formula incorporating temperature effect . . baseline parameter estimation temperature plays a crucial role in tire performance. a large number of factors, such as the normal load, ambient conditions, etc., affect the temperature of the tread. since the tread temperature profile was different on the two tires due to normal force fluctuation (two tires were tested simultaneously), it was necessary to align the measurements in such a way that the tires were in a similar temperature range. once the measurements were processed according to similar temperature ranges, the measurement data was converted according to tydex format [ ]. this format enables the usage of mftool software provided by tass international (former tno automotive, helmond, the netherlands) [ ]. the mftool software is a commercial software to determine the coefficients of the empirical model, well known as the magic formula, from the pre-processed measurement data. it also allows to compensate for the variation in normal load and fits the curve for a nominal load. in the content of this study, mftool was used to generate a tire property file for the extended magic formula (appendix a). . . basic magic formula the magic formula is an empirical model commonly used to simulate steady-state tire forces and moments. it is based on a special function instead of look-up tables or other various polynomial functions. by adjusting the function coefficients, the same special function can be used to describe longitudinal and lateral forces (sine function), and self-aligning moment (cosine function). the basic magic formula has the following form [ ]: y(x) = d sin(carctanbx − e(bx − arctanbx)) y(x) = y(x) + sv x = x + sh ( ) where y(x) may represent longitudinal, lateral forces or self-aligning moment; x denotes slip angle α or longitudinal wheel slip κ; b is stiffness factor; c is shape factor; d is peak factor; e is curvature factor; sh is horizontal shift; sv is vertical shift. appl. sci. , , of the peak value is described by the peak factor d. the derivative at the origin provides bcd as a result and defines the longitudinal or cornering stiffness of the tire. the curve shape can be adjusted by both factors c and e, where c, called as shape factor, controls the “stretching” in the x direction and e, called the curvature factor, enables a local stretch or compression. the parameters sh and sv allow to model the influence of plysteer and conicity. the extended version of the magic formula [ ] also covers effects of load dependency, combined slip, camber angle, inflation pressure, etc. . . proposed modification for temperature effect to change the peak friction and the stiffness of the force curve according to temperature variation, the magic formula was modified to affect the tire forces and moments. the proposed modifications are only demonstrated for the basic magic formula; meanwhile, they were integrated with the extended version of the magic formula. to represent temperature change, a delta temperature ∆t was defined: ∆t = t − tre f tre f ( ) where tref is reference temperature (the performance is known from measurements). to evaluate the performance of the proposed modification, the normalized friction values from measurement data and simulation results were compared. the measured temperature was used as an input to the modified magic formula. longitudinal force. four temperature coefficients were introduced to capture the temperature effect on the longitudinal force. the coefficients related to temperature effect are: tx linear temperature effect on longitudinal slip stiffness tx quadratic temperature effect on longitudinal slip stiffness tx linear temperature effect on longitudinal friction tx quadratic temperature effect on longitudinal friction to affect the longitudinal slip stiffness and longitudinal friction, the above-mentioned coefficients were used. the modified peak factor dx and longitudinal slip stiffness kx are: dx = ( + tx ∆t + tx ∆t ) dxbase kx = ( + tx ∆t + tx ∆t ) kxbase ( ) where dxbase is the peak factor of pure longitudinal force d and kxbase is the longitudinal slip stiffness kx according to the basic magic formula. as can be observed in figure , the measurement data and simulation results of normalized longitudinal friction are close to each other. the quantitative comparison between simulation and experimental results for longitudinal force is summarized in table . it also can be noted that temperature changes longitudinal stiffness and the longitudinal friction similar to experimental data. the main difference is related to the measurement noise and fluctuations due to road surface irregularities. table . errors from measurements for different temperatures. temperature range longitudinal force fx lateral force fy t < ◦c - . % ◦c ≤ t < ◦c . % . % ◦c ≤ t < ◦c . % . % ◦c ≤ t < ◦c . % - ◦c ≤ t < ◦c . % - t > ◦c % - appl. sci. , , of appl. sci. , , x for peer review of (a) (b) figure . normalized longitudinal force fx: (a) experimental measurements; (b) modified magic formula. lateral force and self-aligning moment. four temperature coefficients were introduced to incorporate the temperature effect in lateral force variation: ty temperature effect on cornering stiffness magnitude ty temperature effect on location of cornering stiffness peak ty linear temperature effect on lateral friction ty quadratic temperature effect on lateral friction to affect the cornering stiffness and lateral friction respectively, the above-mentioned coefficients were used. the modified peak factor dy and the cornering stiffness ky are described: ( ) ( ) ( ) sin arctan y y y ybase z y y ky z ky z y d t t t t d f k t t p f p f t t = + Δ + Δ    = + Δ     + Δ   ( ) where dybase is the peak factor of pure lateral force d in the basic magic formula; fz is the nominal normal load; pky is the maximum value of stiffness kfy/fz ; pky is the load at which kfy reaches maximum value. the results of normalized lateral force from measurement data and simulation results are shown in figure and demonstrate a similar behavior while temperature changes. the quantitative assessment between simulation and experimental results for lateral force is summarized in table . the pneumatic trail would remain constant under the assumption of the same pressure distribution in a contact patch. since the self-aligning moment is the product between the lateral force and the tire pneumatic trail, it will change only due to the lateral force variation. therefore, there is no need to introduce the coefficients affecting the self-aligning moment as shown in figure . figure . normalized longitudinal force fx: (a) experimental measurements; (b) modified magic formula. lateral force and self-aligning moment. four temperature coefficients were introduced to incorporate the temperature effect in lateral force variation: ty temperature effect on cornering stiffness magnitude ty temperature effect on location of cornering stiffness peak ty linear temperature effect on lateral friction ty quadratic temperature effect on lateral friction to affect the cornering stiffness and lateral friction respectively, the above-mentioned coefficients were used. the modified peak factor dy and the cornering stiffness ky are described: dy = ( + ty ∆t + ty ∆t ) dybase ky = ( + ty ∆t)pky fz sin ( arctan ( fz pky fz ( +ty ∆t) )) ( ) where dybase is the peak factor of pure lateral force d in the basic magic formula; fz is the nominal normal load; pky is the maximum value of stiffness kfy/fz ; pky is the load at which kfy reaches maximum value. the results of normalized lateral force from measurement data and simulation results are shown in figure and demonstrate a similar behavior while temperature changes. the quantitative assessment between simulation and experimental results for lateral force is summarized in table . the pneumatic trail would remain constant under the assumption of the same pressure distribution in a contact patch. since the self-aligning moment is the product between the lateral force and the tire pneumatic trail, it will change only due to the lateral force variation. therefore, there is no need to introduce the coefficients affecting the self-aligning moment as shown in figure . the error to compare the performance of the modified magic formula is defined as: emeas = % √ (fsimulation − fmeasurement) |fmeasurement| ( ) data for different temperature ranges were taken to evaluate the error values for the longitudinal and lateral forces, and the summary is presented in table . as it can be observed, the modified magic formula performs sufficiently close to the actual measurement data taking into account that the absolute measurement error was %. the largest errors occur at lower temperatures and a general trend of improved accuracy is seen as the temperature increases. for lower temperatures, the potential reason of the deviation is related to a high variance of appl. sci. , , of measured data. the errors for the self-aligning moment are not shown due to the high variance in the measurement data related to the measurement setup. appl. sci. , , x for peer review of (a) (b) figure . normalized lateral force fy: (a) experimental measurements; (b) modified magic formula. (a) (b) figure . normalized self-aligning moment mz: (a) experimental measurements; (b) modified magic formula. the error to compare the performance of the modified magic formula is defined as: ( ) % simulation measurementmeas measurement f f e f − = ( ) data for different temperature ranges were taken to evaluate the error values for the longitudinal and lateral forces, and the summary is presented in table . table . errors from measurements for different temperatures. temperature range longitudinal force fx lateral force fy t < °c - . % °c ≤ t < °c . % . % °c ≤ t < °c . % . % °c ≤ t < °c . % - °c ≤ t < °c . % - t > °c % - as it can be observed, the modified magic formula performs sufficiently close to the actual measurement data taking into account that the absolute measurement error was %. the largest errors occur at lower temperatures and a general trend of improved accuracy is seen as the figure . normalized lateral force fy: (a) experimental measurements; (b) modified magic formula. appl. sci. , , x for peer review of (a) (b) figure . normalized lateral force fy: (a) experimental measurements; (b) modified magic formula. (a) (b) figure . normalized self-aligning moment mz: (a) experimental measurements; (b) modified magic formula. the error to compare the performance of the modified magic formula is defined as: ( ) % simulation measurementmeas measurement f f e f − = ( ) data for different temperature ranges were taken to evaluate the error values for the longitudinal and lateral forces, and the summary is presented in table . table . errors from measurements for different temperatures. temperature range longitudinal force fx lateral force fy t < °c - . % °c ≤ t < °c . % . % °c ≤ t < °c . % . % °c ≤ t < °c . % - °c ≤ t < °c . % - t > °c % - as it can be observed, the modified magic formula performs sufficiently close to the actual measurement data taking into account that the absolute measurement error was %. the largest errors occur at lower temperatures and a general trend of improved accuracy is seen as the figure . normalized self-aligning moment mz: (a) experimental measurements; (b) modified magic formula. . integration into full fsae vehicle model . . full vehicle modeling the following assumptions have been made for the multi-body model of the fsae vehicle: • all joints and bodies besides the tires and springs are considered as rigid. since the suspension and chassis elements are designed to be stiff, compliance is neglected. joint friction is omitted, except modeling of the transmission friction. for this purpose, damping in the revolute joints between the suspension and the wheels is added. • aerodynamic forces are applied on a constant center of pressure. its position was calculated based on computational fluid dynamic simulations in steady-state conditions and changes according to the velocity, roll, pitch, etc. • the characteristics of suspension springs and dampers are linear. the model was developed in simscape multibody (simulink toolbox). the coordinate system and axis orientation was according to the sae coordinate system. center of gravity position and mass-inertial characteristics of the chassis were obtained from the computer-aided design (cad) model or/and actual measurements on the fsae vehicle. the model includes four suspension subsystems appl. sci. , , of connected to the chassis body. using revolute joints from the upright center, the tire was connected to the suspension. the suspension geometry was imported from the cad model corresponding to the actual fsae vehicle. a steering system was connected to the front suspension and the chassis body through different hardpoints. more details regarding the vehicle modeling can be found in [ ]. . . model structure the completed structure of the simulation setup is shown in figure . the tire states (position, velocity, acceleration) are calculated on each time step based on the multi-body vehicle model and transferred to the modified magic formula. according to kinematic states, the magic formula calculates the forces and moments. then tire forces and moments are sent to the thermal model calculating the temperature and its change. finally, tire forces and moments are transferred to the multibody vehicle model, while temperature and pressure are transferred to the tire model. appl. sci. , , x for peer review of temperature increases. for lower temperatures, the potential reason of the deviation is related to a high variance of measured data. the errors for the self-aligning moment are not shown due to the high variance in the measurement data related to the measurement setup. . integration into full fsae vehicle model . . full vehicle modeling the following assumptions have been made for the multi-body model of the fsae vehicle: • all joints and bodies besides the tires and springs are considered as rigid. since the suspension and chassis elements are designed to be stiff, compliance is neglected. joint friction is omitted, except modeling of the transmission friction. for this purpose, damping in the revolute joints between the suspension and the wheels is added. • aerodynamic forces are applied on a constant center of pressure. its position was calculated based on computational fluid dynamic simulations in steady-state conditions and changes according to the velocity, roll, pitch, etc. • the characteristics of suspension springs and dampers are linear. the model was developed in simscape multibody (simulink toolbox). the coordinate system and axis orientation was according to the sae coordinate system. center of gravity position and mass-inertial characteristics of the chassis were obtained from the computer-aided design (cad) model or/and actual measurements on the fsae vehicle. the model includes four suspension subsystems connected to the chassis body. using revolute joints from the upright center, the tire was connected to the suspension. the suspension geometry was imported from the cad model corresponding to the actual fsae vehicle. a steering system was connected to the front suspension and the chassis body through different hardpoints. more details regarding the vehicle modeling can be found in [ ]. . . model structure the completed structure of the simulation setup is shown in figure . the tire states (position, velocity, acceleration) are calculated on each time step based on the multi-body vehicle model and transferred to the modified magic formula. according to kinematic states, the magic formula calculates the forces and moments. then tire forces and moments are sent to the thermal model calculating the temperature and its change. finally, tire forces and moments are transferred to the multibody vehicle model, while temperature and pressure are transferred to the tire model. figure . model structure. . . validation to validate the overall model, two test scenarios were considered. the first test scenario was a skidpad maneuver to evaluate a steady-state lateral behavior. the second test scenario was a full lap maneuver performed for a longer time. skidpad maneuver. the skidpad course consists of two pairs of concentric circles in a figure of eight pattern. the centers of these circles are . m apart. the inner circles are . m in diameter and the outer circles are . m in diameter. the driving path is the m wide path between the inner and outer circles [ ]. to increase the signal-to-noise ratio without significant signal distortion, the raw sensor data was filtered using a savitzky–golay digital filter [ ]. the filtered data was used for comparison with simulation results. the longitudinal and lateral accelerations, and yaw rate are shown in figure . the simulation shows more oversteered behavior compared to the measured data. the mismatch can be caused by the modeling of chassis roll stiffness. a real chassis has a limited roll stiffness; however, as mentioned, the chassis was assumed as a rigid body. hence, the chassis was infinitely stiff in the simulation model. it could result in a higher yaw rate and increase the root mean square error (rmse) of the yaw rate. the longitudinal and lateral accelerations, and yaw rate were better aligned with the measurement data in the left turn circles. this observation can be explained by the fact that the elevation change in the skidpad circuit affected the measurements from the accelerometer sensor. appl. sci. , , of figure shows the evolution of tire temperature during skidpad maneuver for four wheels: front left (fl), front right (fr), rear left (rl), and rear right (rr). as can be observed, due to higher normal loads, the outer tires (initially left tires for the right turn) start heating up first. when the vehicle steered in another direction (at s), the right tires start heating up. appl. sci. , , x for peer review of figure . model structure. . . validation to validate the overall model, two test scenarios were considered. the first test scenario was a skidpad maneuver to evaluate a steady-state lateral behavior. the second test scenario was a full lap maneuver performed for a longer time. skidpad maneuver. the skidpad course consists of two pairs of concentric circles in a figure of eight pattern. the centers of these circles are . m apart. the inner circles are . m in diameter and the outer circles are . m in diameter. the driving path is the m wide path between the inner and outer circles [ ]. to increase the signal-to-noise ratio without significant signal distortion, the raw sensor data was filtered using a savitzky–golay digital filter [ ]. the filtered data was used for comparison with simulation results. the longitudinal and lateral accelerations, and yaw rate are shown in figure . figure . longitudinal acceleration ax, lateral acceleration ay, and yaw rate r for skidpad. the simulation shows more oversteered behavior compared to the measured data. the mismatch can be caused by the modeling of chassis roll stiffness. a real chassis has a limited roll stiffness; however, as mentioned, the chassis was assumed as a rigid body. hence, the chassis was infinitely stiff in the simulation model. it could result in a higher yaw rate and increase the root mean square error (rmse) of the yaw rate. the longitudinal and lateral accelerations, and yaw rate were better aligned with the measurement data in the left turn circles. this observation can be explained by the fact that the elevation change in the skidpad circuit affected the measurements from the accelerometer sensor. figure shows the evolution of tire temperature during skidpad maneuver for four wheels: front left (fl), front right (fr), rear left (rl), and rear right (rr). as can be observed, due to higher normal loads, the outer tires (initially left tires for the right turn) start heating up first. when the vehicle steered in another direction (at s), the right tires start heating up. figure . longitudinal acceleration ax, lateral acceleration ay, and yaw rate r for skidpad. full lap. to validate transient behavior of the simulation model, a full lap maneuver was investigated. the measurement data correspond to one of the endurance runs with a duration of ~ km driving ( – laps). the simulation was conducted for one lap of the circuit performed approx. in s. in a similar way as was done for the skidpad scenario, the raw measurement data of longitudinal and lateral accelerations, and yaw rate, were filtered and the rmse calculated. the rmse for the longitudinal and lateral accelerations, as well as yaw rate are summarized in table , including also the simulation results performed without the proposed model. the comparison between simulation results with a temperature effect and measurement data is shown in figure . it can be noted that the yaw rate from the simulation results was aligned with the measured data; however, an offset can be observed which might relate to the above-mentioned difference in the chassis roll stiffness. table . the root mean square error (rmse) for both simulation maneuvers. skidpad maneuver full lap maneuver temperature effect without with without with longitudinal acceleration, m/s . . . . lateral acceleration, m/s . . . . yaw rate, rad/s . . . . the rmse for longitudinal acceleration is slightly higher compared to the skidpad scenario; however, the longitudinal acceleration in the skidpad maneuver was much lower compared to a full lap scenario. thus, the simulation results of longitudinal dynamics using wheel torques as an input matched the longitudinal acceleration as sufficiently accurate. the highest difference was found in lateral dynamics by the rmse of lateral acceleration of . m/s . the potential reasons can be related to: (i) combined slip conditions; (ii) camber effect on lateral appl. sci. , , of force, (iii) longitudinal velocity not tacked precisely. due to a limited financial budget, the tire testing was performed only for pure longitudinal and lateral conditions. the coefficients corresponding to combined slip behavior were estimated based on similarly sized fs tires. it can be observed, that the maximum mismatch happens in a corner where the tires were cambered due to chassis roll and performed in combined slip conditions. another aspect is the body slip of the actual vehicle which might affect the differences between the simulation and real fsae vehicles. figure shows the evolution of tire temperature during a full lap maneuver. the tread temperature steadily increased and its reduction corresponds to straight lines or long corners. it can also be observed that a higher temperature for right tires was achieved compared to the left tires. due to the circuit orientation, the right tires have higher loads. in this case, the gas temperature increased and due to the deflections, the carcass heated up. as can be concluded from table , the rmse of lateral acceleration was up to % and % in steady-state and transient maneuvers, correspondingly. regarding the yaw rate, the rmse was up to % in steady-state maneuver and up to % in the transient one. therefore, the modeling of temperature effect has a significant effect on fsae vehicle dynamics. appl. sci. , , x for peer review of (a) (b) (c) figure . evolution of tire temperatures for skidpad maneuver: (a) tire tread temperature; (b) tire carcass temperature; (c) tire inflation gas temperature. full lap. to validate transient behavior of the simulation model, a full lap maneuver was investigated. the measurement data correspond to one of the endurance runs with a duration of ~ km driving ( – laps). the simulation was conducted for one lap of the circuit performed approx. in s. in a similar way as was done for the skidpad scenario, the raw measurement data of longitudinal and lateral accelerations, and yaw rate, were filtered and the rmse calculated. the rmse for the longitudinal and lateral accelerations, as well as yaw rate are summarized in table , including also the simulation results performed without the proposed model. the comparison between simulation results with a temperature effect and measurement data is shown in figure . it can be noted that the yaw rate from the simulation results was aligned with the measured data; however, an offset can be observed which might relate to the above-mentioned difference in the chassis roll stiffness. the rmse for longitudinal acceleration is slightly higher compared to the skidpad scenario; however, the longitudinal acceleration in the skidpad maneuver was much lower compared to a full lap scenario. thus, the simulation results of longitudinal dynamics using wheel torques as an input matched the longitudinal acceleration as sufficiently accurate. the highest difference was found in lateral dynamics by the rmse of lateral acceleration of . m/s . the potential reasons can be related to: (i) combined slip conditions; (ii) camber effect on lateral force, (iii) longitudinal velocity not tacked precisely. due to a limited financial budget, the tire testing was performed only for pure longitudinal and lateral conditions. the coefficients corresponding to combined slip behavior were estimated based on similarly sized fs tires. it can be observed, that the figure . evolution of tire temperatures for skidpad maneuver: (a) tire tread temperature; (b) tire carcass temperature; (c) tire inflation gas temperature. appl. sci. , , of appl. sci. , , x for peer review of maximum mismatch happens in a corner where the tires were cambered due to chassis roll and performed in combined slip conditions. another aspect is the body slip of the actual vehicle which might affect the differences between the simulation and real fsae vehicles. table . the root mean square error (rmse) for both simulation maneuvers. skidpad maneuver full lap maneuver temperature effect without with without with longitudinal acceleration, m/s . . . . lateral acceleration, m/s . . . . yaw rate, rad/s . . . . figure shows the evolution of tire temperature during a full lap maneuver. the tread temperature steadily increased and its reduction corresponds to straight lines or long corners. it can also be observed that a higher temperature for right tires was achieved compared to the left tires. due to the circuit orientation, the right tires have higher loads. in this case, the gas temperature increased and due to the deflections, the carcass heated up. as can be concluded from table , the rmse of lateral acceleration was up to % and % in steady-state and transient maneuvers, correspondingly. regarding the yaw rate, the rmse was up to % in steady-state maneuver and up to % in the transient one. therefore, the modeling of temperature effect has a significant effect on fsae vehicle dynamics. figure . longitudinal acceleration ax, lateral acceleration ay, and yaw rate r for a full lap. figure . longitudinal acceleration ax, lateral acceleration ay, and yaw rate r for a full lap. appl. sci. , , x for peer review of (a) (b) (c) figure . evolution of tire temperatures for full lap: (a) tire tread temperature; (b) tire carcass temperature; (c) tire inflation gas temperature. . conclusions this paper proposes an empirical tire model (modified magic formula) coupled with a thermal model to incorporate temperature effects on the force capability of the tire. the thermal model was proposed to model the influence of temperature on peak friction and the shear modulus, allowing for a better representation of the temperature variation under different load conditions. the magic formula was modified introducing new coefficients allowing for the change in longitudinal and cornering stiffness, and peak factors. the simulation results of the modified magic formula with temperature sufficiently matched the variation in longitudinal and lateral force, as well as self- aligning moment, compared to the experimental tire tests. the error between simulation results and experimental data was in the range between % and %. however, limited accuracy with an error of . % was achieved at lower temperatures due to a high variance in the measured data. to evaluate the proposed tire model, the thermal and modified magic formula models were integrated to a multi-body vehicle model. the simulation results match the behavior of the real fsae vehicle in steady-state maneuver. for transient maneuvers, limited accuracy was achieved due to estimated parameters or unmodeled dynamics such as tire behavior in combined slip conditions, neglect of camber effect on lateral force, etc. comparing the proposed model with an empirical-based tire model without temperature effect, the rmse of lateral acceleration was reduced by up to % and % in the steady-state and the transient maneuvers, correspondingly. for yaw motion, the figure . evolution of tire temperatures for full lap: (a) tire tread temperature; (b) tire carcass temperature; (c) tire inflation gas temperature. appl. sci. , , of . conclusions this paper proposes an empirical tire model (modified magic formula) coupled with a thermal model to incorporate temperature effects on the force capability of the tire. the thermal model was proposed to model the influence of temperature on peak friction and the shear modulus, allowing for a better representation of the temperature variation under different load conditions. the magic formula was modified introducing new coefficients allowing for the change in longitudinal and cornering stiffness, and peak factors. the simulation results of the modified magic formula with temperature sufficiently matched the variation in longitudinal and lateral force, as well as self-aligning moment, compared to the experimental tire tests. the error between simulation results and experimental data was in the range between % and %. however, limited accuracy with an error of . % was achieved at lower temperatures due to a high variance in the measured data. to evaluate the proposed tire model, the thermal and modified magic formula models were integrated to a multi-body vehicle model. the simulation results match the behavior of the real fsae vehicle in steady-state maneuver. for transient maneuvers, limited accuracy was achieved due to estimated parameters or unmodeled dynamics such as tire behavior in combined slip conditions, neglect of camber effect on lateral force, etc. comparing the proposed model with an empirical-based tire model without temperature effect, the rmse of lateral acceleration was reduced by up to % and % in the steady-state and the transient maneuvers, correspondingly. for yaw motion, the rmse of the yaw rate was reduced up to % in the steady-state maneuver and up to % in the transient one. further work related to tire testing which would result in a more precise identification of the magic formula parameters and thus improve the tire model. the following improvements should be considered: (i) combined slip conditions, (ii) camber influence on performance, and (iii) relaxation behavior. author contributions: conceptualization, d.h. and b.s.; testing, d.h.; modeling, d.h.; supervision, b.s.; validation, d.h.; writing—original draft, d.h.; writing—review and editing, b.s. funding: this research received no external funding. acknowledgments: the authors would like to thank formula student team delft, the netherlands, for their support with tire and vehicle testing. conflicts of interest: the authors declare no conflicts of interest. appendix a —tire property file [model] fittyp = $magic formula version number [dimension] unloaded_radius = . $free tyre radius width = . $nominal section width of the tyre rim_radius = . $nominal rim radius rim_width = . $rim width aspect_ratio = . $nominal aspect ratio [inertia] mass = . $tyre mass ixx = . $tyre diametral moment of inertia iyy = . $tyre polar moment of inertia belt_mass = . $belt mass belt_ixx = . $belt diametral moment of inertia belt_iyy = . $belt polar moment of inertia gravity =− . $gravity acting on belt in z direction appl. sci. , , of [vertical] fnomin = $nominal wheel load vertical_stiffness = $tyre vertical stiffness vertical_damping = $tyre vertical damping breff = . $low load stiffness e.r.r. dreff = . $peak value of e.r.r. freff = . $high load stiffness e.r.r. q_re = . $ratio of free tyre radius with nominal tyre radius [longitudinal_coefficients] pcx = . $shape factor cfx for longitudinal force pdx = . $longitudinal friction mux at fznom pdx =− . $variation of friction mux with load pex = . $longitudinal curvature efx at fznom pex = . $variation of curvature efx with load pex = . $factor in curvature efx while driving pkx = . $longitudinal slip stiffness kfx/fz at fznom pkx = . $variation of slip stiffness kfx/fz with load pkx = . $exponent in slip stiffness kfx/fz with load phx =− . $horizontal shift shx at fznom phx = . $variation of shift shx with load pvx = . $vertical shift svx/fz at fznom pvx = . $variation of shift svx/fz with load rbx = $slope factor for combined slip fx reduction rbx = $variation of slope fx reduction with kappa rcx = $shape factor for combined slip fx reduction [lateral_coefficients] pcy = . $shape factor cfy for lateral forces pdy = . $lateral friction muy pdy =− . $variation of friction muy with load pey = . $lateral curvature efy at fznom pey =− . e- $variation of curvature efy with load pky =− $maximum value of stiffness kfy/fznom pky = $load at which kfy reaches maximum value pky = . $curvature of stiffness kfy phy = . $horizontal shift shy at fznom pvy = . $vertical shift in svy/fz at fznom pvy =− . $variation of shift svy/fz with load rby = $slope factor for combined fy reduction rcy = $shape factor for combined fy reduction [aligning_coefficients] qbz = $trail slope factor for trail bpt at fznom qbz = $variation of slope bpt with load qcz = . $shape factor cpt for pneumatic trail qdz = . $peak trail dpt = dpt*(fz/fznom*r ) qdz =− . $variation of peak dpt with load qez =− . $trail curvature ept at fznom qez = $variation of curvature ept with load appl. sci. , , of [temperature_coefficients] ty =− . $temperature effect on cornering stiffness magnitude ty = . $temperature effect on location of cornering stiffness peak ty = . $linear temperature effect on lateral friction ty =− . $quadratic temperature effect on lateral friction tx =− . $linear temperature effect on slip stiffness tx = . $quadratic temperature effect on slip stiffness tx = . $linear temperature effect on longitudinal friction tx =− . $quadratic temperature effect on longitudinal friction tref = $reference temperature note: the non-mentioned coefficients are equal to zero. references . formula sae. available online: https://www.fsaeonline.com (accessed on october ). . formula student team delft. available online: https://www.fsteamdelft.nl (accessed on october ). . ammon, d. vehicle dynamics analysis tasks and related tyre simulation challenges. veh. syst. dyn. , , – . [crossref] . hirschberg, w.; rill, g.; weinfurter, h. user-appropriate tyre-modelling for vehicle dynamics in standard and limit situations. veh. sys. dyn. , , – . [crossref] . li, j.; zhang, y.; yi, j. a hybrid physical-dynamic tire/road friction model. j. dyn. syst. meas. control , , . [crossref] . yavari, b.; tworzydlo, w.; bass, j. a thermomechanical model to predict the temperature distribution of steady state rolling tires. tire sci. technol. , , – . [crossref] . clark, s.k. temperature rise times in pneumatic tires. tire sci. technol. , , – . [crossref] . maniowski, m. optimisation of driver actions in rwd race car including tyre thermodynamics. veh. sys. dyn. , , – . [crossref] . durand-gasselin, b.; dailliez, t.; mössner-beigel, m.; knorr, s.; rauh, j. assessing the thermo-mechanical tametire model in offline vehicle simulation and driving simulator tests. veh. sys. dyn. , , – . [crossref] . pearson, m.; blanco-hague, o.; pawlowski, r. tametire: introduction to the model. tire sci. technol. , , – . [crossref] . farroni, f.; giordano, d.; russo, m.; timpone, f. trt: thermo racing tyre a physical model to predict the tyre temperature distribution. meccanica , , – . [crossref] . sorniotti, a. tire thermal model for enhanced vehicle dynamics simulation; no. - - , sae technical paper ; sae: warrendale, pa, usa, . . ozerem, o.; morrey, d. a brush-based thermo-physical tyre model and its effectiveness in handling simulation of a formula sae vehicle. imeche part d j. automob. eng. , , – . [crossref] . kasprzak, e.m.; gentz, d. the formula sae tire test consortium-tire testing and data handling; no. - - , sae technical paper ; sae: warrendale, pa, usa, . . dynamic test center. available online: https://www.dtc-ag.ch (accessed on october ). . kistler wheel force transducer. available online: https://www.kistler.com (accessed on october ). . milliken, w.f.; milliken, d.l. race car vehicle dynamics; society of automotive engineers warrendale: warrendale, pa, usa, . . kelly, d.; sharp, r. time-optimal control of the race car: influence of a thermodynamic tyre model. veh. sys. dyn. , , – . [crossref] . fevrier, p.; le maitre, o. tire temperature modeling: application to race tires. in proceedings of the vdi th international congress numerical analysis and simulation in vehicle engineering, wurzburg, germany, – september ; pp. – . . savkoor, a. some aspects of friction and wear of tyres arising from deformations, slip and stresses at the ground contact. wear , , – . [crossref] . falk, k.; nazarinezhad, a.; kaliske, m. tire simulations using a slip velocity, pressure and temperature dependent friction law; deutsche simulia anwenderkonferenz: dresden, germany, . https://www.fsaeonline.com https://www.fsteamdelft.nl http://dx.doi.org/ . / http://dx.doi.org/ . /vesd. . . . http://dx.doi.org/ . / . http://dx.doi.org/ . / . http://dx.doi.org/ . / . http://dx.doi.org/ . / . . http://dx.doi.org/ . / http://dx.doi.org/ . /tire. . http://dx.doi.org/ . /s - - - http://dx.doi.org/ . / https://www.dtc-ag.ch https://www.kistler.com http://dx.doi.org/ . / . . http://dx.doi.org/ . / - ( ) - appl. sci. , , of . huemer, t.; liu, w.n.; eberhardsteiner, j.; mang, h.a. a d finite element formulation describing the frictional behavior of rubber on ice and concrete surfaces. eng. comput. , , – . [crossref] . lemaitre, j. nonlinear models and properties. in handbook of materials behavior models, nd ed.; elsevier: amsterdam, the netherlands, . . svendenius, j.; gäfvert, m.; bruzelius, f.; hultén, j. experimental validation of the brush tire model. tire sci. technol. , , – . [crossref] . van rijk, s. development and modelling of a formula student racing tyre. master’s thesis, eindhoven university of technology, eindhoven, the netherlands, . . oosten, j.; unrau, h.; riedel, a.; bakker, e. tydex workshop: standardisation of data exchange in tyre testing and tyre modelling. veh. sys. dyn. , , – . [crossref] . mf-tyre. mf-swift . -help manual; tno automotive: helmond, the netherlands, . . bakker, e.; pacejka, h.b.; lidner, l. a new tire model with an application in vehicle dynamics studies. sae trans. , – . [crossref] . pacejka, h.b.; besselink, i. tire and vehicle dynamics, rd ed.; elsevier: amsterdam, the netherlands, . . harsh, d. full vehicle model of a formula student car: full-car vehicle dynamics model incorporating a tyre model which includes the effects of temperature on it’s performance. master’s thesis, delft university of technology, delft, the netherlands, . . formula student germany e.v. formula student rules. available online: https://www.formulastudent.de/ fsg/rules/ (accessed on october ). . schafer, r.w. what is a savitzky-golay filter? ieee signal process. mag. , , – . [crossref] © by the authors. licensee mdpi, basel, switzerland. this article is an open access article distributed under the terms and conditions of the creative commons attribution (cc by) license (http://creativecommons.org/licenses/by/ . /). http://dx.doi.org/ . / http://dx.doi.org/ . / . http://dx.doi.org/ . / http://dx.doi.org/ . / https://www.formulastudent.de/fsg/rules/ https://www.formulastudent.de/fsg/rules/ http://dx.doi.org/ . /msp. . http://creativecommons.org/ http://creativecommons.org/licenses/by/ . /. 	introduction 	experimental setup and results 	setup and test program 	pure cornering results 	pure acceleration and braking results 	results related to temperature effect 	thermal model 	proposed thermal model 	comparison with the established models 	comparison results 	modified magic formula incorporating temperature effect 	baseline parameter estimation 	basic magic formula 	proposed modification for temperature effect 	integration into full fsae vehicle model 	full vehicle modeling 	model structure 	validation 	conclusions 	—tire property file 	references r hat.qxp c o v e r f e a t u r e - / /$ . © ieee computer p u b l i s h e d b y t h e i e e e c o m p u t e r s o c i e t y a middleware-based, component-oriented software system consists of • an integrating middleware layer that abstracts the execution environment and implements services and communication channels, and • a network of cooperating components that run on the middleware services and enforce the business logic. this clear separation of infrastructure and applica- tion modules, and the ability to easily compose such modules, naturally suggests three distinct development roles: the product-line architect sets the system archi- tecture, selects infrastructure platforms, and organizes the development process; the component developer builds the business-logic modules; and the component integrator assembles the components into systems. the “product-line development roles” sidebar describes these roles in more detail. advances in architecture description languages and metamodeling environments have enabled software product managers to hide the complexities of lower- level implementation details by defining structural abstractions of components, interfaces, connectors, and system assemblies that can be visualized and analyzed existing software modeling approaches fail to support many product-line development activities. the cadena platform, together with its core modeling concept, the cadena architecture language with metamodeling, addresses this deficiency by providing a highly adaptive type-centric modeling framework with robust, flexible, and extensible tool support. adam childs, jesse greenwald, georg jung, matthew hoosier, and john hatcliff kansas state university l arge-scale software development efforts are increasingly based on product lines, a develop- ment process in which developers build the soft- ware for similar product families from reusable infrastructure and common application compo- nents. by emphasizing systematic reuse, the product-line approach can reduce development and production time as well as overall costs by a factor of times or more. using component middleware frameworks supports the product-line approach by • providing well-defined interfaces that prevent client code from becoming unnecessarily tangled with low-level implementations; and • making units easier to plug and unplug, which facil- itates reuse and system evolution. product-line development based on such frameworks has been successful in numerous application domains ranging from large-scale distributed real-time and embed- ded computing systems used in mission- and safety-crit- ical domains—including commercial air traffic control, military systems, electrical power grids, industrial process control, and medical imaging—to user-level operating- system and desktop-application integration. calm and cadena: metamodeling for component-based product-line development february and that can drive automatic generation of various forms of infrastructure code. however, these modeling tools are often not directly targeted to product-line development roles. calm/cadena the cadena development-tool platform, together with its core modeling concept, the cadena architecture language with metamodeling (calm), addresses this deficiency by providing a highly adaptive type-centric modeling framework with robust, flexible, and exten- sible tool support. calm calm (www.cadena.projects.cis.ksu.edu/calm) is an architecture description language that enables strongly typed modeling of platforms, components on these plat- forms, and component assemblies of concrete scenar- ios. it also supports inheritance-based hierarchical organization of platforms with aspect mechanisms for product-line development roles product-line architects, component developers, and component integrators each play a distinct role in middle- ware-based, component-oriented software product-line development. product-line architects product-line architects initially analyze commonalities and differences among systems in the product family as well as any inherent dependencies between their fea- tures and capabilities. guided by this analysis, they con- struct a common software infrastructure, including a component model and supporting services, that devel- opers can reuse for all products within the family. product-line architects also define coding and modeling guidelines to constrain and organize the development process. developers can use numerous industry-standard compo- nent platforms, such as the corba component model and enterprise javabeans, to support product-line development. most out-of-the box implementations of ccm and espe- cially ejb also add functionality that is not part of the plat- form specification. in addition, specialized platforms have been developed to build specific, large-scale projects—for example, the boeing bold stroke/prism model, which is used for avionics-control systems, and the gnome-bonobo platform, which is the basis for the gnome desktop-man- ager event system. given the number of different component models and implementations targeted at various classes of applications (such as embedded versus nonembedded), it is often benefi- cial to organize the product line in a platform-independent manner by working at a level of abstraction that avoids com- mitting to specific underlying component models or imple- mentations.this allows using different middleware implementations for different products and can facilitate migration to middleware platforms across the product line’s lifetime. the product-line architect is responsible for developing rules, guidelines, and even automated transformations for migrating platform-independent models and artifacts to platform-specific models and implementations. in large- scale systems of systems, architectures may need to support multiple component models and platforms. component developers component developers fall into two broad categories. common component developers design components that are intended to be reused across multiple projects, as well as abstract versions of more special components that are refined—for example, via inheritance—and implemented for use in specific projects. project-specific component developers implement com- ponents for particular systems and customize or refine common components to satisfy functional requirements for their portion of the system. components that offer related functionality are grouped into modules or libraries. the component developer carefully models the event and data interfaces provided to client components as well as those on which the component depends as component ports.the component integrator resolves these dependen- cies later during system assembly and configuration. components can use infrastructure services such as a persistence, time, event, or replication. often the configura- tion settings for such services are exposed by the compo- nent developer in metadata associated with the component so that the component integrator can choose a particular setting after determining the component’s context. component integrators component integrators work on specific projects.they attempt to satisfy the functional and real-time requirements of a particular system by hooking together general-purpose and project-specific components drawn from a component library and by selecting distribution strategies, execution priorities, and infrastructure-specific communication services. this phase benefits most obviously from tool support, such as automatic valuation of metadata or deployment information through implemented heuristics, or more sim- ply through graphical display and editing of assemblies. partial assemblies that support certain recurring functional- ities (subassemblies) can be saved as libraries similar to component-type libraries. computer product-line architects use the top style tier to define architectures and capture middleware infrastructure properties by specifying languages for building types of components, interfaces, and connectors, along with appropriate metadata schemas. component developers use those languages on the middle module tier to define component and interface types within a particular archi- tecture. system integrators use the bottom scenario tier to allocate instances of declared component and con- nector types, connect allocated component instances, and configure underlying middleware, services, and network deployments by setting model-level attribute values. while calm exists as a textual language, develop- ers use specialized spreadsheet-based or graphical input/output and editing environments to integrate it into cadena, making calm/cadena intuitively easy to use with little training. cadena cadena (www.cadena.projects.cis.ksu.edu) provides a variety of forms of support for creating, editing, query- ing, and browsing and transforming calm models. calm models are connected to underlying component middleware frameworks as well as analysis and code- generation facilities via cadena plug-ins. these plug-ins effectively serve as model interpreters that realize the semantics of calm models. cadena provides a spec- trum of plug-in points, letting users add specialized func- tionality for the component systems modeled at each calm stage. each plug-in is associated with a partic- ular calm architecture type, and inheritance on archi- tecture types guarantees effective plug-in reuse on models as they are migrated to more specific platforms. cadena is engineered from the ground up to serve as an extensible tool platform rather than as a single tool. based on eclipse (www.eclipse.org), ibm’s widely used open source integrated development environment, cadena itself is implemented as a series of eclipse plugs-ins. cadena exploits the eclipse modeling framework’s autocoding facilities so that developers can create extremely robust, efficient tool features such as incremental type checking and model-change propagation. cadena plug-in points are available to associate emf adapter factories with calm styles, enabling cadena plug-in developers to add specialized constraint languages, constraint checkers, involved analysis frameworks, automated design heuris- tics, and autocoding and deployment facilities. modeling in calm/cadena the starting point of a calm specification is the def- inition of component, connector, and interface kinds to describe a metamodel of a system’s architectural ele- ments at the style tier. in accordance with type theory, each kind defines a language of types at the module tier, and every software code unit, or instance, at the sce- nario tier is derived from a type. for example, a single incorporating specific platform attributes into general architectural descriptions. the framework offers a rigorous, type-based approach for transitioning platform-independent models to plat- form-specific models via a series of tool-supported refine- ment steps. these steps incrementally incorporate more structural details regarding particular component mod- els, underlying middleware, and product development contexts. they also augment the models with increasingly precise and detailed metadata, attributes, and weave-ins. basic modeling primitives in calm are based on the three fundamental types of entities that define every component-based system and, more generally, every sys- tem of communicating processes: • business-logic containers—the locations of compu- tation (components); • services—the locations of communication (connec- tors); and • the interaction points between components and con- nectors that define the assumptions and guarantees (interfaces). the formation and use of these entities is realized in calm using a tiered metamodeling approach aligned with the three product-line development roles. a model in a particular tier defines the language or vocabulary of entities that can be used in constructing models in the tier below it. figure . cadena style editor. the table, or spreadsheet, view reveals basic corba component model entities in their calm definition. component instance has a certain compo- nent type, and the component type is defined within its component kind. to spec- ify component kinds, calm uses the con- cept of metakinds—toolboxes that serve as a factory for new types of components, interfaces, and connectors. the relation between instance and type is the same as that between type and kind. although the concept is simple, this addi- tional level of abstraction gives calm/ cadena the flexibility to provide specialized development environments for a huge class of platforms. style tier figure shows the cadena style editor with the common object request broker architecture (corba) component model specification (www.omg.org/technology/ documents/formal/components.htm). the table, or spreadsheet, view reveals basic ccm entities in their calm definition, such as the highlighted ccmcomponent with its capabilities to send and receive events (consumes, emits, publishes) or to provide and use data interfaces (provides, uses). note that the ccmcomponent is derived from a specific template, the metakind mccmcomponent. component kinds. these describe a platform’s soft- ware unit primitives and possible mechanisms that these units may use to interact with one another. the possible interaction points, called port options in calm, pro- vide a means to declare explicit context dependencies— a central theme of component-based development. the product-line architect uses port options to specify the kinds of interfaces a component kind can feature, together with the necessary keywords—for example, consumes for ccm event input ports, and publishes or emits for ccm event output ports, as shown in figure . further, developers can add attributes, typed through calm’s own extensible type system, to the component kind to define its ability to carry metadata. these attrib- utes—which can be specified for the whole kind, types therein, or concrete instances—can hold data for deploy- ment information, initialization values, and all sorts of functional, structural, or organizational content. connector kinds. these model the services the plat- form provides. the chosen platform can include vari- ous middleware services supporting intercomponent communication, distribution, persistence, and state replication, among others. each connector kind repre- sents a distinct platform service or family of services. each connector definition may contain multiple role dec- larations, much like port options in component kinds, and interface-type constraints governing each role. single-role connectors abstract services such as time-out generators, while multirole connectors model inter- component communication services. note that connectors are abstractions, which are not meant as descriptions of the services’ implementations. rather, they describe the service’s intended usage and thus enforce certain coding styles. interface kinds. these categorize interaction points of platform components and check component-connec- tor compatibility. for example, the ccm component interacts with the asynchronous connector through event interfaces and the synchronous connector through ccm data interfaces. interface kinds are defined by a set of attributes in the same manner as component kinds. for the calm ccm style, for example, module-tier attributes for interface kinds describe possible method signatures of ccm interface types. after an architectural style is specified within cadena, the tool suite does not remain static. rather, upon alter- ation of a style, the lower tiers adapt and offer new and changed primitives in that style’s language. module tier working with the kinds—that is, type languages— defined for a particular architectural style, component developers use the module tier to define libraries of design-time component types that conform to that style. component and interface kinds from the style are avail- able in the cadena module editor, shown in figure , to february figure . cadena module editor. while operating under the calm ccm style, the editor adapts to provide customized ccm support. computer create component and interface types. the library of design-time types in the module is then available for allo- cating and connecting component instances at the sce- nario tier. building types at the module tier is analogous to defin- ing component types using the ccm interface definition language in the standard ccm development process. module-tier component types are typically associated with a cadena autocoding plug-in that generates stubs, skeletons, and other infrastructure code required for component implementation. after applying autocoding plug-ins to generate skeleton implementations, compo- nent developers can use eclipse development environ- ments—for java or c++, for example—to complete component implementations by coding the business logic associated with components. scenario tier to allocate and connect component instances at the scenario tier, cadena offers two standard development environments: a tree-based editor and a graphical editor. figure shows the graphical editor open to a ccm sys- tem. at the top is an instance diagram representing the communication component shown in figure . within the graphical editor, the user can insert new components and configure connections through drag-and-drop matching of the connector role and component port. large system assemblies with many components can overwhelm developers and inhibit their ability to reason effectively about a system. to help developers build impor- tant structural abstractions that collect multiple cooper- ating components together in a single unit, cadena supports the notion of a nested assembly—essentially a virtual component that is actually realized as a nested net- work of connected component instances. cadena’s graph- ical scenario editor allows a developer to form a nested assembly from a set of components, view this nested assembly as one entity in the graphical scenario editor, and expand and edit the nested assembly as needed. product-line support product-line development includes many involved tasks such as designing and establishing platform spec- ifications, refining these specifications for more specific projects, combining distinct platform specifications into hybrid platforms, migrating components and assemblies from one platform to the next, augmenting a platform specification to accommodate deployment or metadata, figure . cadena scenario editor. the user can insert new components and configure connections through drag-and-drop matching. and integrating standard functionality into different platforms. inheritance hierarchies to support these tasks, calm introduces inheritance hierarchies, an object-oriented programming concept, into style design. for one calm style specification to inherit from another means that the component, connector, and interface kinds defined in the parent style are also present in the substyle. calm further allows multiple inheri- tance—that is, a style can name multiple parent styles. in addition to facilitating style specifications, a well- organized hierarchy of architectural styles serves as a con- ceptual basis for code reuse over distinct platforms by providing a guide for type-safe migration of artifacts from compliance with one style to compliance with another. as figure a shows, the specialization of generic plat- form specifications toward more individual platforms corresponds to a parent-to-child transition, while the abstraction from a particular platform to a more generic, reusable form corresponds to a child-to-parent transi- tion, as figure b shows. in addition, the reuse of elements from one platform on another generally corresponds to model migration, as figure c illustrates. finally, as figure d shows, one platform inheriting from multiple platforms yields a hybrid or combination platform. developers also can use this operation to establish integrative styles that mit- igate between platforms as well as to factor in standard definitions that are reusable over multiple, otherwise unrelated styles. attribute attachment a second mechanism for refining architectural styles, and models therein, is attribute attachment, shown in figure e. calm offers an extendable value-type sys- tem that allows the definition of new types through names or combinations of existing types. attributes typed in this system can be defined directly within a style specification or on separate attribute sheets, which can be added at every model tier. some values that attributes capture can be specific to particular scenarios, such as deployment or location information. other sorts of attributes recur in a general context, such as structures used to describe method sig- natures in interfaces, or in specific contexts such as com- ponent run rates in real-time systems. developers can use attribute sheets to factor in general data over mul- tiple systems as well as to factor out specific data in par- ticular systems—for example, to allow viewing a projected system’s structural interrelations without hav- ing to set real-time data. example product line one of the benchmark test cases for calm/cadena is boeing’s avionics open experimental platform (oep), developed as part of the defense advanced research projects agency’s program composition for embedded systems project and available for exchange within the academic research community. the avionics oep is inspired by boeing’s bold stroke mission-control soft- ware product-line architecture for its military avionics platforms such as the f- . the oep implements prism, a model that consists of three component kinds. the business component is equivalent to the ccm component—that is, it is the generic container for business logic. the correlator is a specialized component that lowers network load through asynchronous message filtering. the event- channel component serves as the source for periodic time-out events. the event-channel component models an infrastructure unit that also manages threading and buffers asynchronous communication, but these tasks are deliberately hidden in the abstraction of the services. february (a) (b) (c) (d) (e) style inheritance model transformation figure . calm style hierarchies. (a) model specialization. (b) model abstraction. (c) model migration. (d) hybrid model construction. (e) attribute sheet attachment. computer figure shows a table view of the prism style in cadena. prism offers two types of services: an untyped, asynchronous, event-propagation mechanism with a sta- tically set publish-subscribe connection scheme (event connection) and a synchronous, typed, data transmis- sion service (interface connection). figure a illustrates using cadena to model a single- processor modal scenario from the oep. as figure b shows, cadena seamlessly supports the three different kinds of prism components, which can be assigned cus- tom shapes, within the three calm tiers. adding to the communication infrastructure like many other prism examples, modalsp exploits control-push data-pull, a dual connection in which a component supplying data first sends an asynchronous message announcing that new data is available, then the receiving component uses a synchronous connection to retrieve the data. this guarantees that the synchronous connection never blocks or waits for updates. the con- trol-push data-pull strategy links many components by two connectors instead of one. when developers use this approach, tool support must be available to ensure that both connectors’ end points match. in calm, it is also possible to define a dedicated connector with both syn- chronous and asynchronous connection points that cap- tures the control-push data-pull protocol in a single entity. adding real-time metadata in addition to the communication network, a prism system configuration consists of deployment informa- tion and metadata such as initialization values, security levels, run rates, and location data. calm features an attribute-attachment system that lets users weave in attributes orthogonal to the style hierarchy on each tier of the development process. cadena provides multiple forms of support for attribute management, including declaration and checking of domain-specific attribute types, and plug-in facilities for processing attribute data values. for example, prism developers must assign a priority value to each component event-handler port following a particular set of development heuristics that involve trac- ing data, event, and control dependencies across net- works of components. in the cadena prism development environment, a prism plug-in automates calculation of these priority settings by coding the previously manually implemented heuristics, significantly reducing the time and effort required for this particular task. this is an instance of a general strategy for using a model-centric approach to minimize development effort and increase overall confidence levels: important system attributes are presented at the modeling level and vari- ous domain-specific analyses, configuration mecha- nisms, automated developer advice, and integrity checks are automated through plug-ins tailored to a particular development context. advanced platform organization the calm style hierarchy is more than a mechanism for capturing features or abstractions of existing compo- nent platforms; it provides a structured approach for leveraging commonalities and capturing differences across families of closely related platforms. for example, the boeing bold stroke line entails some abstract parent styles from which the actual prism platform definition inherits, as well as child styles for particular platforms such as the f/a- hornet fighter, which in turn branches into the f/a- e and f/a- f super hornet aircraft. the guiding principle for style inheritance is the reuse of infrastructure implementation as represented by calm kinds. reimplementations and new additions within the substyle, which appear in the definition of the substyle as new kinds, integrate naturally into the specification. systems designed for distinct platforms often must cooperate in personal computing—for example, desk- figure . boeing bold stroke/prism style in cadena. prism offers an untyped, asynchronous, event-propagation mechanism with a statically set publish-subscribe connection scheme as well as a synchronous, typed, data transmission service. top systems integrate ccm and bonobo components— and even more frequently in companywide business inte- gration. calm is the first conceptual framework to accomplish straightforward type-safe merging of plat- forms. the strategy is to create a child style that inher- its from each style to be merged. within the child style, developers can easily specify necessary components or services that act as mitigators between the parent styles. t he design principles underlying calm/cadenahave been influenced by extensive interaction withproduct-line architects at boeing phantom works and lockheed martin. in addition, collaboration with real-time middleware and modeling experts at vanderbilt university to integrate cadena with the school’s cosmic (component synthesis with model-integrated com- puting) ccm environment (www.dre.vanderbilt.edu/ february instances kinds instance of extends types module tier meta-kinds style tier scenario tier minterfacemcomponent mconnector mprismcomponent mprismeventconnector mprismtimeout mprismevent mprisminterfacemprismconnector prismcorrelatorprismcomponent prismtimer prismtimeout prismevent prisminterface device modal modal modesource modechange dataavailable data time-out binaryand binaryand binaryand navdisplay tacticalsteering navsteering lazyactive airframe pilotcontrol navsteeringpoints gps navigator passive modal display pushdatasource mprismcorrelator mprismtimer mprismtimeoutconnector mprisminterfaceconnector figure . boeing’s single-processor modal scenario in cadena. (a) graphic view. (b) specifying modalsp within calm. (a) (b) computer cosmic), generic modeling environment (www.isis.van- derbilt.edu/projects/gme), and ciao (component integrated ace orb) real-time component middleware (www.cs.wustl.edu/~schmidt/ciao.html) have provided significant opportunities to enhance and refine our prod- uct-line modeling ideas. compared to many existing approaches that focus on modeling and analyzing single-system assemblies in a fixed component model, calm/cadena is a rigorous type- based framework for modeling multiple component mid- dleware platforms, systematically organizing and tran- sitioning between platform definitions, and creating cus- tomized development environments that leverage domain knowledge and automate development process steps to enable early design decisions for entire product lines. the framework continues to evolve. the initial set of features in cadena . , soon to be released, focuses on capturing and specifying basic entities with their struc- tural interrelations. this will enable users to establish platform terminologies, build and assess topologies, pro- vide basic autocoding and deployment facilities, design attribute-seeding mechanisms, and conduct fundamen- tal analysis such as slicing and cycle detection. because a main focus of cadena is its extensibility, custom views of component systems can also easily be added. as for calm, research is under way to flexibly attach various forms of behavioral specification. this will enable much finer grained dependence analysis and eval- uation of systems’ temporal behavior, both absolute tim- ing (real-time and schedulability analysis) and relative timing (temporal logic properties, model checking, and safety analysis). more information about calm/cadena and other analysis and verification tools built by the laboratory for specification, analysis, and transformation of software at kansas state university can be found at the santos lab home page (www.cis.ksu.edu/santos). ■ acknowledgment this work was supported in part by the defense advanced research projects agency as part of the program composition for embedded systems project (contract f - -c- ) and the army research office’s high-confidence embedded systems initiative (contract daad - - - ). references . p. clements and l. northrop, software product lines: prac- tices and patterns, addison-wesley professional, . . k. schmid and m. verlage, “the economic impact of prod- uct line adoption and evolution,” ieee software, vol. , no. , , pp. - . . a. ledeczi et al., “the generic modeling environment,” proc. ieee int’l workshop intelligent signal processing, ; www.isis.vanderbilt.edu/publications/archive/ledeczi_a_ _ _ _the_generi.pdf. . b.c. pierce, types and programming languages, mit press, . adam childs is pursuing an ms in computer science at kansas state university. his research interests include soft- ware architecture, model-driven development, and compo- nent-based software. he is a member of the acm. contact him at achilds@cis.ksu.edu. jesse greenwald is a research associate in the department of computing and information sciences at kansas state university. his research interests include model-driven development and component-based software. greenwald received an ms in software engineering from kansas state university. contact him at jesse@cis.ksu.edu. georg jung is a phd candidate and graduate research assis- tant in the department of computing and information sci- ences at kansas state university. his research interests include specification, typing, and verification of software architectures and component-based systems. jung received a diplom in computer science from saarland university, germany. contact him at jung@cis.ksu.edu. matthew hoosier is a research associate in the department of computing and information sciences at kansas state university. his research interests include model-checking and software specification and verification techniques. hoosier received an ms in computer science from kansas state university. contact him at matt@cis.ksu.edu. john hatcliff is a professor in the department of comput- ing and information sciences at kansas state university, where he also leads the santos laboratory. his research interests include model-driven development, program analy- sis and verification, and software engineering for safety- and mission-critical systems. hatcliff received a phd in com- puter science from kansas state university. he is a mem- ber of the acm. contact him at hatcliff@cis.ksu.edu. n o t a m e m b e r ? j o i n o n l i n e t o d a y ! www.computer.org/join doi: . /j.foreco. . . forecasting landscape-scale, cumulative effects of forest management on vegetation and wildlife habitat: a case study of issues, limitations, and opportunities stephen r. shifley a,*, frank r. thompson iii a , william d. dijak a , zhaofei fan b, a northern research station, u.s. department of agriculture, forest service, anheuser-busch natural resources building, university of missouri, columbia, mo - , usa b department of forestry, university of missouri, anheuser-busch natural resources building, university of missouri, columbia, mo - , usa received march ; received in revised form august ; accepted august abstract forest landscape disturbance and succession models have become practical tools for large-scale, long-term analyses of the cumulative effects of forest management on real landscapes. they can provide essential information in a spatial context to address management and policy issues related to forest planning, wildlife habitat quality, timber harvesting, fire effects, and land use change. widespread application of landscape disturbance and succession models is hampered by the difficulty of mapping the initial landscape layers needed for model implementation and by the complexity of calibrating forest landscape models for new geographic regions. applications are complicated by issues of scale related to the size of the landscape of interest (bigger is better), the resolution at which the landscape is modeled and analyzed (finer is better), and the cost or complexity of applying a landscape model (cheaper and easier is better). these issues spill over to associated analyses that build on model outputs or become integrated as auxiliary model capabilities. continued development and application of forest landscape disturbance and simulation models can be facilitated by ( ) cooperative efforts to initialize more and larger landscapes for model applications, ( ) partnerships of practitioners and scientists to address current management issues, ( ) developing permanent mechanisms for user support, ( ) adding new capabilities to models, either directly or as compatible auxiliary models, ( ) increasing efforts to evaluate model performance and compare multiple models running on the same landscape, and ( ) developing methods to choose among complex, multi-resource alternatives with outputs that vary over space and time. # elsevier b.v. all rights reserved. www.elsevier.com/locate/foreco available online at www.sciencedirect.com forest ecology and management ( ) – keywords: forest landscape model; landis; harvest; landsum; simpple; vddt; validation; multi-resource evaluation; succession; disturbance . introduction over the past years, forest landscape disturbance and succession models have matured into practical tools for large- scale analysis and planning. prior to the s it was relatively easy to conceive of the components necessary for a landscape model that could forecast the spatial and temporal outcomes of forest management, but implementation was hampered by available software and computing capacity. today, bolstered by * corresponding author. tel.: + x ; fax: + . e-mail address: sshifley@fs.fed.us (s.r. shifley). currently at college of forest resources, forest and wildlife research center, box , mississippi state university, mississippi state, ms - , usa. - /$ – see front matter # elsevier b.v. all rights reserved. doi: . /j.foreco. . . vast improvements in computing capacity and geographic information system (gis) software, a number of forest landscape models and decision support systems have evolved into applied tools for evaluating effects of forest disturbance, including management practices, at the landscape scale. this evolution has been propelled by the needs of forest mangers to account for large-scale cumulative effects of proposed management alternatives. many of the available forest landscape change models and decision support systems have been summarized in key documents and databases including barrett ( ), mladenoff and baker ( ), mowrer et al. ( ) and gordon et al. ( ) http://ncseonline.org/ncssf/dss/documents/index.htm. the subset of forest landscape disturbance and succession models generally fall in to three broad groups: http://ncseonline.org/ncssf/dss/documents/index.htm mailto:sshifley@fs.fed.us http://dx.doi.org/ . /j.foreco. . . s.r. shifley et al. / forest ecology and management ( ) – ( ) p olygon-based systems that model changes through time for somewhat homogenous vegetation units (e.g., forest stands > ha in extent) represented as mapped polygons. for each polygon, forest change over time due to succession or disturbance is modeled with a finite set of forest conditions (states) and pathways (deterministic or prob- abilistic transitions among states). cumulative landscape change is defined by the collective spatial and temporal changes for the polygons. examples of polygon-based models include landsum (keane et al., , ), simpple (chew, ; chew et al., in press) and vddt/ telsa (beukema and kurz, ). ( ) r aster-based models that uniformly divide the landscape into a mosaic of square sites (i.e., rasters or pixels). the area of a site is user defined but typically falls within the range . ha to km . modeled spatial processes such as seed dispersal, tree establishment, harvest or fire affect individual sites. cumulative changes across the entire landscape or any subset are represented as the cumulative change (and variation) for the included sites. examples of raster-based models include landis (he et al., , ; mladenoff and he, ; mladenoff, ) and harvest (gustafson and crow, ; gustafson and rasmussen, ). ( ) t ree-based growth and yield modeling systems with spatial modeling capability. these are an extension of individual- tree-based models that predict changes (growth, mortality, harvest, and ingrowth) for a statistically representative sample of trees from a forest stand. predicted changes to the individual sampled trees can be aggregated to estimate stand-level change. over the past two decades improve- ments in spatial analysis tools and computing capacity have provided options to model stand dynamics within multiple stands simultaneously while also modeling spatially explicit process that move among stands and/or depend on the pattern of stand adjacencies (e.g., harvest, insect dispersal, or fire). cumulative landscape impacts are represented by the combined changes and patterns in individual trees and stands over time. examples of this class of models include fvs (dixon, ) and lms (e.g., mccarter, ; mccarter et al., ). this confluence of technologies has blurred the distinction between traditional growth and yield models and landscape disturbance and succession models.the order ( – ) represents a general pattern of increasing detail in (a) initial data requirements, (b) model complexity, (c) computation load, and (d) options for mapping or summarizing results. it is important to match management information needs with model cap- abilities, complexity, and data requirements. in theory, the simplest model that can provide the required information is preferable; in practice the model that has the fewest perceived barriers to implementation, regardless of com- plexity, is often the one applied. for more than a decade, we have applied the landis model of forest landscape disturbance and succession in the midwestern united states and concurrently applied gis-based wildlife habitat suitability index models (larson et al., ). as landis development continues and the mechanics of model implementation become less onerous, we face an emerging set of issues common to application of many landscape disturbance and succession models: (a) matching appropriate models to the tasks at hand, (b) overcoming complexity of calibration for new geographic regions, (c) defining initial conditions for a large landscape, (d) testing or validating results at the landscape scale, (e) understanding each model’s strengths and weaknesses, (f) securing computing resources for large landscapes and/or multi- resource evaluations, (g) finding user support, and (f) quantita- tively comparing anticipated cumulative effects of complex management alternatives. in this paper we discuss our experience with landis and our observations relative to these issues. where possible we relate the issues to the broader set of forest landscape disturbance and succession models. . landis applications in the midwest . . the landis model the landis landscape disturbance and succession model is thoroughly described in numerous sources (e.g., he et al., , ; mladenoff and he, ; mladenoff, ), and landis web sites provide additional background, reference material and access to software (http://web.missouri.edu/ �umcsnrlandis/, http://landis.forest.wisc.edu/documentation). landis was initially developed and applied in wisconsin (usa) (e.g., he et al., ; he and mladenoff, ; scheller and mladenoff, ; sturtevant et al., b; zollner et al., ; radeloff et al., ), with subsequent applications elsewhere in the united states (larson et al., ; syphard and franklin, ; wimberly, ; franklin et al., ; shifley et al., ), canada (van damme et al., ), china (he et al., ; wang et al., ), and europe (pennanen and kuuluvainen, ; schumacher et al., ). briefly, landis estimates at - or -year intervals the presence or absence of tree species or species groups by age class for each site (i.e., raster). tree species succession is governed by a set of stochastic rules based on species’ vital attributes (e.g., longevity, age at initiation of seed production, seed dispersal distance, fire tolerance, shade tolerance, and sprouting probability). species and age class dynamics are affected by stochastically modeled disturbances due to harvest, fire, or wind (weather); disturbance frequency and size distribution are user-defined and vary by ecological land type. . . application on national forests our applications of landis have been in the ozark highlands of southeastern missouri and in south-central indiana (shifley et al., , , ; larson et al., ). those landscapes range from steeply dissected, rocky missouri ozark plateaus dominated by black oak (quercus velutina lam.), scarlet oak (quercus coccinea muenchh.), white oak (quercus alba l.), post-oak (quercus stellata wangenh.) and shortleaf pine (pinus echinata mill.) to mesic, http://web.missouri.edu/~umcsnrlandis/ http://web.missouri.edu/~umcsnrlandis/ http://landis.forest.wisc.edu/documentation s.r. shifley et al. / forest ecology and management ( ) – rolling, fertile hills in indiana with complex mixtures of the trees in the red oak and white oak groups, sugar maple (acer saccharum marsh.), yellow-poplar (liriodendron tulipifera l.) and numerous species of lesser abundance. landis applications in missouri have primarily been on contiguous portions of the mark twain national forest ranging from a few thousand (shifley et al., ) to more than , ha in extent (shifley et al., ). on the mark twain we have examined the long-term effects of alternative management regimes on forest structure and tree species composition and on wildlife habitat suitability (larson et al., , ; shifley et al., ). in indiana our applications have been directed at forecasting, as realistically as possible, the outcomes of the alternatives proposed in the forest plan (united states department of agriculture forest service, ) for the , ha hoosier national forest. that analysis included the development and application of landscape-level habitat suitability models (rittenhouse et al., ) that utilized landis output to forecast changes in wildlife habitat suitability. national forests are partitioned into discrete purchase units (proclamation boundaries) within which the national forest ownership is concentrated but which also include intermixed private lands. while working with the hoosier national forest we also made provisional forecasts of expected changes on adjacent private lands within each purchase unit. . . landis initialization as the landis model has evolved the documentation and user interface have improved to the point where application of landis software is within the grasp of any determined analyst with a modern computer. the contemporary barriers to application are more often related to finding suitable data to define initial landscape conditions and to calibrating the relationships that govern forest dynamics (e.g., seed dispersal, species establishment, windthrow, and wildfire). maps of landforms or ecological landtypes are user-defined in landis and required for differentiating spatial processes and event probabilities that are expected to differ among landtypes (he et al., ). although, ecological classification systems are widely available, in our applications ecological landtypes and landforms were not fully mapped across our study areas. consequently, mapping ecological landtypes across our study region, personally or via contractors, was an initial requirement. elevation changes across our study areas are generally less than m and a reduced set of ecological landtypes representing northeast (cool) slopes, southwest (warm, dry) slopes, wide ridges, narrow ridges, small drainages, and mesic riparian areas proved sufficient and could be derived from a digital elevation model. mapping ecological landtypes can be time consuming, but there are numerous examples of methods and/or algorithms for doing so from spatial data (e.g., franklin, ; shao et al., ; syphard et al., a,b). some investigators have used remotely sensed data and ecological landtypes to coarsely define dominant forest cover and age class (e.g., for sites > ha in size), the two essential vegetation characteristics needed to describe initial vegetation conditions (he et al., ) in our applications we needed a smaller site (raster size) to more realistically model selection harvests, windthrow, and landscape elements (e.g., length of edge habitat) relevant to wildlife species of interest. conse- quently, we used site sizes of . or . ha and based initial estimates of dominant forest cover and age on stand-level inventories maintained on the hoosier and mark twain national forests. mapping initial forest age and dominant cover type on privately owned lands was more problematic; private owner- ships in our study area typically range from to acres and have no site-specific forest inventory data. existing gis coverages are available to ( ) distinguish private lands from public lands, ( ) distinguish forest land from nonforest, and ( ) in some cases to distinguish conifer cover from hardwoods. we used those data layers to map the location of private forest ownerships and then used state-level data on the frequency distribution of ownership sizes (birch, ) to create and overlay a hypothetical, ownership grid with approximately the same frequency distribution of parcel sizes. we used the intersection of ownership boundaries and mapped ecological landtypes to define locations for private land management units (i.e., a substitute for stand boundaries). we then summarized forest inventory and analysis (fia) field data (united states department of agriculture forest service, ) for private ownerships in the region and used the frequency distribution of forest area by dominant cover type, age class, and ecological landtype to populate initial forest conditions for each privately owned management unit. thus, we knew that the univariate frequency distributions of forest area by age class and by dominant species generally matched that for private forest lands as a whole, but the spatial patterning of those units was only spatially representative, not spatially exact. the process for deriving initial map layers differs with available data sources; other published methodologies include species distribution modeling (franklin, ) and hierarchical bayesian techni- ques (he et al., ). . . landis calibration tree species establishment coefficients in landis differ by landtype to characterize differential rates of species recruitment among landtypes. we used fia and other inventory data to guide initial estimates of species establishment coefficients and then used long-term projections of species change to estimate trends, discuss trends with local experts, and refine coefficients. we found derivation of the species establishment coefficients the most difficult part of the calibration process because they do not correspond to any readily measurable inventory character- istic. nevertheless landis has proved amenable to calibration for more than a dozen different ecosystems by users in north america, europe and asia (mladenoff, ). calibration of algorithms for exogenous disturbances caused by fire, windthrow, and harvest add additional layers of complexity to landis. federal and state agencies generally have good historical records of wildfires that can be used to estimate frequency and size of fire events. the heterogeneous s.r. shifley et al. / forest ecology and management ( ) – spatial patterns of wildfire events are increasingly being described in mathematical and mapped formats (haight et al., ; yang et al., ; shang et al., ; guyette et al., ). historical estimates of weather damage to forests are not as common. for our study regions we were limited to a few sources for small-scale windthrow (rebertus and meier, ), tornadoes (national oceanic and atmospheric administration, ), and ice damage (rebertus et al., ). landis currently has no mechanism for simulating the intense, linear impacts of tornado events which are infrequent, but not inconsequential in their impacts on forest structure over a century of change. harvest disturbance is typically user defined and supported by well documented historical records or well defined future plans. consequently, simulated harvest events are generally easier to model that wind or fire events (gustafson et al., ). . . linking habitat suitability models forecasting changes in wildlife habitat quality or population size is often an objective of landscape simulation, and we developed and applied gis-based wildlife habitat suitability models in conjunction with the landis for this purpose. habitat suitability index (hsi) models estimate habitat suitability on a scale of (not suitable) to (highly suitable) based on an assessment of resource attributes considered important to a species’ abundance, survival, or reproduction (u.s. fish and wildlife service, , ). individual habitat attributes are modeled as suitability indices (sis) represented as mathematical or graphical relationships. overall habitat suitability, or the hsi, is calculated as a mathematical combination of the individual sis. the hsi is typically calculated as the geometric mean of the individual sis, although more complex formulas can be used depending on how the sis are thought to interact. recent developments in habitat suitability index modeling have resulted in models that can be applied to large landscapes through the utilization of gis (duncan et al., ; gustafson et al., ; larson et al., , ; rittenhouse et al., ). landscape-scale, gis-based hsi models can address ecologi- cal and landscape effects on wildlife such as area sensitivity, edge effects, interspersion, landscape composition, and juxtaposition of resources (larson et al., , ; ritten- house et al., ). these models rely on data layers derived from remote sensing and other existing spatial databases or from large-scale inventories and can include data layers produced by landis or other simulation models that represent future forest conditions. with gis-based models, si and hsi values are calculated for each pixel in the landscape and the distribution of hsi values for all the pixels in a landscape can be summarized with maps or descriptive statistics, or used as inputs to other models (e.g., larson et al., ; shifley et al., ) (fig. ). the models we applied in these projects are available in a stand-alone software package, landscape his models (http://www.nrs.fs.fed.us/hsi; dijak et al., ), that can be applied to data from a wide range of sources. . discussion . . model scale and detail in our work we faced some obvious and some non-obvious issues of scale. our use of a . ha or . ha site (raster size) for landis and hsi model implementation in the midwest was due to our desire to assess the impacts to wildlife habitat of the harvest or death of single trees or small groups of trees. the . ha raster size had an intuitive appeal—one raster cell was roughly equivalent to the crown area of a mature tree and allowed realistic spatial representation of the uneven-aged, individual-tree selection silviculture that is commonly prac- ticed in indiana. the . ha raster sized was approximately the size of a regeneration opening used with group selection silviculture in missouri. disturbance events smaller than the selected model resolution (e.g., small windthrow events) cannot be explicitly modeled, and the spatial arrangement of vegetation within a site (raster) is considered to be (a) homogenous, (b) indeterminate, or (c) inconsequential. raster cell size is a critical decision for wildlife hsi modeling as well as for vegetation modeling. gis-based hsi models frequently include suitability functions that assess habitat interspersion, patch size, and distance to edge; these metrics differ depending on how patches and edges are defined. we were aware that a decrease in raster cell size results in an exponential increase in the number of raster cells and creates at least an exponential increase in data processing time. never- theless, for landscapes about million raster cells in size landis could process a century of landscape change in about h using a high-end computer workstation. for the , ha hoosier national forest which we processed in five geogra- phically discrete sections using a . ha raster cell size, we could model five different management alternatives for years in about week of computer time. our experience has been that we run every project at least three times. this is usually due to unanticipated issues with data processing or to errors in the design of a model scenario. corrections generally mean rerunning the entire scenario. in our work, a larger concern than running landis to model future tree species age structure and species composition is the effort required for post-processing landis outputs via gis or custom spatial analysis programs to summarize landscape statistics and habitat suitability. for example, to use landscape hsimodels (dijak et al., ) to calculate habitat suitability for nine species on the hoosier national forest, at three time steps ( , , and years of elapsed time), for five management alternatives required h of computer processing time. as we acquire the ability to work with larger landscapes and derive more characteristics of interest from modeled scenarios, post processing considerations will become even more influential in defining the scope of tractable projects. to some extent, buying more computing capacity can mitigate issues of project size and processing time, but it is surprisingly easy to design a project that imposes unreasonable computing demands. the available forest landscape succession and disturbance models and decision support systems vary in complexity and http://www.nrs.fs.fed.us/hsi fig. . starting from initial gis layers mapping landforms, tree age classes, tree species composition, and management units (stands) for the mark twain national forest, we used landis to analyze and communicate expected outcomes of alternative management strategies over time and space. we could forecast forest age structure spatially over for four alternatives (shown for the final year of a years projection; darker sites have older trees). from the modeled scenarios it is possible to derive habitat suitability for numerous wildlife species (two shown, darker is better habitat), and a wide variety of other derived values. similar maps and values can be estimated for each year or each decade of the modeled scenarios. this figure highlighting ha within the , ha modeled landscape illustrates some of the many types of information that can be generated, how that information can be communicated to professionals or the public, and the complexity of choosing a preferred alternative in a spatially explicit, multi-resource modeling environment. s.r. shifley et al. / forest ecology and management ( ) – capability. thus, when addressing complex forest landscape modeling problems it is important to match the model to the question at hand and then continue to revaluate options and required outcomes. for example, for applications where harvest effects are the primary agent of forest change and detailed information about species succession is not required, a model such as harvest (gustafson and crow, ; gustafson and rasmussen, ) may be more appropriate than landis. harvest carries less detail about each site but is easier to initialize and can manipulate large landscapes that are impractical to investigate using landis (or can run equivalent landscapes significantly faster). similarly, if the management questions can be addressed using the aggregate information for larger sections of forest (e.g., based on forest stands that are roughly – ha in extent) then models such as vddt/telsa (beukema and kurz, ), simmpple (chew et al., in press) or landsum (keane et al., , ) may be fully adequate yet easier to calibrate and require less computational processing. despite the tradeoffs among scale, detail, conceptual complexity, and computational complexity, the effort and expense of implementing any landscape disturbance and succession model are so demanding that we are unaware of any side-by-side comparisons of different landscape models applied to address the same scenarios on the same large landscape. there are, however, some useful publications and web-based tools to compare model features and capabilities (barrett, ; mowrer et al., ; gordon et al., ; http:// ncseonline.org/ncssf/dss/documents/index.htm). the high overhead associated with initializing and applying any land- scape disturbance and succession model creates a high incentive to continue with the simulation model that one starts with, limitations notwithstanding. thus, model selection should also consider adaptability of the model to address the range of http://ncseonline.org/ncssf/dss/documents/index.htm http://ncseonline.org/ncssf/dss/documents/index.htm s.r. shifley et al. / forest ecology and management ( ) – other landscape-scale management issues that are likely to arise after a model has been successfully implemented in given region. . . organizational issues applying a landscape model to address practical or policy- relevant problems requires a team approach. maps of initial landscape conditions (e.g., land types, forest cover, and forest age) rely on shared data from multiple sources. information about disturbance patterns from fire, wind, and harvest often come from specialists with local experience. it is not difficult for scientists to conjure up theoretical scenarios for application of a landscape disturbance and succession model, and those exercises can be enormously instructive in understanding cumulative effects of management decisions (e.g., shifley et al., ). however, some of the most challenging problems are those faced by public land managers with specific management issues where they must define objectives and alternatives, quantify expected outcomes, interpret the implications, select a course of action, communicate decisions to the public, implement the practice, and live with the consequences of either an accurate or an erroneous prognostication. strong partnerships among practitioners, scientists, and technical experts are necessary to allow landscape simulation science to continue the transition from theory to practice. in our experience, the individuals with the greatest expertise for interpreting the ecological, social, and policy implications of model results are generally not those with the specialized skills for running the landscape model. for example, staff specialists with forest management agencies are often the local experts on the ecological and social implications of present and future forest conditions. however, they often lack the time or motivation to learn how to calibrate and apply a landscape succession and disturbance model. this tendency is exacerbated by the fact that forest planning efforts are usually periodic rather than continuous, and planning involves periods of intense modeling and data analysis followed by years of plan implementation. thus, there are potential efficiencies in creating a service center approach—maintaining a core group of landscape modeling experts who are available to serve multiple clients with planning or other issues suitable for model applications. we have observed that when a forest landscape succession and disturbance model has been initialized and demonstrated, new model applications for that landscape usually follow. for example, the ability to apply landis on national forests in missouri and indiana has generated new lines of research on fuel management and fire risk, additional habitat analyses for wildlife species of conservation concern, and implications for public land management priorities when evaluated within a surrounding matrix of private land managed with different management practices. . . missing pieces landscape disturbance and succession models or related decision support systems, despite their substantial capabilities, are ripe for further research and development. there are significant factors affecting landscape change that are not addressed in contemporary models. perhaps the most obvious is land use change (e.g., the movement of land among categories of forest, agriculture, urban, and suburban). although the total area of forest land in the united states has decreased by only about percent in the last years, the total proportion of affected acres is vastly greater because over time acres that shift out of forest cover are offset by tracts elsewhere that move back to forest cover (alig et al., ). urbanization tends to permanently erode forest area whereas shifts between forest and agricultural lands are cyclical. although we lack the ability to forecast many types of land use change in a spatially explicit manner within forest landscape disturbance and succession models, the ability to link external landscape change models with forest landscape disturbance and succession models has been demonstrated for a large landscape in california (syphard et al., a,b), and that capability will undoubtedly continue to improve. in fact, the feasibility of incorporating complex spread processes in raster- based landscape disturbance and succession models has already been demonstrated through the incorporation of routines that forecast how insects and fire differentially move across a landscape in response to resource availability, topography, and/or weather (e.g., sturtevant et al., a; yang et al., ). . . wildlife population change and viability hsi models assess habitat quality—the potential for a pixel, patch, or landscape to support a species. however, actual wildlife population sizes or population growth rates can be limited by factors other than habitat availability (e.g., survival, fecundity, dispersal, age structure, spatial population structure). one approach to addressing population viability or growth is to link spatially explicit wildlife species viability models to landscape disturbance and succession models (liu et al., ; larson et al., ; akcakaya et al., , ). we used ramas gis software (applied biomathematics, setauket, ny, http://www.ramas.com/software.htm) to model population growth for individual wildlife species in con- junction with landis projections and to assess population viability and growth in a changing landscape (larson et al., ). the process of simultaneously applying landscape change and wildlife population growth models is becoming easier; the ramas software now offers a version that is integrated with landis (http://www.ramas.com/landsc.htm) and it has been used to predict wildlife viability under different fire and tree harvest scenarios (akcakaya et al., , ). . recommendations several steps can be taken to improve the utility and applicability of forest landscape disturbance and succession models, and they are important to the continued evolution of the technology. http://www.ramas.com/software.htm http://www.ramas.com/landsc.htm s.r. shifley et al. / forest ecology and management ( ) – . . initialize large landscapes for application the biggest barriers to implementing landis and similar landscape disturbance and succession models are (a) mapping initial landscape conditions (e.g., land cover, ecological land type, forest composition, age structure) for large landscapes with mixed ownership, (b) calibrating the tree species successional dynamics, and (c) calibrating the disturbance and management events (e.g., wind, fire, harvest, fuel treatment, biological agents). a rough estimate is that about % of our effort went into landscape initialization and model calibration, % to running scenarios, and % to post-processing and summarizing results. although the calibration and initialization tasks can be time consuming, they are certainly not overwhelming and have been completed in the course of landscape model applications in a wide array of settings (mladenoff, ; barrett, ; mowrer et al., ; gordon et al., ). after a model is initialized for a given landscape, it can usually support a wide range of analyses that address many different issues. often the process of initializing a model for a large landscape is only marginally more difficult and time consuming than doing if for a smaller landscape. thus, it is efficient and benevolent to initialize forest conditions across large landscapes and to share the maps and data among cooperators interested a wide array of landscape issues. that way more effort can be directed toward model application and analyses than toward landscape initialization. . . provide continuing software support as landscape models move from research tools to application tools for forest planning, continuing technical support is increasingly important. this can occur through agency and institutional support or through software commer- cialization. to date, support for landscape models has primarily been from model developers who themselves are scientists. an example of software support that has worked for other forestry computer applications is the national service center (http:// www.fs.fed.us/fmsc/) approach used for the forest vegetation simulator (fvs) (dixon, http://www.fs.fed.us/fmsc/fvs/ index.shtml) and other forest growth and yield software. as growth and yield modeling capacity became integral to public land management, the u.s. forest service established a work group that provides user support, develops standardized user interfaces, adds new features, calibrates models for new ecoregions, conducts validation tests, and integrates the models with silvicultural and planning operations. one difference between forest growth and yield models and forest landscape models is that the latter have fewer individual users working on projects that are much larger in scale and scope. it took more than years for that on-going support system to evolve for fvs, and once established it has continued for more than years. landscape disturbance and succession models are now more than years into development and consideration of a similar system of ongoing support is timely. presumably such an approach would provide support for multiple modeling systems and assist users in public and private sectors. this approach could result in faster model implemen- tation, creation of auxiliary software to interface with agency databases, and development of specialized reporting formats. . . validate models validation of landscape disturbance and succession models or wildlife hsi models is challenging. statistical comparisons of observed and predicted patterns of landscape change or habitat suitability are limited by (a) lack of long-term or large- scale temporal and spatial data for forest landscapes and wildlife, and by (b) the stochastic nature of most landscape models in which disturbance events at a particular location are the result of random draws from a probability distribution for that event. most often validation efforts are brief affairs where a structured series of simulation runs are used to compare patterns of change with expert opinion. over time, experience with model predictions under a wide range of forest landscape conditions and disturbance scenarios affords increased under- standing of the range of conditions where the model behavior is reasonable and useful. greater attention to evaluation and validation of landscape disturbance and succession models and wildlife habitat or population models is warranted, and there is substantial guidance in the literature (e.g., morrison et al., ; rykiel, ; vanclay and skovsgaard, ; pontius et al., ; burgman, ). we are currently validating some of the hsi models described in this paper with independent data sets on abundance and reproductive success. . . develop methods to rank the multidimensional outcomes of management alternatives the ability to analyze multiple scenarios and simultaneously consider the interactions and cumulative effects of decisions is essential to making informed management choices. landscape disturbance and succession models are valuable because they provide a framework for simultaneously assessing and mapping effects of management alternatives not only for landscape pattern, vegetation composition, and vegetation age structure, but also for timber resources, wildlife habitat, water resources, and aesthetics. the emerging challenge is to develop techniques to simultaneously display, analyze, and understand the levels, interactions and trade-offs among all these factors (fig. ). although quantitative methods previously have been applied to similar kinds of natural resource decision making problems (e.g., haight et al., ), the field is relatively new and small efforts to merge decision making techniques with landscape disturbance and succession models could yield significant accomplishments. applied landscape modeling projects are so complex that they may never be amenable to traditional optimization techniques, but there certainly are quantitative tools that could be used in a more rigorous approach to selecting among management alternatives. . conclusions landscape disturbance and succession models have demonstrated utility in estimation of large-scale, long-term http://www.fs.fed.us/fmsc/ http://www.fs.fed.us/fmsc/ http://www.fs.fed.us/fmsc/fvs/index.shtml http://www.fs.fed.us/fmsc/fvs/index.shtml s.r. shifley et al. / forest ecology and management ( ) – cumulative effects of forest management. they can be used for traditional forest planning and to address emerging scientific and policy issues related to forests and wildlife. barriers to application are the ( ) technical skills need to run the models and analyze the findings, ( ) lack of sufficiently detailed information about initial landscape conditions, ( ) effort needed for model calibration for new regions, ( ) limited number of suitable links to other factors of interest (e.g., wildlife habitat, fuels, forest amenities), ( ) limited user support, and ( ) difficulty selecting among complex manage- ment alternatives with multi-resource interactions that vary over space and time. several scientific and administrative actions can help landscape disturbance and succession models continue their evolution as tools for policy-relevant analyses. first, continue to initialize more and larger landscapes so they are ready for model applications. second, foster ongoing collaborative model applications for forest planning and related analyses in cooperation with forest managers. third, provide support for continuing software improvement. fourth, add new capabilities to the models, either directly or as compatible auxiliary models and software. these capabilities may include forecasts of land use change, impact of insects and disease agents, wildlife population viability estimators, or models for catastrophic weather events such as hurricanes and tornadoes. fifth, more formally evaluate model performance; identify and quantify conditions where the models perform well and where they do not. sixth, provide support for multiple modeling methodol- ogies and multiple spatial scales. landscape models differ in their strengths and limitations. the ability to apply multiple models to the same problem provides greater insight to the individual models and to the range of potential outcomes. seventh, explore quantitative methods that can guide evaluation and comparison of management alternatives with complex, multi-resource tradeoffs that vary over space and time. acknowledgements our applications of the landis software were made possible through the support and encouragement of landis developers hong he (department of forestry, university of missouri, columbia) and david mladenoff (department of forest ecology and management, university of wisconsin- madison). michael schanta (mark twain national forest), guofan shao (purdue university), judi perez (hoosier national forest) and dale weigel (hoosier national forest) provided essential maps, data layers, and advice. janet franklin and an anonymous reviewer provided insightful comments that substantially improved an earlier version of this manuscript. we thank them all. references akcakaya, h.r., radeloff, v.c., mlandenoff, d.j., he, h.s., . integrating landscape and metapopulation modeling approaches: viability of the sharp- tailed grouse in a dynamic landscape. conserv. biol. , – . akcakaya, h.r., franklin, j., syphard, a.d., stephenson, j.r., . viability of bell’s sage sparrow (amphispiza belli ssp. belli) under altered fire regimes. ecol. appl. , – . alig, r.j., plantinga, a.j., ahn, s.e., kline, j.d., . land use changes involving forestry in the united states: – , with projections to . general technical report pnw- . u.s. department of agriculture, forest service. pp. barrett, t.m., . models of vegetation change for landscape planning: a comparison of fetm, landsum, simpplle, and vddt. general technical report rmrs- -www. u.s. department of agriculture, forest service. pp. beukema, s.j., kurz, w.a., . vegetation dynamics development tool: user’s guide (v. . ). available from: essa technologies ltd., # - west th avenue, vancouver, bc v j c . pp. published on line. http://www. essa.com/downloads/telsa/ http://www.essa.com/downloads/vddt/index.htm. birch, t.w., . private forest-land owners of the northern united states, . resource bulletin ne- . u.s. department of agriculture, forest service. pp. burgman, m.a., . risks and decisions for conservation and environ- mental management. cambridge university press, cambridge, pp. chew, j.d., . development of a system for simulating vegetative patterns and processes at landscape scales. ph.d. dissertation. university of mon- tana, missoula. pp. chew j.d., stalling, c., moeller k., bella, e., schreiner, r., ahl, r., jones, t., in press. user guide for simpplle, version . (draft). general technical report rmrs-xxx. u.s. department of agriculture, forest service. pp. published on line. http://www.fs.fed.us/rm/missoula/ /simpplle/ files/publications/simpplle-manual- - .pdf. dijak, w.d., rittenhouse, c.d., larson, m.a., thompson iii, f.r., millspaugh, j.j., . landscape habitat suitability index software. j. wildl. manage. ( ), – . dixon, g.e. (comp.), . essential fvs: a user’s guide to the forest vegetation simulator. internal report. u.s. department of agriculture, forest service, forest management service center, fort collins, co. pp. published on line. http://www.fs.fed.us/fmsc/fvs/description/index.shtml. duncan, b.w., breininger, d.r., schmalzer, p.a., larson, v.l., . validat- ing a florida scrub jay habitat suitability model using demographic data on kennedy space center. photogramm. eng. remote sens. , – . franklin, j., . predicting the distribution of shrub species in southern california from climate and terrain-derived variables. j. veg. sci. , – . franklin, j., . clustering versus regression trees for determining ecological land units in the southern california mountains and foothills. for. sci. , – . franklin, j., syphard, a.d., he, h.s., mladenoff, d.j., . altered fire regimes affect landscape patterns of plant succession in the foothills and mountains of southern california. ecosystems , – . gordon, s.n., johnson k.n., reynolds, k.m., crist, p., brown, n., ., decision support systems for forest biodiversity: evaluation of current systems and future needs. final report project a . national commis- sion on science and sustainable forestry. published on line. http:// www.ncssf.org. gustafson, e.j., crow, t.r., . harvest: linking timber harvesting stra- tegies to landscape patterns. in: mladenoff, d.j., baker, w.l. (eds.), spatial modeling of forest landscape change: approaches and applications. cam- bridge university press, cambridge, uk, pp. – . gustafson, e.j., rasmussen, l.v., . harvest for windows v . : user’s guide. u.s. department of agriculture, forest service, north central research station, st. paul, mn. published on line. http://www.ncrs.fs. fed.us/ /harvest/v /documentation/default.asp. gustafson, e.j., shifley, s.r., mladenoff, d.j., nimerfro, k.k., he, h.s., . spatial simulation of forest succession and timber harvesting using landis. can. j. for. res. , – . gustafson, e.j., murphy, n.l., crow, t.r., . using a gis model to assess terrestrial salamander response to alternative forest management plans. j. environ. manage. , – . guyette, r.p., muzika, r.m., dey, d.c., . dynamics of an anthropogenic fire regime. ecosystems , – . http://www.essa.com/downloads/telsa/ http://www.essa.com/downloads/telsa/ http://www.essa.com/downloads/vddt/index.htm http://www.fs.fed.us/rm/missoula/ /simpplle/files/publications/simpplle-manual- - .pdf http://www.fs.fed.us/rm/missoula/ /simpplle/files/publications/simpplle-manual- - .pdf http://www.fs.fed.us/fmsc/fvs/description/index.shtml http://www.ncssf.org/ http://www.ncssf.org/ http://www.ncrs.fs.fed.us/ /harvest/v /documentation/default.asp http://www.ncrs.fs.fed.us/ /harvest/v /documentation/default.asp s.r. shifley et al. / forest ecology and management ( ) – haight, r.g., cleland, d.t., hammer, r.b., radeloff, v.c., rupp, t.s., . assessing fire risk in the wildland–urban interface. j. forest. ( ), – . haight, r.g., snyder, s.a., revelle, c.s., . metropolitan open-space protection with uncertain site availability. conserv. biol. , – . he, h.s., mladenoff, d.j., . spatially explicit and stochastic simulation of forest landscape fire disturbance and succession. ecology , – . he, h.s., mladenoff, d.j., radeloff, v.c., crow, t.r., . integration of gis data and classified satellite imagery for regional forest assessment. ecol. appl. , – . he, h.s., mladenoff, d.j., boeder, j., . an object-oriented forest landscape model and its representation of tree species. ecol. model. , – . he, h.s., hao, z., larsen, d.r., dai, l., hu, y.m., chang, y., . a simulation study of landscape scale forest succession in northeastern china. ecol. model. , – . he, h.s., li, w., sturtevant, b.r., yang, j., shang, z.b., gustafson, e.j., mladenoff, d. j., . landis, a spatially explicit model of forest landscape disturbance, management, and succession: landis . user’s guide. general technical report nc- . u.s. department of agriculture, forest service. pp. he, h.s., dey, d.c., fan, x., hooten, m.b., kabrick, j.m., wikle, c.k., fan, z., . mapping pre-european settlement vegetation at fine resolutions using a hierarchical bayesian model and gis. plant ecol. ( ), – . keane, r.e., long, d., basford, d., levesque, b.a., . simulating vegeta- tion dynamics across multiple scales to assess alternative management strategies. in: conference proceedings – gis , th annual symposium on geographic information systems – integrating spatial information technologies for tomorrow, gis world, inc., vancouver, british columbia, canada, pp. – . keane, r.e., parsons, r., hessburg, p., . estimating historical range and variation of landscape patch dynamics: limitations of the simulation approach. ecol. model. , – . larson, m.a., dijak, w.d., thompson iii, f.r., millspaugh, j.j., . general technical report nc- . u.s. department of agriculture, forest service. pp. larson, m.a., thompson iii, f.r., millspaugh, j.j., dijak, w.d., shifley, s.r., . linking population viability, habitat suitability, and landscape simu- lation models for conservation planning. ecol. model. , – . liu, j., duning jr., j.b., pulliam, h.r., . potential effects of a forest management plan on bachman’s sparrows (aimophila aestivalis): linking a spatially explicit model with gis. conserv. biol. ( ), – . mccarter, j.b., . integrating forest inventory, growth and yield, and computer visualization into a landscape management system. in: teck, r., moeur, m., adams, j. (comps.), proceedings of the forest vegetation simulator conference. general technical report int-gtr- . ogden, ut. usda forest service, intermountain research station. pp. – . mccarter, j.b., wilson, j.s., baker, p.j., moffett, j.l., oliver, c.d., . landscape management through integration of existing tools and emerging technologies. j. forest. , – . mladenoff, d.j., . landis and forest landscape models. ecol. model. , – . mladenoff, d.j., baker, w.l. (eds.), . advances in spatial modeling of forest landscape change: approaches and applications. cambridge uni- versity press, cambridge, uk, pp. mladenoff, d.j., he, h.s., . design and behavior of landis, an object- oriented model of forest landscape disturbance and succession. in: mladen- off, d.j., baker, w.l. (eds.), advances in spatial modeling of forest landscape change: approaches and applications. cambridge university press, cambridge, uk, pp. – . morrison, m.l., marcot, b.g., mannan, r.w., . wildlife-habitat relation- ships: concepts and applications. the university of wisconsin press, madison, pp. mowrer, h.t. (technical compiler), barber, k., campbell, j., crookston, n., dahms, c., day, j., laacke, j., merzenich, j., mighton, s., rauscher, m., sojda, r., thompson, j., trenchi, p., twery, m., . decision support systems for ecosystem management: an evaluation of existing systems. general technical report rm- . u.s. department of agriculture, forest service. pp. national oceanic and atmospheric administration, . tornado climatol- ogy. national oceanic and atmospheric administration. published on line. http://www.ncdc.noaa.gov/oa/climate/severeweather/tornadoes.html. pennanen, j., kuuluvainen, t., . a spatial simulation approach to natural forest landscape dynamics in boreal fennoscandia. for. ecol. manage. , – . pontius jr., r.g., huffaker, d., denman, k., . useful techniques of validation for spatially explicit land-change models. ecol. model. ( ), – . radeloff, v.c., mladenoff, d.j., gustafson, e.j., scheller, r.m., he, h.s., zollner, p.a., akçakaya, r., . modeling forest harvesting effects on landscape pattern in the northwest wisconsin pine barrens. for. ecol. manage. , – . rebertus, a.j., meier, a.j., . blowdown dynamics in oak-hickory forests of the missouri ozarks. j. torrey bot. soc. ( ), – . rebertus, a.j., shifley, s.r., richards, r.h., roovers, l.m., . ice storm damage to an old-growth oak-hickory forest in missouri. am. midl. nat. , – . rittenhouse, c.d., dijak, w.d., thompson iii, f.r., millspaugh, j.j., . development of landscape-level habitat suitability models for wildlife species in the central hardwoods region. general technical report nrs- . u.s. department of agriculture, forest service. pp. rykiel jr., e.j., . testing ecological models: the meaning of validation. ecol. model. ( ), – . scheller, r.m., mladenoff, d.j., . a spatially interactive simulation of climate change, harvesting, wind, and tree species migration and projected changes to forest composition and biomass in northern wisconsin, usa. glob. change biol. , – . schumacher, s., bugmann, h., mladenoff, d.j., . improving the formula- tion of tree growth and succession in a spatially explicit landscape model. ecol. model. , – . shang, z., he, h.s., lytle, d.e., shifley, s.r., crow, t.c., . long-term effects of fire suppression on central hardwood forests in missouri ozarks, usa. for. ecol. manage. , – . shao, g., parker, g.r., zhalnin, a.v., merchant, p., albright, d., . gis protocols in mapping ecological landtypes for the hoosier national forest. north. j. appl. for. , – . shifley, s.r., thompson iii, f.r., larsen, d.r., mladenoff, d.j., . model- ing forest landscape changes in the ozarks: guiding principles and pre- liminary implementation. general technical report nc- . u.s. department of agriculture, forest service, pp. – . shifley, s.r., thompson iii, f.r., larsen, d.r., dijak, w.d., . modeling forest landscape change in the missouri ozarks under alternative manage- ment practices. comput. electron. agric. , – . shifley, s.r., thomson iii, f.r., dijak, w.d., larson, m.a., millspaugh, j.j., . simulated effects of forest management alternatives on landscape structures and habitat suitability in the midwestern united states. for. ecol. manage. , – . sturtevant, b.r., gustafson, e.j., li, w., he, h., a. modeling biological disturbances in landis: a module description and demonstration using spruce budworm. ecol. model. , – . sturtevant, b.r., zollner, p.a., gustafson, e.j., cleland, d.t., b. human influence on the abundance and connectivity of high-risk fuels in mixed forests of northern wisconsin, usa. landscape ecol. ( ), – . syphard, a.d., franklin, j., . spatial aggregation effects on the simulation of landscape pattern and ecological processes in southern california plant communities. ecol. model. , – . syphard, a.d., clarke, k., franklin, c.j., a. simulating fire frequency and urban growth in southern california coastal shrublands, usa. landscape ecol. , – . syphard, a.d., yang, j., franklin, j., he, h.s., keeley, j.e., b. calibrating a forest landscape model to simulate high fire frequency in mediterranean- type shrublands. environ. model. softw. ( ), – . united states fish and wildlife service, . habitat evaluation procedures (hep). division of ecological services manual , washington, dc. united states fish and wildlife service, . standards for the develo- pment of habitat suitability index models for use in the habitat http://www.ncdc.noaa.gov/oa/climate/severeweather/tornadoes.html s.r. shifley et al. / forest ecology and management ( ) – evaluation procedure. division of ecological services manual , washington, dc. united states department of agriculture forest service, . land and resource management plan, hoosier national forest. u.s. department of agriculture, forest service. published on line. http://www.fs.fed.us/r / hoosier/planningdocs/final_docs/ _documents.htm. united states department of agriculture forest service, . the forest inventory and analysis database: database description and users guide version . . u.s. department of agriculture, forest service. pp. published on line at http://fia.fs.fed.us/tools-data/. van damme, l., russell, j.s., doyon, f., duinker, p.n., gooding, t., hirsch, k., rothwell, r., rudy, a., . the development and application of a decision support system for sustainable forest management on the boreal plain. j. environ. eng. sci. , s –s . vanclay, j.k., skovsgaard, j.p., . evaluating forest growth models. ecol. model. ( ), – . wang, x., he, h.s., li, x., chang, y., hu, y., xu, c., bu, r., xie, f., . simulating the effects of reforestation on a large catastrophic fire landscape in northeastern china. for. ecol. manage. , – . wimberly, m.c., . fire and forest landscapes in the georgia piedmont: an assessment of spatial modeling assumptions. ecol. model. , – . yang, j., he, h.s., shifley, s.r., gustafson, e.j., . spatial patterns of modern period human-caused fire occurrence in the missouri ozark high- lands. for. sci. , – . zollner, p.a., gustafson, e.j., he, h.s., radeloff, v.c., mladenoff, d.j., . modeling the influence of dynamic zoning of forest harvesting on a northern wisconsin landscape. environ. manage. , – . http://www.fs.fed.us/r /hoosier/planningdocs/final_docs/ _documents.htm http://www.fs.fed.us/r /hoosier/planningdocs/final_docs/ _documents.htm http://fia.fs.fed.us/tools-data/ 	forecasting landscape-scale, cumulative effects of forest �management on vegetation and wildlife habitat: a case study �of issues, limitations, and opportunities 	introduction 	landis applications in the midwest 	the landis model 	application on national forests 	landis initialization 	landis calibration 	linking habitat suitability models 	discussion 	model scale and detail 	organizational issues 	missing pieces 	wildlife population change and viability 	recommendations 	initialize large landscapes for application 	provide continuing software support 	validate models 	develop methods to rank the multidimensional outcomes of management alternatives 	conclusions 	acknowledgements 	references plant physiol. ( ) , - comparison of propylene-induced responses of immature fruit of normal and rin mutant tomatoes" received for publication january , and in revised form september , william b. mcglasson, herbert c. dostal, and edward c. tigchelaar department oj horticulture, purduie university, west lafayette, indiana abstract continuous application of propylene to to % mature fruits of normal tomato strains (lycopersicon esculentum mill.) advanced ripening in fruits of all ages by at least %. although preclimacteric respiration was stimulated by pro- pylene treatment, there was no concomitant increase in ethylene production. once ripening commenced, the rates of endogenous ethylene production were similar in both propylene-treated and untreated fruits. continuous exposure to propylene also stimu- lated respiration in inimature fruits of rin, a nonripening mu- tant. although respiration reached rates similar to those during the climacteric of comparable normal fruits there was no change in endogenous ethylene production which remained at a low level. internal ethylene concentrations in attached to % mature fruits of rin and a normal strain were similar. it is suggested that the onset of ripening in normal tomato fruit is not controlled by endogenous ethylene, although increased ethylene production is probably an integral part of the ripening processes. in commercial practice, tomato fruits are harvested at the mature green stage to facilitate shipment. satisfactory non- destructive indices of maturity are not available for tomatoes, and commercial consignments usually comprise fruits of a wide range of physiological ages. in other immature climacteric fruits, e.g., bananas and melons, uniform ripening can be achieved by treating the fruit with ethylene ( ). however, treatment of immature tomato fruits with ethylene even at high concentrations, does not result in uniform ripening. although ripening in individual fruits is advanced. lyons and pratt ( ) showed that continuous treatment with -dl/l ethylene in air reduced the time to ripen by about one-half in fruit picked at from to of the total growth period. at c, the commencement of natural ripening in tomato fruits is indicated by a simultaneous increase in respiration and ethylene production. red or orange pigments usually appear ' this research was supported by a contract grant from general foods, inc., technical center, white plains, n.y. and by grant gb from the national science foundation to e.c.t. journal paper no. . purdue university agricultural ex- periment station, purdue university, west lafayette, ind. . 'this research was completed while w. b. mcg. was on leave from the plant physiology unit, commonwealth scientific and industrial research organization, division of food research, and school of biological sciences, macquiarie university. north ryde . sydney, australia. to days later. the role of endogenous ethylene in ripening of climacteric fruits has been debated for many years. presently, most workers suggest from the accumulated evidence that ethylene plays a specific role in the initiation of ripening in most climacteric fruits ( , ). stronger evidence for the involvement of ethylene production in the ripening processes, once initiated, can be deduced from several recent studies. it has been shown for bananas that ethylene is required for nor- mal integration of various ripening processes ( , ). other workers have demonstrated that an autocatalytic increase in ethylene production in conjunction with increased respiration and other indications of ripening is a characteristic of climacteric fruits ( , ). if enhanced ethylene production acts normally to initiate and integrate the ripening processes, a major problem is to determine how ethylene evolution is con- trolled. we have approached this problem by examining the responses of immature tomato fruit to treatment with propylene. propylene. an active analogue of ethylene, was substituted for ethylene because it enables measurement of endogenous ethylene production by treated fruits ( . . i ). our studies included work with immature fruit of the mutant type rin. ri,z fruit have been reported to develop and mature normally but to remain green when normal fruit turn red ( ). little or no change in ethylene or co production has been reported in rini fruit monitored for up to days after harvest even though a gradual softening and yellowing occurred dur- ing this period ( ). while a climacteric-like increase in respira- tion can be induced in rin fruit by a - to -day exposure to ethylene or propylene. ethylene production remains low unless the fruit are injured ( ). although yellowing and softening occur in both ethylene-treated and untreated rin fruit, the fruit do not develop the flavor and texture characteristic of nor- mal cultivars. the data reported in this paper lead us to re- examine the role of endogenous ethylene in the initiation of natural ripening in the tomato fruit and to reinforce the view that increased ethylene production is an integral part of the ripening processes. materials and methods tomatoes (lycopersicon esciulentuln mill.) were grown in a greenhouse. the cultivars used were an fl ('purdue ' x 'hoosier x - ,' bred for commercial production by davis gardens inc., terre haute, ind.), 'rutgers,' and a partially isogenic strain of the rin mutant developed by two backcrosses to 'rutgers.' uniform populations of fruits were produced by restricting the fruit load per plant ( ). the fruits were picked at various intervals after anthesis. to facilitate comparisons between fruit populations and cultivars, the ages of the fruit at picking were expressed as percentages of the total growth period, % being the average time after anthesis for the appearance of red or pink color of representative fruits of each population still attached to the plant ( ). rini fruits were https://plantphysiol.orgdownloaded on april , . - published by copyright (c) american society of plant biologists. all rights reserved. https://plantphysiol.org responses of tomato fruit to propylene picked on the same time scale as were rutgers, since it was previously established that fruits set on the same date grew to full size in about the same time in both strains. after harvest the fruits were weighed, then dipped for min in a fungicidal suspension composed of . % (w/v) benomyl (methyl -butylcarbomyl- -benzimidazolcarbamate, dupont), . % (w/v) botran ( - -dichloro- -nitroaniline, upjohn), and . % (w/v) of a nonionic wetting agent. after dipping, the fruits were enclosed individually in glass jars which were ven- tilated continuously with humidified co,-free air, or humidified cog-free air containing to ,tl/l propylene, at about l/hr- g fruit tissue. air streams containing propylene were prepared by mixing metered flows of humidified co,-free air and of a purchased mixture of % (v/v) propylene in n . co and ethylene pro- duction levels were measured daily with an infrared gas analyzer and a gas chromatograph, respectively. the gas chromatograph was equipped with a flame ionization detector. the hydrocarbons were separated on a x . cm column of to mesh a at c. the lowest concentration of ethylene that could be measured was . ./l in a -ml sample. the retention times were and min for ethylene and propylene, respectively. to measure internal concentra- tions of ethylene in developing fruit, samples of gas were withdrawn by syringe from fruit immersed in h . reslults to establish the concentrations of propylene required to produce the maximum advancement of ripening, rutgers fruits, picked at and % of the total growth period were treated continuously with , , and ,tl/l. the onset of the exponential rise in endogenous ethylene production was used as a parameter of ripening. table i shows that all three con- centrations of propylene were similarly effective in advancing ripening. however, in supplementary studies with immature fruits of field grown cv. heinz , ftl! was less effec- tive than higher concentrations. therefore, ,al/l was adopted as a standard treatment to ensure the presence of a moderate excess of the gas. figure shows the response of the hybrid cultivar to propylene as a function of age at harvest. co, production by untreated and % mature fruit declined for or days after picking, rose to a peak to days after picking and then declined to a preclimacteric minimum. these changes which also have been observed consistently in rutgers (fig. ) were not accompanied by changes in ethylene production rates. treatment with propylene caused an increase in co produc- table i. comparison of effectiveniess of three conicentrations of propylente in advancing ripeninlg in inmmature rutgers fruits of ages days to ripen' % mature % mature air . i . . i . propylene , /i . ± . ± . aii/l . i . . i . ,a . . . + . the onset of the exponential rise in ethylene production was used as a parameter of ripening. the statistical limits are estimates of the standard deviations of the population (n, = ). table ii. initernial conicenztrationls of ethylene in attached fruits of rutgers anid ritl total growth rutgers rin c-( elhylene pi/i . i . ' . . . i . . i . . i . . i . . the statistical limits are estimates of the standard deviations of the population (nl = ). the % sample comprised only two fruits. tion which, in and % fruit, merged into the respiratory climacteric. these initial increases in respiration were not accompanied by increased ethylene production which remained low until the onset of ripening. as in the untreated fruits, ethylene production in propylene- treated fruit rose abruptly with the onset of ripening. color development was observed on the nd day after the beginning of the rise in ethylene production in both control and treated fruits. treatment with propylene reduced the time between harvest and ripening of fruits of all ages by at least one-half. (fig. and table i). the responses of % mature rin and rutgers fruits to propylene are compared in figure . for the first few days the rates of respiration in untreated fruits and the respiration responses to propylene were similar in both strains. rutgers fruits subsequently ripened and conformed to the patterns described for the hybrid cultivar. in contrast, respiration rates in untreated rin fruits declined gradually to a low level and ethylene production remained barely measurable. these fruits eventually turned yellow. in treated rin fruits respiration con- tinued at a high level for several days past the time of the post- climacteric decline in respiration in treated rutgers fruits but ethylene production remained at a barely measurable level. the appearance of yellow color was advanced by propylene treatment. table ii shows that the internal concentration of ethylene in attached fruits of rutgers and rin were similar. the % fruits of rin had been left on the plants for about days after comparable rutgers fruits had reached the % stage and at the time of sampling had begun to turn yellow. discussion propylene at ju/ applied continuously to immature fruits ( - % mature) of normal tomato strains advances ripening in all stages by at least %. the close agreement between this finding and that of lyons and pratt ( ) for ethylene indicates that the responses to ethylene and propylene are equivalent. our studies provide the additional information that propylene treatment of normal fruit does not stimulate ethylene production during the preclimacteric period, and furthermore, the rise in endogenous ethylene production in propylene-treated fruit is normal in rate and timing in rela- tion to color development (fig. ). the data reviewed by pratt and goeschl ( ) suggest that the autocatalytic burst in ethylene production, the respiratory climacteric, and other concomitant changes which comprise ripening in climacteric fruits result from either a gradual increase in cellular sensitivity to endogenous ethylene, a gradual increase in endogenous ethylene to a threshold con- centration, or a combination of these two possibilities. if suffi- cient exogenous ethylene (or propylene) is added to immature plant physiol. vol. , https://plantphysiol.orgdownloaded on april , . - published by copyright (c) american society of plant biologists. all rights reserved. https://plantphysiol.org mcglasson, dostal, and tigchelaar :. t e w t i-~~~~~ z i-l in w .~ z o d w z w -j )- w . ± . .± ± . . ± ± . ± . i%ii ~~~~~~i days days days fig. . average rates of co and ethylene production by propylene-treated and untreated fruits of fl tomatoes (purdue x hoosier x - ) at three stages of maturity. the vertical bars indicate estimates of the standard deviations of the population (nl = ). for clarity, estimates for alternate points on the curves were omitted. the arrows indicate the average time at which the fruits showed the first detectable red color and the numbers, with estimates of the standard deviations of the population, indicate the average elapsed time to the onset of the rapid rise in ethyl- ene production. to provide a more accurate presentation of the rates of co and ethylene production during ripening, composite curves were prepared by matching the data for individual fruits in each treatment. the day of the first significant increase in ethylene production was taken as day i of ripening. the double slashes mark the breaks in the curves necessitated by variation in time taken by individual fruits to initiate ripen- ing. tissue, the system responsible for the natural tolerance to this gas should be overwhelmed and fruits of all physiological ages should commence to ripen at about the same rate. such an observation has been made for several climacteric fruits ( ). however, in tomato fruits of normal strains the tolerance of immature fruits cannot be completely overwhelmed by ethylene or propylene treatment. ethylene or propylene treatment stimulates respiration in rin fruits as it does in preclimacteric fruits of normal ripening strains (fig. ) ( ). even after prolonged periods of treatment or storage, no increases in ethylene production have been ob- served. nevertheless, rin fruit tissues are capable of ethylene production. table ii shows that the internal levels of ethylene in developing rin and rutgers fruits are similar and herner and sink ( ) reported that wounding stimulated ethylene produc- tion by r-in fruit tissue. mcmurchie et al. ( ) showed that treating banana fruit with propylene stimulated a typical respiratory climacteric and a rise in ethylene production. in contrast, lemons and oranges, examples of nonclimacteric fruits, showed only a climacteric-like rise in respiration in response to propylene. these results were considered to sup- port previous suggestions that the critical difference between climacteric and nonclimacteric fruits rests in their relative abilities to produce ethylene in response to low concentrations of ethylene ( , ). on the basis of the similarity of the re- sponses of rin and citrus fruits to propylene, we concur with the conclusion of herner and sink ( ) that rin fruit behave as a nonclimacteric tissue. from these comparisons of the responses of different fruits, we suggest that the onset of ripening in normal strains of tomato fruit is not controlled by endogenous ethylene, although increased ethylene production appears to be an essential com- plant physiol. vol. , https://plantphysiol.orgdownloaded on april , . - published by copyright (c) american society of plant biologists. all rights reserved. https://plantphysiol.org responses of tomato fruit to propylene % '- -~ n' r w i- - o rutgers -air * rutgers - ppm propylene u rin - pptrpln - l days fig. . average rates of co and ethylene production by propylene-treated and untreated fruits of rutgers and rin. other details as for fig. . ponent of normal ripening once it has been initiated. rather, it seems that some as yet unidentified and relatively slow change must precede the burst in ethylene production and other concomitant ripening changes. the advanced ripening resulting from treatment with ethylene or propylene could be a result of the acceleration of respiration and of general senescence, a response similar to that found in detached citrus fruits ( , ). in keeping with these suggestions the lack of characteristic ripening in rin fruits, and perhaps in other non- climacteric fruits, may be due to an inability to form a specific cellular component which binds ethylene or its analogues. this specific cellular component is conceived to develop in fruits of normal strains during growth or during aging in detached fruits. the system for the autocatalytic production of ethylene and of other normal components of ripening would be integrated with the development of this specific cellular component. literature cited . biale, j. b. . growth, maturation and senescence in fruits. science : - . . burg, s. p. and e. a. burg. . role of ethylene in fruit ripening. plant physiol. : - . . herner, r. c. and k. c. sink, jr. . ethylene production and respiratory behavior of the rin tomato mutant. plant physiol. : - . . hesselman, c. w. and h. t. freebairn. . rate of ripening of initiated bananas as influenced by oxygen and ethylene. j. amer. soc. hort. sci. : - . . lyon-s, j. m. and h. k. pratt. . effect of stage of maturity and ethylene treatment on respiration and ripening of tomato fruits. proc. amer. soc. hort. sci. : - . . mcglasson-, w. b. . the ethylene factor. in: a. c. hulne, ed., the plant physiol. vol. , https://plantphysiol.orgdownloaded on april , . - published by copyright (c) american society of plant biologists. all rights reserved. https://plantphysiol.org mcglasson, dostal, and tigchelaar plant physiol. vol. , biochemistry of fruits and their products, vol. . academic press, new . reid, m. s. and h. k. pratt. . ethylene and the respiration climacteric. york. pp. - . nature : - . . micmurchie, e. j., w. b. 'mcglasson, and i. l. eaks. . treatment of . robinson, r. w. and m. l. tomes. . ripening inhibitor: a gene with fruit with propylene gives information about the biogenesis of ethylene. multiple effect on ripening. tomato genet. coop. : - . nature : - . . sfakiotakis, e. m. and d. r. dilley. . induction of autocatalytic . pratt, h. k. and j. d. goeschl. . plhysiological roles of ethylene in ethylene production in apple fruits by propylene in relation to maturity and plants. anntu. rev. plant physiol. : - . oxygen. j. amer. soc. hort. sci. : - . . quazi, am. h. and h. t. freebairn. . the influence of ethylene, oxygen, . winston, j. r. . the coloring or degreening of mature citrus fruits with and caibon dioxide on the ripening of bananas. bot. gaz. : - . ethylene. u. s. dept. agric. circ. . https://plantphysiol.orgdownloaded on april , . - published by copyright (c) american society of plant biologists. all rights reserved. https://plantphysiol.org commission - woodard, m. and noyes, r.: , nature , vii. instrumentation for the study of solar radiation and structure (j. harvey) this report complements the report on solar instrumentation presented by commission in that subjects treated there are not repeated here. both reports taken together should give a fair, though necessarily incomplete, idea of solar instrumentation activity during the period june through june . a. absolute spectral radiometry broad-band radiometers on the solar maximum mission and nimbus- satellites continue to operate and have revealed interesting changes in solar irradiance. radiometers have been flown on spacelabs and with good results (crommelynck et al., ; labs et al., ; van hoosier et al., ). the difficult region between and nm has been investigated by carlson et al. ( ). a long term program, using modern equipment and techniques, to measure solar irradiance from the ground has started (palmer et al. ). b. spectro- and differential photometry one of the most active research areas during the past three years has been precision differential photometry, often directed toward limb observations. most of these instruments involve rapid scanning of the image with detector arrays to reduce atmospheric noise. seykora ( ) uses low amplitude, high frequency chopping to detect tiny intensity variations. rosch and yerle ( ), hirayama et al. ( a), dicke et al. ( ) and herzog et al. ( ) scan parallel to the solar limb to measure facular contrast and image geometry. limb-to-limb scanning is employed by chapman et al. ( ). a unique instrument to measure solar dia- meter fluctuations, the solar disk sextant, has recently been ground tested by sofia and his colleagues. ultraviolet spectrometers and results have been described by samain and lemaire ( ), hirayama et al. ( b), parkinson and gabriel ( ) and epstein et al. ( ). a -element, linear array detector has been added to the spectrograph at purple mountain observatory (wang et al., ). at the crimean astrophysical observatory, an ingenious spectrometer using a michelson interfero- meter and lazer stabilization, has been used for precise doppler shift measurements (didkovsky et al. ). c. polarimetry careful design has allowed henson and kemp ( ) to achieve broadband circular polarization measurements to one part per million. new stokes polarlmeters are under construction by the high altitude observatory and the applied physics laboratory of johns hopkins university for installation at the national solar observatory. the high sensitivity video magnetograph of the big bear solar observatory is described by wang et al. ( ) and the video solar magnetic field telescope of the beijng observatory has been described in great detail (beijng, ). video techniques are also used in a spectromagnetograph being built at the national solar observatory (jones, ). markov ( ) present the characteristics of the -element magnetograph operated by siblzmir. d. high angular resolution the flight of spacelab in the summer of at last provided extended periods of solar observations with sub-arc-second resolution and no atmospheric at https://www.cambridge.org/core/terms. https://doi.org/ . /s x downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core terms of use, available https://www.cambridge.org/core/terms https://doi.org/ . /s x https://www.cambridge.org/core radiation & structure of the solar atmosphere distortion. the visible and ultraviolet instruments and preliminary results are described by title et al. ( ) and brueckner et al. ( ) respectively. rocket flights, though brief, continue to be an important method for obtaining high resolution observations. the transition region camera was flown in (dame' et al., ) and first experimental flights of normal incidence x-ray telescopes have been conducted by groups from the center for astrophysics (golub et al., ), lockheed and stanford. on the ground, speckle imaging techniques have at last been developed into a practical tool for achieving nearly diffraction- limited, white-light images of small solar structures (laville conde et al., ). adaptive optics experiments continue at the national solar observatory by groups headed by r. smithson and j. beckers. the msdp spectrograph at pic du midi has been improved and now delivers half-arc-second, two-dimensional spectroscopy (insu), ). e. infrared instrumentation using equipment designed for stellar observations, remarkable far-infrared observations of the solar limb structure, limb darkening and oscillations have been made both from high-altitude aircraft (lindsey et al., ) and from the ground (lindsey and kaminsky, ). balloons have been used to observe the thermal emission of dust near the sun (mizutani et al., ), and for mapping the disk at . and micrometers (zou, ) and , and micrometers (degiacomi et al., ). excellent spectra were obtained by a fourier transform spectrometer flown on spacelab and a michelson interferometer flown on a balloon (borelko and clark, ). two-dimensional arrays of sensitive infrared detectors are becoming available for use in solar observations (graves et al., ) and a remarkable new pair of photographic emulsions have been produced that are sensitive as far as . micrometers (shcherbakova et al. ). f. radio instrumentation and techniques frequency agile receiving equipment has become a popular addition to solar radio telescopes (hurford et al., ; white et al. ). earth rotation image synthesis has been employed by alissandrakis et al. ( ) to obtain solar images at mhz and by a group at siblzmir to obtain microwave images (smolkov et al., ). very long baseline interferometry of solar features has been attempted by tapping ( ) as a way of obtaining very high resolution. a fast . ghz receiver was described by jin et al. ( ) and a sensitive ghz whole sun radiometer and polarlmeter have been developed by nakajima et al. ( ). the outer corona is observed by the low frequency image synthesis telescope at clark lake (kundu et al., ) and the magnetic field is probed using faraday rotation of helios transmissions (patzold et al., ). g. helioseismology a new generation of instruments is under development to make observations of solar velocity oscillations. magneto-optic filters that use resonance absorption effects in atomic vapor to reveal doppler shifts have been developed by cacciani and rhodes ( ), appourchaux ( ) and koyama ( ). a laser-stabilized, electrically-tunable, fabry-perot filter has been developed by rust et al. ( ). a birefringent filter is used by libbrecht and zirin ( ) to observe degrees > . glenar et al. ( ) have shown that laser heterodyne spectroscopy is an effective way of measuring solar velocity oscillations. michelson interferometers continue to be developed for doppler shift measurements (gong, ). a classical spectrograph has been equipped with three linear diode arrays to obtain phase shift measurements of oscillations (staiger, ). equipment to measure solar intensity oscillations has been described by duvall et al. ( ), didkovsky and kotov ( ), kotov et al. ( ), didkovsky at https://www.cambridge.org/core/terms. https://doi.org/ . /s x downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core terms of use, available https://www.cambridge.org/core/terms https://doi.org/ . /s x https://www.cambridge.org/core commission ( ), jimenez et al. ( ) and nishikawa et al. ( ). networks of oscillation observing equipment have been or are being establish- ed at numerous sites around the world to provide continuous observations until suitable space experiments can be conducted. a group from birmingham has stations in operation in australia, the canary islands and hawaii to obtain low-degree observations. the iris project based in nice has placed one low-degree instrument at la silla and plans additional instruments at stanford and tashkent (insu, ). the gong project based in tucson plans to install six instruments for degrees up to about at sites that are currently being investigated. a proto- type instrument is under construction (gong, ). references alissandrakis, c.e., lantos, p. and nicolaides, e.: , solar phys. , appourchaux, t.: , solar phys. , beijing: , publ. beijing astron. obs. , - boreiko, r.t. and clark, t.a.: , astron. astrophys. , brueckner, g.e., bartoe, j.-d.f., cook, j.w., dere, k.p. and socker, d.g.: , adv. space res. , caccianl, a. and rhodes, e.j., jr.: , solar seismology from space, carlson, r.w., ogawa, h.s., phillips, e. and judge, d.l.: , appl. opt. , chapman, g.a., herzog, a.d. and templar, s.: , bull. amer. astron. soc. , crommelynck, d.a., brusa, r.w. and domingo, v.: , solar phys. , damfi, l., foing, b.h., martic, m., bruner, m., brown, w., decaudin, m. and bonnet, r.m.: , adv. space res. , degiacomi, c.g., kneubuhl, f.k. and huguenin, d.: , astrophys. j. , dicke, r.h., kuhn, j.r. and libbrecht, k.g.: , astrophys. j. , didkovsky, l.v.: , izv. krym. astrofiz. obs. , didkovsky, l.v. and kotov, v.a.: , izv. krym. astrofiz. obs. , didkovsky, l.v., kozhevatov, i.e. and stepanyan, n.n.: , izv. krym. astrofiz. obs. , duvall, r.l., jr., harvey, j.w. and pomerantz, m.a.: , nature , epstein, g.l., thomas, r.j. and neupert, w.m.: , bull. amer. astron. soc. , in press glenar, d.a., deming, d., espenak, f., kostiuk, t. and mumma, m.: , appl. opt. , golub, l., nystrom, g., spiller, e. and wilczynski, j.: , spie , gong: , newsletter no. , national solar obs., tucson graves, b., foukal, p., rieke, m. and fowler, a.: , bull. amer. astron. soc. , in press henson, g.d. and kemp, j.c.: , solar phys. , herzog, a.d., chapman, g.a., lawrence j.k. and templer, s.: , bull. amer. astron. soc. , hirayama, t., hamana, s. and mizugaki, k.: a, solar phys. , hirayama, r, tanaka, k., watanabe, t., akita, k., sakurai, t. and nishi, k. : b, solar phys. , hurford, g.j., read, r.b. and zirin h.: , solar phys. , insu: , ann. rep. inst. nat. sciences de l'universe (cnrs) - , jimenez, a., pallg, p.l., roca cortes, t. and domingo, v.: , astron. astrophys. , jin, s.-z., zhao, r.-y. and fu, q.-j.: , solar phys. , jones, h.p.: , bull. amer. astron. soc. , in press kotov, v.a., koutchmy, s., kononovlch, e.v., ryzhykova, n.n. and tsap, t.t.: , izv. krym. astrofiz. obs. , koyama, k. : , "hydrodynamic and magnetohydrodynamic problems in the sun and stars", y. osaki (ed.), univ. tokyo, p. at https://www.cambridge.org/core/terms. https://doi.org/ . /s x downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core terms of use, available https://www.cambridge.org/core/terms https://doi.org/ . /s x https://www.cambridge.org/core radiation & structure of the solar atmosphere kundu, m.r., gergely, t.e., schmahl, e.j., szabo, a., loiacono, r., wang, z. and howard, r.a.: , solar phys. , ubs, d., neckel, h., simon, p.c. and thuiller, g.: , solar phys. , laville conde, a., von der liihe, . and radick, r.r.: , bull. amer. astron. soc. , libbrecht, k.g. and zirin, j.: , astrophys. j. , lindsay, c. and kaminksy, c.: , astrophys. j. , l lindsay, c , becklin, e.e., jefferies, j.t., orrall, f.q. and werner, m.w.: , astrophys. j. , markov, v.s.: , issled. geomag. aehron. fiz. solntsa , mlzutani, k., maihara, t., hiromoto, n. and takami, h.: , nature , nakajima, h., sekiguchi, h., sawa, m., kai, k. kawashima, s., kosugi, r., shibya, n., shinohara, n. and shiomi, y.: , publ. astron. soc. japan , nishikawa, j., hamana, s., mizugaki, k. and hirayama, r.: , pub. astron. soc. japan , palmer, j.m., brad, l.g., perry, d.l. and wolfe, w.l.: , j. opt. soc. amer. , parkinson, j.h. and gabriel, a.h.: , adv. space res. , patzold, m., bird, m.k., volland, h., levy, g.f.s., seidel, b.l. and stelzried, c.t.: , solar phys. , rosch, j. and yerle, r.: , esa sp- , rust, d.m., burton, c. and leistner, a.j.: , spie , samain, d. and lemaire, p.: , astrophys. space sci. , shcherbakova, z.a., shcherbakov, a.g. and shapiro, b.i.: , pis'ma astron. zh. , seykora, e.: , solar phys. , smolkov, g.y., pistolkors, a.a., treskov, t.a., krissinel, b.b., putilov, v.a. and potapov, n.n.: , astrophys. space sci. , staiger, j.: , astron. astrophys. , tapping, k.f.: , solar phys. , title, a.m. et al.: , adv. space res. , van hoosier, m.e., bartoe, j.-d.f., brueckner, g.e. and prinz, d.k.: , bull. amer. astron. soc. , wang, c.-j., lu, j., ni, z.-r., you, j.-q. and fan, c.-j.: , acta astron. sinica , wang, j., zirin, h. and shi, z.: , solar phys. , white, s.m., kundu, m.r. and szabo, a.: , solar phys. , zou, h.-c.: , acta astron. sinica , at https://www.cambridge.org/core/terms. https://doi.org/ . /s x downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core terms of use, available https://www.cambridge.org/core/terms https://doi.org/ . /s x https://www.cambridge.org/core _ _ _article .. obituary john a. ‘jack’ smith, md (aug. , –march , ) richard b. gunderman & mervyn d. cohen received: june /accepted: june /published online: august # springer-verlag berlin heidelberg john arthur “jack” smith, long-time pediatric radiologist at riley hospital for children and professor of radiology at indiana university school of medicine, died on march , , at the age of years. he grew up in cincinnati, oh, graduating from walnut hills high school in . when he was young he wanted to be a minister, but the example of nobel laureate and medical missionary albert schweitzer led him to pursue a medical career. jack attended miami university of ohio, where he became the first black resident assistant. he then attended the univer- sity of cincinnati college of medicine, graduating with honors in and becoming the first black in the history of the school to be elected to membership in alpha omega al- pha honor medical society. he completed a rotating in- ternship at the university of iowa, then returned to cin- cinnati to complete his radiology residency, serving as chief resident with dr. ben felson. in jack was drafted and served at the u.s. air force hospital in ankara, turkey, as the chief of radiology for the middle east. after completing his tour of duty, he took a fellowship in pediatric radiology with dr. herb kaufman at the kinderspital in basel, switzerland. this led him to a career in academic medicine, first at downstate hospital in brooklyn, ny, and then beginning in as assistant professor at riley hospital and indiana university. at indiana, jack progressed rapidly up the ranks, becoming associate professor in , professor in , and director of pediatric radiology in . he was the first black tenured professor at the indiana university school of medicine. be- fore he retired from indiana in , he completed two sab- baticals, one at boston children’s hospital in and the other at university hospital of the west indies in kingston, jamaica, in . author and co-author of more than refereed arti- cles, jack’s scientific presentation on the mri appear- ance of blood and blood products, co-authored with mervyn cohen and david cory, received the caffey award of the society for pediatric radiology in . his scientific exhibits received certificates of merit from * richard b. gunderman rbgunder@iu.edu department of radiology, riley hospital for children, indiana university medical center, barnhill drive, indianapolis, in - , usa pediatr radiol ( ) : – doi . /s - - - http://crossmark.crossref.org/dialog/?doi= . /s - - - &domain=pdf the american roentgen ray society in , and , and cum laude citations from the radiological society of north america in and . active in professional service, jack was president of the hoosier state medical society in . he served as an ex- aminer for the american board of radiology from to and as an examiner for the certificate of added qualifi- cation in pediatric radiology in . he was also very active in community service. he served on the boards of the sickle cell anemia foundation, the boys club and the ensemble music society, and he served as president of the national junior tennis league. jack enjoyed driving sports cars, listening to jazz music, and making peach and lemon meringue pies. a big sports fan, he followed racing, tennis, golf and skiing especially closely, and he was for decades a season ticket holder for the indiana university basketball team. he was also a world traveler and was known among friends and family as “mapquest before there was mapquest,” possessing an eidetic memory for every place he had ever visited. jack was a people person. he was genuinely curious about others and made friends with people from every walk of life. a natural networker, he helped to create many connections, both professional and personal, between people. he seemed to know everyone, including their kids’ names and their dogs’ names. he was also remarkably easy-going and seemed to never get flustered, which helped to build a friendly, cohesive radiology department. back in the days when people took lunch breaks, jack could regularly be found in the lunchroom with a group of technologists, catching up on the soap operas “days of our lives” and “the young and the restless.” he was the person technologists and other staff came to with personal problems, because everyone knew that he was genuinely interested in them, would give them his full attention for as long as it took, and would offer wise words of counsel. jack will be missed by all who knew him. he is survived by his wife, donna; his four daughters, ann, janis, gwen, and jill; and five grandchildren, anna, river, jonathan, jackson, and julian. pediatr radiol ( ) : – 	john a. ‘jack’ smith, md (aug.� , –march� , ) wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ document resume ed so author madison, james h. title extending liberty westward: the northwest ordinance of . pub date may note p.; for related documents, see so and so . pub type historical materials ( ) edrs price mf /pc plus postage. descriptors civil rights; governance; government (administrative body); land settlement; slavery; *social studies; state government; *state history; united states history identifiers bicentennial; *indiana; *northwest ordinance ; northwest territories abstract the northwest ordinance of , which proviled for government of the largely unsettled frontier area north of the ohio river and for an orderly, three-stage transition of the territories from control by national government to full and equal statehood, joins the declaration of independence and the u.s. constitution to form a trinity of founding-era documents that charted the course of a new nation. the northwest ordinance aroused conflict in indiana and more broadly among u.s. citizens over slavery and sectionalism, states' rights, representative government, western expansion, and individual freedom. the indiana territory was created in , and its first governor was william henry harrison. many of harrison's fellow settlers resented their lack of a voice in government and pushed hard for transition to the seconl, semi-representative stage as promised in the ordinance. the force of numbers pushed harrison to agree to movement to the second stage. the pattern of conflict was repeated in moving to the third stage although the conflict was more intense and enduring. a population census in counted , hoosiers, more than enough to meet the requirements of the northwest ordinance for statehood. in , men wrote a constitution for indiana, and on december , , president james madison approved statehood. the indiana constitution of ended the controversy over slavery by making indiana a free state. (sm) ***********************************************t*********************** * reproductions supplied by edrs are the best that can be made * * from the original locument. * ************************************** ************* .**************** j. madison/may' - cy% extending liberty westward: cr the northwest ordinance of by james h. madison james h. madison, ma' , phd' , is associate professor of history at indiana university bloomington, editor of the indiana magazine of history, and author of the recently published book, the indiana way: a state history. currently he is working on a biography of eli lilly. "permission to reproduce this materiai has been granted by tclyvn s h"' y to the educational resources information center (eric)." s dcrartment of education . ov.ce duce'.onal gesearch and improvement bc: eoucat:onal resources information r center (eric) relhis document ,s been reproduced as received irom the purson or organization jriginattrg it miner changes have been made to improve reoroduction qua lity (. ) points of view or ogynions stater' m !hie docu-ment do not necessaniy represent of oeri pothhon or policy j. madison/may' page the birth was not promising. passed on friday, the th of july, by a nation- al government that was moribund, the northwest ordinance of nonethe- less became a fundamental doom', lt in american history. indeed, it belongs to the near-holy trinity of founding- era documents, joining the declaraticn of independence and the constitution of in charting the course of the new nation. for the frontier west es- pecially, including pioneer indiana, the northwest ordinance determined the nature of government and ensured that basic american freedoms extende west- ward. a vision for self-,overnment the immediate purpose of the north- west ordinance was to provide for gov- ernment in the largely unsettled lands west of the appalachian mountains and north of the ohio river. while there was no strong sentiment among american a j. madison/may' page leaders to maintain the west in per- petual colonial dependency, neither was there confidence that this sparse- ly populated, undeveloped land was yet suited for representative government and statehood. indeed, some conserva- tive easterners feared '-hat the sort of people most likely to move west were those least "civilized" and least capable of self-government. the ordi- nance suggested a pragmatic compromise that allowed for a transition from full control by the national govern- ment to gradual representative govern- ment and eventual statehood. it there- by posited a developmental model, one that planned for evolutionary growth toward self-government. this transition would occur in three stages, carefully delinated in the ordinance. in the least democrat- ic, first stage a territorial governor and judges appointed by the national government would rule. when the popu- lation reached , free, adult j. madison/may, page males, the territorial government would move to the second stage in which those men who owned acres of land could elect representatives to a territorial legislature. when the total free population reached , inhabitants the territory could claim statehood and enter the union on an equal footing with the original states. by this three-step process five states of the northwest territory entered the union: ohio ( ), indi- ana ( ), illinois ( ), michigan ( ), and wisconsin ( ). so suc- cessful was this visionary prescrip- tion for development that the model was used to create states across the great plains to the pacific ocean. william henry harrison: frontier ruler the success of the northwest ordi- nance over the long term should not hide the immediate conflicts it gener- ated, for in these struggles were re- j. madison/may' page fleeted some of the basic challenges of american political life, particu- larly the tensions between liberty and order. the application of the ordi- nance to indiana's early history illustrates the nature of the debate. the indiana territory was created in , as ohio was becoming a state. initially its boundf,ries extended west to the mississippi river and north to the canadian border. its first gover- nor was william henry harrison. only when he assumed office, harrison was a virginia gentleman who had served on the frontier as early as when he fought beside general anthony wayne at the battle of fallen timbers. harrison continued as an in- dian fighter, most notably at the bat- tle of tippecanoe in . even more significantly, though less bloodily, he served as a negotiator of indian land cession treatief by which much of indiana passed from native to white occupancy. in addition to indians, j. madison/may' page harrison's foes came to include many of his fellow settlers, most of whom approved his aggressive indian policy but resented his undemocratic leader- ship. the nearly , pioneers who re- sided in indiana territory in had not elected harrison as their gover- nor. as stipulated in the northwest ordinance, he was appointed by presi- dent john adams. and his word was law. resenting their lack of a voice in government, indianans pushed hard for transition to the second, semi-repre- sentative stage as promised in the or- dinance. harrison resisted for a time, but the force of numbers, guaranteed in , pushed him to agree to move- ment to the second stage. indiana's first elections were held in , and the first general assembly met at vincennes that same year. the pattern was repeated in moving to the third stage, although the con- flict was more intense and enduring. j. m-dison/may' page governor harrison still retained large power after . he could, for exam- ple, exercise an absolute veto over any territorial legislation. opposition grew as newcomers set- tled in indiana. the whitewater valley in the southeast became a hotbed of anti-harrison feelings. accustomed to self-rule in north carolina, ohio, or pennsylvania, these westward moving pioneers demanded more democratic gov- ernment. they compared their situation to that of the revolutionary patriots, complaining in an petition to washington that their indiana govern- ment had a "monarchal shape." harrison and the national government responded with concessions, but the self-styled democratic opposition remained unsat- isfied. the solution was obvious. only the statehood promised in the ordi- nance of would quiet the discon- tent. hoosiers eager for statehood j. madison/may' page the leader of indiana's pro-state- hood faction was jonathan jennings. a newcomer from pennsylvania, jennings gathered around him men opposed to harrison, men eager for statehood, and, in some cases, men hungry also for office for themselves. harrison and his followers resisted the push for statehood, arguing that indiana taxpayers could not afford tha costs of state government. moreover, one of harrison's followers asserted, there was in indiana "a great scarcety (sic) of talents, or men of such information as are necessary to fill the respec- tive stations, & offices of govern- ment." the debate over statehood was not settled by measuring either wealth or talent. all that counted was popula- tion. with the defeat of the indians and the british in the war of , pioneer families poured into indiana territory, taking up rich, abundant land and bringing with them fundamen- j. madison/may' page tal attachments to representive gov- ernment. a population census in counted , hoosiers, more than enough to meet the requirements of the northwest ordinance. the nineteenth state in the heat of a corydon summer in , men wrote a constitution for indiana. the majority of the elected delegates had been pro-statehood men and supporters of jennings, whom they elected to chair the convention. a critic of the jennings faction labeled them "empty babblers, democratic to madness," but they accomplished what they set out to do. with the experi- ence of territorial government fresh in their minds, jennings and his asso- ciates created a government in which the elected legislature was the strongest branch, closely dependent on grassroots approval. the powers of the state governor were severely re- j. madison/may' page stricted; he would enjoy little of the gubernatorial reach possessed by harrison. on december , , president james madison approved statehood. in- diana now had a full voice in the na- tional government, including the vote of two senators in washington, equal to that of virginia, massachusetts, and the other original states. resent- ment of quasi-colonial status that may have lingered from the years before soon washed away in the waves of state and national pride that burst across indiana. the commitment made in had been fulfilled. slavery is prohibited the promise of statehood was .he most important promise made in , but there were other guarantees in the northwest ordinance. these were con- tained i a section called the arti- cles of compact, which pledged that j. madison/may' page from the beginning settlers in the west would enjoy certain fundamental freedoms and rights, many of which the new government would soon include in the bill of rights. the articles of compact guaranteed religious freedom and due process of law, including trial by jury, and it urged the en- couragement of public education and justice toward indians. the sixth and final article stipulated that "there shall be neither slavery nor involun- tary servitude in the said territory." the articles reassured pioneers that they would take american freedoms along west with their rifles, spinning wheels, and axes. the importance of the articles of compact is indicated in the fact that indiana's constitu- tion of repeated and expanded these liberties guaranteed in . the prohibition of slavery caused the most controversy in indiana. gov- ernor harrison owned slaves and so did many of his prominent supporters. the j. madison/may' page census of counted slaves and free blacke the territory. it is likely that _.:e lives of many of those blacks countea as free differed little if at all from those listed as slaves. harrison's opponents soonadded jlevery to their growing list of objections to the governor's rule. the south's peculiar institution, they asserted, was "repugnant to the ines- timable principles of a republican government." jennings and his sup- por* grs castigated the harrison faction as a slaveholding aristocracy that was determined to deny represen- tative government and democracy on in- diana's frontier. this argument, com- bining antislavery with 'reedom, would ring loud in the nation's politics down to appcnattox courthouse. in in- diana the debate closed with the con- stitution fo , which sealed the fate of slavery in the new free-soil state. j. madison/may' an enduring document page two hundred years later the north- west ordinance remains a fundamental document. not only did it shape the history of westward settlement andstatemaking, but it enunciated princi- ples of broad and enduring signifi- cance. the northwest ordinance aroused conflict, immediately in indiana in the struggles between the harrison and jennings factions, and more broadly in the controversies among americans over slavery and sectionalism, states' rights, representative government, western expansion, and individual freedom. ultimately, however, it was among america's great achievements. today we rightly celebrate that achievement. the document promulgated one of the most generous colonial policies in the history of nations, a policy j. madison/may' page dedicated to creating equal rather than dependent units of government. in this purpose it was radically differ- ent from british colonial policy and that of most empires. it was remark- ably optimistic in assuming that the nation would expand westward, and itencouraged that expansion by guaran- teeing that liberty and representative government would move across the moun- tains with the pioneers. at the same time, this expansion would be an or- derly process, one in which the na- tional government would remain a re- sponsible party in allowing time for the growth and maturation necessary for representative government. the framers of the northwest ordi- nance tied the growing west to the new nation, not with cords of imperial power but with sentiments of recipro- cal, heartfelt nationalism. from the early th century to the present, the people of indiana and of th. . midwest have often perceived themselves as the ' j. madison/may' page most patriotic and the most typical cf americans. such sentiments originate in part with a document approved on july , . the hoosier vocal emotions corpus: a validated set of north american english pseudo-words for evaluating emotion processing the hoosier vocal emotions corpus: a validated set of north american english pseudo-words for evaluating emotion processing isabelle darcy & nathalie m. g. fontaine # the psychonomic society, inc. abstract this article presents the development of the “hoosier vocal emotions corpus,” a stimulus set of recorded pseudo-words based on the pronunciation rules of english. the corpus contains controlled audio pseudo-words uttered by two actresses in five different emotions (i.e., happiness, sadness, fear, anger, and disgust) and in a neutral tone, yielding , audio files. in this article, we describe the corpus as well as a validation study of the pseudo-words. a total of native english speakers completed a forced choice emotion identification task. all emotions were recognized better than chance overall, with substantial variability among the different tokens. all of the recordings, including the ambiguous stimuli, are made freely available, and the recognition rates and the full confusion matrices for each stimulus are provided in order to assist researchers and clinicians in the selection of stimuli. the corpus has unique characteristics that can be useful for experimental paradigms that require controlled stimuli (e.g., electroencephalographic or fmri studies). stimuli from this corpus could be used by researchers and clinicians to answer a variety of questions, including investigations of emotion processing in individuals with certain temperamental or behavioral characteristics associated with difficulties in emotion recognition (e.g., individuals with psychopathic traits); in bilingual indi- viduals or nonnative english speakers; in patients with aphasia, schizophrenia, or other mental health disorders (e.g., depression); or in training automatic emotion recognition algorithms. the hoosier vocal emotions corpus is available at https:// psycholinguistics.indiana.edu/hoosiervocalemotions.htm. keywords vocal emotions . forced choice identification . emotion perception . speech corpus . validation . english . pseudo-words . emotion stimulus set the ability to process salient emotional and social cues is critical for adaptive behavior. a failure to process expressions of emotion adequately can have important negative and long- term effects on social behavior and can be a risk factor for adaptation problems, including aggressive and antisocial be- havior (herba & phillips, ). the majority of studies on emotion processing have focused on facial expressions of emotion (e.g., pollak & sinha, ; tottenham et al., ). there is less research on vocal expressions of emotion, notably because of the difficulty in obtaining naturalistic re- cordings of vocal expressions of specific emotions (scherer, banse, wallbott, & goldbeck, ). still, vocal cues play an important role in the expression of emotions. by “vocal,” we refer to “everything that remains present in a spoken message after lexical and syntactic information has been removed” (van bezooijen, , p. ). a growing number of studies conducted in the past decade have indicated that humans, across languages and cultures, can infer emotion from vocal expression alone because of differential acoustic patterns (e.g., banse & scherer, ; bänziger, mortillaro, & scherer, ; castro & lima, ; juslin & laukka, ; liu & pell, ; livingstone & russo, ; pell, paulmann, dara, alasseri, & kotz, ; sauter, eisner, ekman, & scott, ; scherer, banse, & wallbott, ). a number of emotion corpora have been produced (see scherer, clarke-polner, & mortillaro, ; ververidis & kotropoulos, , for reviews). they all have their particular features and are composed of diverse vocal stimuli. table presents a sample of data collections of vocal expressions of emotion. we developed and validated a set of pseudo-words based on the phonology and pronunciation rules of north american english, which we aim to make available to the research and clinical communities. the corpus, named the hoosier vocal emotions corpus (hvec), includes important unique * isabelle darcy idarcy@indiana.edu indiana university, bloomington, in, usa university of montreal, montreal, quebec, canada behavior research methods https://doi.org/ . /s - - - ( ) : – published online: september http://crossmark.crossref.org/dialog/?doi= . /s - - - &domain=pdf https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm mailto:idarcy@indiana.edu ta b le s am pl e o f d at a co ll ec ti o ns o f v o ca l ex p re ss io n s of em o ti o n r ef er en ce s n am e/ d es cr ip ti on o f th e d at a co ll ec ti o n l an g u ag e s p ea k er s t y p e an d n u m b er o f v o ca l st im u li k in d o f sp ee ch e m ot io n s (i n te rm s u se d b y au th o rs ) o th er p er ce p tu al m o d al it ie s b än zi g er et al . ( ) g en ev a m ul ti m o da l e m o ti o n p or tr ay al s c o re s et (g e m e p -c s ) n on la ng u ag e (p se ud o -s pe ec h se nt en ce s an d a no n ve rb al v o ca li za ti o n ; “a aa ” b y f re nc h sp ea ke rs ) w om en an d m en (p ro fe ss io na l f re nc h -s pe ak in g th ea te r ac to rs ) em ot io n ex pr es si o ns (p se ud o - sp ee ch se n te nc es ) a ct ed sp ee ch em ot io n s (e .g ., am us em en t, de sp ai r, ho t an g er , fe ar /p an ic , jo y /e la ti on , sa dn es s, co nt em pt , di sg us t, su rp ri se ) v id eo (i .e ., pr es en ta ti on of dy n am ic pi ct ur e w it ho u t so u nd ) an d au d io – vi d eo (i .e ., pr es en ta ti on of dy n am ic pi ct u re an d so un d) b el in , f il li on -b il od ea u, an d g o ss el in ( ) m on tr ea l a ff ec ti ve v oi ce s (m a v ) n on ve rb al af fe ct bu rs ts us in g th e f re nc h v ow el “a h” d if fe re nt ac to rs ( w om en an d m en ) n on v er ba l af fe ct b ur st s a ct ed sp ee ch a ng er , d is gu st , p ai n, sa dn es s, su rp ri se , ha pp in es s, p le as u re an d ne ut ra l — b ur k ha rd t, p ae sc hk e, r ol fe s, s en dl m ei er , an d w ei ss ( ) b er li n e m o ti o n al s pe ec h d at ab as e (e m o -d b ) g er m an w om en an d m en m ea ni n gf ul se nt en ce s by em ot io ns (p lu s th e ne ut ra l st at e) by ac to rs , in ad di ti on to so m e se co nd ve rs io n s (n = ab ou t se n te n ce s) a ct ed sp ee ch a ng er , fe ar , jo y, sa d ne ss , di sg us t, bo re d om an d ne ut ra l — c as tr o an d l im a ( ) s et of p o rt u gu es e se n te n ce s an d p se ud o se nt en ce s e ur op ea n p or tu gu es e w om en p or tu gu es e se n te nc es an d ps eu d os en te nc es by em ot io ns (p lu s th e ne ut ra l st at e) m ea n le ng th = sy ll ab le s (r an ge – ) a ct ed sp ee ch h ap pi ne ss , sa dn es s, an ge r, fe ar , di sg us t, su rp ri se an d ne ut ra l — c os ta nt in i, ia da ro la , p ao lo ni , an d t o di sc o ( ) e m o v o c or pu s it al ia n ac to rs ( w o m en an d m en ) se n te n ce s b y em ot io ns (p lu s th e ne ut ra l st at e) by ac to rs ( se nt en ce s) a ct ed sp ee ch d is gu st , jo y, fe ar , an ge r, su rp ri se , sa dn es s an d ne u tr al — l au kk a et al . ( ) v oc al e x pr es si on s of n in et ee n e m ot io ns ac ro ss c ul tu re s (v e n e c ) e ng li sh pr of es si on al ac to rs fr o m e ng li sh sp ea k in g cu lt ur es (u s a , in di a, k en ya , s in ga po re an d a us tr al ia ) ( % w o m en ) a bo ut , vo ca l ex pr es si on s (m ai nl y sh or t ph ra se s w it h em ot io na ll y ne ut ra l co nt en t, ex p re ss ed in th re e le v el s o f in te n si ty ) a ct ed sp ee ch em ot io n s (e .g ., am us em en t, an ge r, co nt em pt , di sg us t, di st re ss , fe ar , gu il t, h ap pi ne ss , sh am e) an d ne ut ra l — l iu an d p el l ( ) a d at ab as e of c hi ne se v oc al em o ti on al st im ul i p se ud o -s en te nc es (s em an ti ca ll y m ea ni n gl es s an d re la ti ve ly p la u si b le as c hi n es e se nt en ce s) n at iv e m an da ri n sp ea ke rs ( w om en an d m en ) p se ud o -s en te nc es by em ot io ns (p lu s th e ne ut ra l st at e) a ct ed sp ee ch a ng er , d is gu st , fe ar , sa d ne ss , ha pp in es s, pl ea sa n t su rp ri se an d ne ut ra l — l im a, c as tr o , an d s co tt ( ) a co rp us of n on v er ba l vo ca li za ti on s n on ve rb al v oc al iz at io n s by e u ro p ea n p or tu gu es e n at iv e sp ea ke rs sp ea ke rs ( w om en an d m en ) w ho d id so u nd s (n o gu id an ce w as pr ov id ed as to th e sp ec if ic a ct ed sp ee ch po si ti ve st at es (a ch ie ve m en t/ tr iu m ph , am us em en t, se n su al pl ea su re — behav res ( ) : – t ab le (c o n ti n u ed ) r ef er en ce s n am e/ d es cr ip ti on o f th e d at a co ll ec ti o n l an g u ag e s p ea k er s t yp e an d n u m b er of v o ca l st im u li k in d o f sp ee ch e m o ti o n s (i n te rm s u se d b y au th o rs ) o th er p er ce pt ua l m o d al it ie s n ot h av e fo rm al ac ti ng tr ai ni ng . k in d of so un ds th e sp ea ke rs ha d to m ak e) an d re li ef ) an d ne ga ti ve st at es (a n ge r, di sg us t, fe ar an d sa dn es s) l iv in g st o ne an d r u ss o ( ) t he r ye rs on a u di o - v is ua l d at ab as e of e m ot io na l s pe ec h an d s on g (r a v d e s s ) e ng li sh n or th a m er ic an e n gl is h- sp ea k in g p ro fe ss io na l ac to rs ( w om en an d m en ) e ng li sh se nt en ce s (t o ta l o f , re co rd in g s) a ct ed sp ee ch an d so ng s pe ec h: ca lm ,h ap py ,s ad ,a ng ry , fe ar fu l, su rp ri se an d di sg us t s on g : ca lm , ha pp y, sa d, an gr y an d fe ar fu l e ac h ex p re ss io n w as pr od uc ed at tw o le ve ls of em ot io na l in te ns it y w it h an ad di ti on al n eu tr al ex pr es si on . f ac e an d v oi ce , fa ce on ly p ar so ns , y o un g , c ra sk e, s te in , an d k ri ng el ba ch ( ) o xf or d v oc al s ou nd s da ta ba se (o xv oc ) n on v er ba l so u nd s in fa n t vo ca li za ti on s ( gi rl s an d b oy s) a d ul t v oc al iz at io n s ( cl ip s by w o m en on ly fo r di st re ss vo ca li za ti on s, w o m en an d m en fo r la ug h te r vo ca li za ti on s an d w om en an d m en fo r n eu tr al vo ca li za ti on s) a n im al vo ca li za ti on s (p et ca ts an d do gs ) t ot al of st im ul i in fa nt s: cr y vo ca li za ti on s (n = ) ; la u gh te r v oc al iz at io ns (n = ); n eu tr al ba bb le s (n = ) a du lt s: di st re ss vo ca li za ti on s (n = ) ; la u gh te r (n = ); n eu tr al (n = ) a n im al s: d is tr es s (n = ) in fa nt s: so un ds fr o m vi d eo re co rd in gs o f in fa nt s fi lm ed in th ei r o w n h om es a du lt s an d an im al s: so un ds fo un d fr o m o n li n e re so ur ce s h ap py (l au gh te r v oc al iz at io ns ), sa d (c ry an d d is tr es s v oc al iz at io n s) an d ne ut ra l — r ig ou lo t, w as si li w iz k y, an d p el l ( ) d at ab as e of em ot io na ll y in fl ec te d ps eu d o- ut te ra nc es p se ud o- ut te ra nc es by na ti ve sp ea ke rs o f c an ad ia n e n gl is h sp ea k er s ( w o m en an d m en ) p se ud o - ut te ra n ce s ( sy ll ab le s in le n g th ) a ct ed sp ee ch a ng er , di sg us t, fe ar , ha pp in es s, sa dn es s, an d n eu tr al — w en dt et al . ( ); w en dt an d s ch ei ch ( ) m ag de bu rg er p ro so di e- k or pu s g er m an ac to rs (w om an an d m an ) l in gu is ti ca ll y m ea ni ng fu l w o rd s (n > , ) an d d is y ll ab ic ps eu d o- w o rd s (n = ) a ct ed sp ee ch a ng er , di sg us t, fe ar , ha pp in es s, sa dn es s an d ne ut ra l — behav res ( ) : – characteristics. first, it focuses on disyllabic pseudo- words, rather than meaningful words or sentences, to remove the semantic meaning and allow for the speech prosody to become the central attribute of emotion pro- cessing (wendt et al., ; wendt & scheich, ). to our knowledge, only one other corpus (the magdeburger prosodie korpus, a set of stimuli respect- ing the phonotactic and phonetic rules of the german language) includes isolated pseudo-words (wendt et al., ; wendt & scheich, ). our corpus’s main fea- tures are based on this german corpus. other corpora of vocal emotions contain pseudo-sentences (e.g., castro & lima, ; liu & pell, ). however, experimental paradigms can require shorter stimuli, which would be difficult to manually extract from sentences and subsequently validate separately. in addition, rigoulot et al. ( ) demonstrated in a gating paradigm study that the length of the stimuli matters for the time course of emotion recognition, and that full sentences are rec- ognized much more easily than truncated ones. other corpora use affect bursts (e.g., “ah”) or emotional sounds such as screams or laughter (e.g., belin et al., ; parsons et al., ). despite the high effectivity of such stimuli to convey specific emotions, they are also not necessarily suitable for experimental paradigms requiring controlled stimuli with medium or normal emotional intensity. the hoosier vocal emotions corpus includes con- trolled audio pseudo-words, uttered twice apiece by two actresses in five different positive or negative emotions (i.e., happiness, sadness, fear, anger, and disgust) and in a neutral tone, yielding , stimuli (some of the stimuli were pronounced more than two times). we selected the emotions on the basis of the basic emotions identified by ekman ( ), except for surprise, because this emotion can have any valence (it can be neutral, positive or neg- ative). in addition, surprise utterances can be difficult to simulate experimentally (pell et al., ). although con- cerns have been raised about the use of acted rather than natural stimuli (bachorowski & owren, ), there are also arguments suggesting that actors can produce realis- tic portrayals and valid instances of vocal expressions of emotion (ververidis & kotropoulos, ). one important argument is that much of our verbal communication is subject to sociocultural censure and involves making im- pressions on others (bachorowski & owren, ; banse & scherer, ). therefore, having people utter an emo- tion as if they were experiencing it may not be signifi- cantly different from a real-life communicative situation. two female voices were preferred over having one male and one female voice, mainly for reasons of comparability and homogeneity between such acoustic dimensions as pitch range, and to facilitate their use in experimental paradigms requiring tight control of the acoustic param- eters of stimuli, such as event-related potential (erp) studies. in this article, we describe the structure of the hoosier vocal emotions corpus, as well as the valida- tion of the pseudo-words in terms of the emotion they portray. we also discuss potential applications of this set of stimuli. method creation of the stimuli the stimulus set is composed of pseudo-words based on real english words. these pseudo-words were created by selecting common english disyllabic words using the cobuild frequency information (per million) from the celex english wordforms database (baayen, piepenbrock, & gulikers, ), and manipulating the order of segments within the word (see wendt & scheich, , or castro & lima, , for a similar procedure). for example, the pseudo-word “elby” was constructed from the noun belly. as a result, there is no clear phonetic relationship between the pseudo- words and their originals, but they are matched in terms of number of syllable and phonemes. care was taken to ensure that the pseudo-words were phonotactically legal—that is, that the sequences of phonemes were permitted and easily pronounceable in english. similarly, slight phonetic adjustments were made to comply with english pronunciation rules. for example, the pseudo-word “domner,” based on modern, did not retain the flapped /d/ found in the north american english pronunciation of modern, since the flap is not found in word-initial position in english. pseudo-words that were too clearly reminiscent of their original or of other real words were excluded. a final list of pseudo-words was generated (see table ). stress al- ways fell on the first syllable, but the vowel in the second syllable was not always fully reduced (indicated by the international phonetic alphabet [ipa] symbols in table , where only “schwa” [ə] represents a reduced vowel). the transcriptions provided in table closely reflect the actual pronunciation of most of the stimuli by both actresses. since each actress pronounced a given pseudo-word times ( × emotions), there are es- sentially pronunciations of the same pseudo-word, thus displaying some variation from one token to the next. the transcription here reflects the most common pronunciation of the stimuli, and there might be some variation across specific stimuli, especially in terms of the vowels. table is provided here to give further guidance to researchers about the possible variations in behav res ( ) : – pronunciation for the same pseudo-word, but we encour- age researchers and clinicians who need an exact con- trol of sound properties to check each stimulus they plan to use. elicitation and recording procedures two actresses were recruited to record the pseudo- words in a neutral tone as well as in five different modal emotions: happiness, sadness, fear, anger, and disgust. female voices were recorded as the basis of another ex- periment (i.e., an electroencephalography [eeg] paradigm involving young children; hoyniak et al., ). both actresses were native speakers of midwestern united states english (north midland dialect region; clopper & pisoni, ), and had lived exclusively in that region prior to the recording. they reported no fluency in any language other than english and have not lived abroad. table list of the pseudo-words included in the corpus, in the roman alphabet and in ipa transcription item number orthographic representation ipa transcription item number orthographic representation ipa transcription nervack /’nɜɹvæk/ vigging /’vɪgɪŋ/ lorack /’loɹæk/ voker /’voʊkəɹ/ lairet /’lɛɹət/ vokered /’voʊkəɹd/ vokered /’voʊkəɹd/ volers /’voʊləɹs/ tairack /’tɛɹək/ winnith /’wɪnɪθ/ domner /’dɑmnəɹ/ ziddy /’zɪdi/ nammy /’næmi/ zilard /’zɪləɹd/ tannock /’tænək/ vercoed /’vɜɹkoʊd/ agerth /’ægəɹθ/ forny /’fɔɹni/ armidge /’ɑɹmɪdʒ/ admage /’ædmɪdʒ/ burish /’bʊɹɪʃ/ affning /’ɑfnɪŋ/ dernom /’dɜɹnəm/ elby /’ɛlbi/ revo /’ɹɛvoʊ/ ervy /’ɜɹvi/ fingill /’fɪŋgəl/ infess /’ɪnfɛs/ jouless /’dʒoʊlɛs/ youssle /’jusəl/ lebby /’lɛbi/ kervo /’kɜɹvoʊ/ lowmen /’loʊmən/ kervoed /’kɜɹvoʊd/ madage /’mædədʒ/ larpy /’lɑɹpi/ menno /’mɛnoʊ/ leknodge /’lɛknədʒ/ merrus /’mɛɹəs/ modner /’mɔdnəɹ/ mowan /’moʊwən/ mokers /’moʊkəɹs/ nabick /’næbɪk/ musser /’mʌsəɹ/ nemmy /’nɛmi/ naffing /’næfɪŋ/ nidder /’nɪdəɹ/ nifish /’nɪfɪʃ/ nillen /’nɪlən/ nipher /’nɪfəɹ/ nomel /’nɔməl/ othening /’ɔθ(ə)nɪŋ/ nomey /’noʊmi/ rackies /’ɹækiːz/ ramidge /’ɹæmɪdʒ/ scopies /’skoʊpiːz/ shavil /’ʃævɪl/ shifin /’ʃɪfɪn/ shibur /’ʃɪbəɹ/ vackner /’væknəɹ/ slover /’sloʊvəɹ/ vashil /’væʃɪl/ terrel /’tɛɹəl/ vishal /’vɪʃəl/ thager /’θægəɹ/ wedick /’wɛdɪk/ thomer /’θoʊməɹ/ winthy /’wɪnθi/ valish /’vælɪʃ/ youshing /’juːʃɪŋ/ venner /’vɛnəɹ/ zuber /’zubəɹ/ verney /’vɜɹni/ boldface in the orthographic representations indicates the syllable carrying the main stress behav res ( ) : – they were students in the department of theatre and drama at a large midwestern higher education institution (indiana university, bloomington, in) and were and years old, respectively, at the time of the recordings. both actresses were paid and gave consent to share the recordings in a publicly accessible database. each actress (henceforth, a.g. and k.m.) was record- ed individually in a single session of approximately . to h. the experimenter first briefly explained the gen- eral procedures to each actress, who was also given time to familiarize herself with the list of stimuli. pronunciation of the pseudo-words was clarified as needed. the different emotions were discussed and ex- plained. the stimuli were elicited using a short sentence preceding the pseudo-word: ‘it starts like /word/, i say /pseudo-word/, i say /pseudo-word/ again’ (see table ). this was done to help maintain consistent pronunciation of the pseudo-words and to enhance fluent delivery and more natural sounding speech. in addition, this form of elicitation was chosen to enable a similar delivery con- text for each pseudo-word across emotions and ensure high comparability. each pseudo-word was thus pro- nounced at least twice (two times per carrier sentence). for each actress, at least stimuli were pronounced for each emotion, yielding a total of at least stimuli per actress. however, some stimuli were pronounced more than two times, when an actress chose to reat- tempt the emotion portrayal for a given carrier sentence, resulting in a total of pseudo-words for a.g. and pseudo-words for k.m., for a grand total of , audio files. the stimuli are overall similar in terms of duration (m = ms, median = ms, sd = ms) and intensity (m = . db, median = . db, sd = . db). actresses were allowed to choose the order in which they preferred to utter each emotion. they were then seated in a recording booth, wearing sennheiser hd dynamic stereo headphones, and before record- ing a set were shown a short presentation of pictures and auditory examples of (non-english) pseudo-words spoken in the corresponding emotion (wendt & scheich, ). the pictures depicted situations in which examples of the specific emotion to be uttered were displayed. for example, various clip art pictures of angry individuals, arguing friends and knit eyebrows were shown to illustrate anger, and to clarify a general mood for each emotion. the experimenter demonstrated a few items in their carrier sentences (without modeling a particular emotion), to help with pronunciation of stimuli (fluency) and overall rhythm. the actresses were also encouraged to imagine situations/scenarios accord- ing to the emotion to be expressed. they were given as much time as they needed to “get into the character” of the emotion before proceeding with the recordings. the experimenter also instructed the actresses not to exag- gerate their expressions of the emotions, but to achieve a “normal” rather than a “strong” level of emotional intensity (see livingstone & russo, ). the stimuli were recorded in a noise-isolated record- ing booth, at a sampling rate of hz with -bit resolution on a mono channel, using a sennheiser e dynamic cardioid microphone and an edirol ua usb stereo audio interface. the distance and orientation of the actresses with regard to the microphone were held as constant as possible. each stimulus (pseudo-word) was then manually cut from its sentence context and saved separately in a .wav format for presentation in the subsequent evaluation procedures. we conducted a validation study with approximately participants rating each sound file of the hoosier vocal emotions corpus, to estimate to what extent each recorded stimulus represents an acceptable rendition of the intended emotion. we included stimuli from both actresses into the corpus validation, that is, a total of , audio files. given the large number of audio files, the time required for a single listener to evaluate all of them would have been prohibitively long. we therefore divided the files into four stimuli lists, which were pre- sented to listeners for evaluation. all emotions were equally balanced in each list. however, we decided against mixing the two voices in each list (see castro & lima, , for a similar design). each list contained stimuli from only one speaker (lists and : a.g., lists and , k.m.). this was done in order to reduce comparison between voices, and to enhance the reliance on actual acoustic properties of the stimuli. an addition- al consideration was the cognitive load of this task, which is demanding for participants. each participant rated only one list. the dataset accompanying the cor- pus contains ratings for each audio file from about persons (see below for the method details). all proce- dures were approved by the indiana university institutional review board. validation of the stimuli procedure to validate the stimuli of the corpus, we opted for a forced choice identification task similar to the one used by van bezooijen ( ) or castro and lima ( ). the stimuli were presented to listeners table example of the materials used to elicit the pseudo-words for each emotion /’sloʊvəɹ/ it starts like ‘slow’ i say slover i say slover again /’loɹæk/ it starts like ‘lord’ i say lorack i say lorack again behav res ( ) : – via headphones, using the praat software (version . . ; boersma & weenink, ) on computers run- ning under windows . participants were tested individ- ually and were seated at a computer station in a partitioned computer lab, wearing high-quality sanako over-the-ear headphones at a self-chosen comfortable listening level. their task was to listen to each sound file and identify what emotion they thought the speaker intended to convey. they were asked to choose one out of six possible emotions and indicate their choice by clicking on the correspondingly labeled button on the screen. the labels were “neutral,” “happy,” “sad,” “fear,” “angry,” and “disgust.” there was no “other/ none of the above” option (livingstone & russo, ). participants were also asked to choose how con- fident they were in their choice by clicking on a num- ber, on a scale ranging from (not sure) to (very sure). the instructions were displayed on the screen as follows: this is a judgment experiment about how actors convey emotions. you will hear an actress say non-words and your task is to choose what emotion you think it con- veys. (some non-words might be repeated a few times). please don’t spend too much time on each non-word. try to do it using your intuition. in addition, we ask that you indicate how confident you are with your choice on a scale of (not sure) to (very sure). there are several breaks. if you have questions, please ask now. the buttons appeared as rectangles on a single line in the middle of the screen, and their order was randomly varied across list (but kept constant for any given participant) to avoid preference effects. the task was not timed, and listeners could replay the sound up to eight times by clicking on a repeat button (fig. ). the presentation order of the sound file was randomized for each participant, and the script implemented a break after every stimuli. no stimulus file was repeated. the average duration of the identification task was about min. as ex- plained above, the sound files were divided into four lists to keep the duration manageable for a single participant. each of the four lists contained roughly the same number of stimuli: lists and (a.g.) each contained sound files, list contained sound files, and list contained sound files (k.m.). participants were randomly assigned to one list upon arrival in the testing room. all participants also filled out a sociodemographic questionnaire (notably to assess their age, sex, and languages spoken) administered through the qualtrics survey software. participants in all, participants were tested. the test- ing took place between february and december . six participants were excluded for various reasons (not native speakers of english or did not grow up in the united states, multiple neurocognitive issues report- ed, incomplete dataset, technical failure, or more than twice the average time needed to complete the task). in total, data from participants ( % female), who were between and years old (m = . , sd = . ), were included in the analysis (list , n = ; list , n = ; list , n = ; list , n = ). most of the participants were college students, and they were predominantly caucasian. only one participant reported not knowing any language other than english. twelve of the participants reported growing up bilingually using english and another language. about half of the partic- ipants ( . %) reported knowledge of spanish, . % of french, and . % of german, with other lan- guages mentioned by fewer than % of the participants (e.g., japanese, . %). a total of . % of the partici- pants reported knowing two languages besides english, and . % reported knowing three languages besides english. two of the participants reported knowing four or more languages besides english. aside from the early bilinguals, three participants reported high proficiency in other languages learned after the first. none reported having any kind of uncorrected speech or hearing dis- order. we recruited the participants using flyers posted in public areas (e.g., various departments at indiana university) and word of mouth. participants were com- pensated for their time. results to ascertain the validity of the corpus, we used two dependent variables: emotion identification accuracy rates and confidence scores (how confident the partici- pants were in their choices). response times (rts) were collected on each trial but is not analyzed as a depen- dent variable given that the task was not speeded. because there were six choice options on each trial, a random selection would yield an overall accuracy of . %. the data were submitted to a chi-square analysis to estimate whether or not the participants were equally likely to choose among the six possibilities for a given stimulus. table provides the confusion matrix overall, across both speakers, and reveals that overall, emotion portrayals were recognized accurately. figure shows the overall median accuracy in emotion identification by the participants, separated by speaker. random performance level (~ %) is indicated by the dotted line. behav res ( ) : – figure suggests that participants were able to iden- tify each stimulus’ intended emotion above chance. the mean recognition accuracy is %. sadness was recognized most accurately (m = %), followed by neutral (m = %), fear (m = %), disgust (m = %), and anger (m = %). the emotion that was recognized least accurately was happiness (m = %). all emotions were recognized better than chance for both stimulus sets, except for happiness for the k.m. stimuli, which was misidentified as neutrality more of- ten than it was identified as happiness (see table below). a global chi-square analysis on the chosen response categories over all data points (across emotions and the pattern of accuracy remained the same even after removing very slow and very fast trials (rt outliers, defined as data points that were more than . sds beyond all participants’ mean rt, or faster than ms; . % of the data were removed). the slow rts on some trials were likely the result of the option of listening to the stimuli multiple times and of the fact that the task was not speeded. fig. screenshots of the praat script interface for the recognition task. the top panel shows the first screen in a trial, where the emotion labels are highlighted (clickable). the bottom panel shows the second screen in a trial, with the confidence scale now also highlighted. the respondent’s choices appear highlighted in red, and a next button is displayed for participants to move to the next trial. the task was self-paced. up to eight replays were allowed behav res ( ) : – speakers) was significant [χ( ) = , . ; p < . , cramer’s v = . ]. this suggests that for each emotion, respondents did not randomly choose among the six options. before evaluating whether this pattern holds for each emotion separately, we first examined whether there is a difference in accuracy between speakers, as is suggested in fig. . a one-way analysis of variance (anova) comparing accuracy for each speaker (k.m., a.g.) revealed that mean recognition accuracy was significantly higher for a.g. (m = %, %ci = – ) than for k.m. (m = %, %ci = – ), f( , ) = . , p = . . this significant effect of speaker indicates that raters were overall slightly more accurate at recognizing emo- tions portrayed by one speaker (a.g.) over the other (k.m.). however, such differences are to be expected among voice actors, and this is unlikely to reflect an inherent difference among our listener groups. if one group of listeners were systematically less concentrated or accurate during the task, we would expect this dif- ference to hold across the emotions for a given speaker. to verify this, a mixed-effect model with speaker and table classification counts of vocal emotion portrayals by the participants’ responses and the overall proportions of accurate responses (%) within each emotion, across both speakers emotion portrayed responses of participants total anger disgust fear happiness neutral sadness anger count , , , , % within emotion . . . . . . . disgust count , , , , % within emotion . . . . . . . fear count , , % within emotion . . . . . . . happiness count , , , % within emotion . . . . . . . neutral count , , % within emotion . . . . . . . sadness count , , , % within emotion . . . . . . . modal response is indicated in boldface (n = , data points) emotion sadnessneutralhappinessfeardisgustanger m ed ia n ac cu ra cy . . . . . error bars: % ci km ag speaker random performance level (. ) fig. overall median accuracy of emotion identifications by the participants, separated by speaker. the random performance level (~ %) is indicated by the dotted line behav res ( ) : – emotion as fixed factors (and participants as a random factor) was conducted in spss . multiple compari- sons were adjusted with the sidak correction. the type iii tests of fixed effects shows a main effect of speaker [f( , ) = . , p = . ], a main effect of emotion [f( , ) = . , p < . ], and crucially, a significant interaction between the two factors [f( , ) = . , p < . ]. the interaction and pairwise comparisons re- veals that for all emotions except disgust and neutral, a.g.’s portrayals were recognized significantly more ac- curately than k.m.’s; conversely, k.m.’s portrayals of disgust and neutral were recognized significantly more accurately than a.g.’s. the presence of an interaction suggests that it is unlikely to be the case that the k.m. listeners were systematically less accurate than the a.g. listeners (otherwise, one would have expected an ab- sence of interaction). tables and provide the confusion matrices obtained for our stimulus set (emotion portrayal by participants’ choices; n = , data points), separated by speaker. given the significant effect of speaker and the speak- er by emotion interaction, we further conducted a series of chi-square analyses (nonparametric goodness-of-fit tests) in spss for each speaker and emotion sepa- rately, which confirmed the global analysis. the results of the tests for each emotion and each speaker are pro- vided in tables and . they show that the tests were significant for all speakers and all emotions, indicating that listeners were not responding randomly. examination of the patterns of misidentifications in tables and revealed the following tendencies. for a.g., all emotions except fear were most often misidentified as neutral, which represents the second- highest proportion of choices in these cases. in the case of fear, items were misidentified most often as happi- ness. however, even though, for instance, happiness was misinterpreted as neutral in % of the cases for a.g., the reverse was not true: neutral items only were misinterpreted as happiness in % of cases, and were more commonly misinterpreted as anger or sadness, each in roughly % of cases (see table ). the error patterns for the k.m. stimuli stand out, in that happi- ness stimuli were most often recognized as neutral, which is the dominant, modal response. happiness choices were given in % of cases, and neutral choices in %. for the other emotions, unlike for the a.g. stimuli, neutral was the second choice after the correct identification only for anger and sadness. disgust was misidentified as anger in % of cases, more often than neutral, and fear was confused with sadness in % of cases (see table ). this overall high proportion of neutral choices is possibly due to the fact that the stimuli were created at a medium/normal intensity level, without emotional exaggeration, rendering the identification task potentially more difficult. to help researchers evaluate how ambig- uous a given recording is, we also provide the full con- fusion matrix for each stimulus in the database (see bänziger et al., , supplemental materials, for a similar approach). some items were identified at very high accuracy rates by all participants who rated them, and conversely, others were almost never identified correctly. figure shows the accuracy variance obtained for each stimulus (each sound file in the corpus represents one dot). the boxplots in the top and bottom panels of the figure show the distribution and median accuracy for each emotion (top, a.g.; bottom, k.m.). the figures reveal that a proportion of items (particularly for happiness) fell below the random performance level (i.e., . %)—suggesting that these particular stimuli are am- biguous and not ideal representations of the intended emotion, at least for the participants who rated the stimuli. we also obtained confidence ratings for each stimulus rated (i.e., how confident the participants were in their table confusion matrix for the a.g. stimuli, with chi-square goodness-of-fit test per emotion emotion (speaker a.g.) response chi-square goodness-of-fit test a d f h n s total anger , , χ( ) = , , p < . disgust , , χ( ) = , , p < . fear , , χ( ) = , , p < . happiness , , χ( ) = , , p < . neutral , , χ( ) = , , p < . sadness , , χ( ) = , , p < . behav res ( ) : – choices). figure shows the correlations (pearson’s r) between accuracy of identification and the confidence ratings of the participants for each emotion and each speaker separately (see top of each panel of figure ). only one relationship (neutral for a.g.) was not significant. discussion the goal of this project was to create a corpus of au- ditory pseudo-words uttered in different emotions. the corpus includes controlled audio pseudo-words uttered by two actresses in five different emotions (i.e., happiness, sadness, fear, anger and disgust) and in a neutral tone, yielding at least stimuli per ac- tress. in addition, the pseudo-words are based on the pronunciation rules of north american english, and they are not caricatures or exaggerations of the emo- tions portrayed. each recording has been validated by native english listeners in terms of recognition accuracy of the intended emotion portrayal. overall, the emotions were recognized at accuracy levels that were clearly higher than chance (m = % across emotions and speakers, for a chance level at about %). the recog- nition proportions obtained for our data were most ac- curate for fear, neutral, and sadness, and least accurate for happiness and disgust, consistent with previous data from other languages (banse & scherer, ; castro & lima, ; liu & pell, ; pell et al., ; scherer et al., ; van bezooijen, ). the one exception is anger. in our stimuli, anger was recognized with surpris- ingly low accuracy ( %). it is often among the best recognized emotions (e.g., bänziger et al., ; scherer et al., ; wendt & scheich, ). this effect was possibly due to the fact that our pseudo-words were produced with a medium/normal emotional intensity lev- el, possibly making them more confusable with neutral stimuli. indeed, for both speakers (and particularly for k.m.), anger was most often confused with neutrality. the resulting accuracy in our dataset was globally sim- ilar to the levels reported in previous studies on vocal emotion (hovering in the %– % range; see scherer et al., ), in particular among the studies that used similar stimuli (words or short sentences, such as rigoulot et al., ) and a similar number of response options. the audio stimuli were created as high-quality re- cordings in the .wav format, which allows experi- menters to run more detailed acoustic analyses in order to match stimuli for specific experimental purposes. for instance, intensity (as loudness, in decibels) and duration measurements (in milliseconds) are provided in the corpus database, but other acoustic parameters can be extracted, such that matched stimuli could be selected for the needs of an eeg study, for instance. the corpus is available at https://psycholinguistics. indiana.edu/hoosiervocalemotions.htm. the website prvides basic information about the corpus and how to request access to the sound files and the database. for each item, recognition accuracy and confusion patterns, as well as speaker, filename, and a number of acoustic details, are provided in an accompanying database, in order to allow researchers to select items specifically for their needs. the list of attributes provided for each sound file in the corpus is detailed in the appendix. a number of methodological issues need to be con- sidered. first, the validation of the stimuli was based on data collected in a laboratory setting using a forced choice methodology with six response alternatives. even though this methodology is commonly used across studies, its ecological validity for real-time interactions in social situations remains limited. it is unclear to what extent these results would generalize to real-life situa- tions outside the laboratory, or to experimental paradigms in which a given stimulus was presented only once without any table confusion matrix for the k.m. stimuli, with chi-square goodness-of-fit test per emotion emotion (speaker k.m.) response chi-square goodness-of-fit test a d f h n s total anger , , χ( ) = , p < . disgust , , χ( ) = , , p < . fear , , χ( ) = , , p < . happiness , , χ( ) = , , p < . neutral , , χ( ) = , , p < . sadness , , χ( ) = , , p < . the underlined number indicates that for these stimuli, happiness was not chosen as the modal response for intended happy stimuli; neutral was the most frequently chosen response behav res ( ) : – https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm available “categorization labels,” because forced choice pro- cedures produce better performance than free-choice tests (see bachorowski & owren, ). second, similar considerations are involved with the specific linguistic context in which an emotion is heard and the type of linguistic materials used. hearing a short (two-syllable) pseudo-word in order to identify an emotion is likely much more difficult than identify- ing it via a longer, meaningful sentence (see rigoulot et al., ), and is likely to lead overall to lower recognition accuracy. similarly, medium/normal emo- tional intensity (as opposed to high, such as in affect bursts) is likely to make emotion recognition less straightforward. taken together, the identification emotion spoken by a.g. sadnessneutralhappinessfeardisgustanger a cc ur ac y . . . . . . random performance level emotion spoken by k.m. sadnessneutralhappinessfeardisgustanger a cc ur ac y . . . . . . random performance level a b fig. box plot with overlaid dot plots for each emotion’s identification accuracy. each dot represents one stimulus (i.e., one sound file in the corpus). horizontal lines represent the medians, boxes show the interquartile range (iqr) representing % of the cases, and whisker bars extend to . times the iqr. (top) a.g. stimuli. (bottom) k.m. stimuli behav res ( ) : – accuracy we obtained in our study was the product of the forced choice methodology, as well as of the medium/normal intensity of the stimuli, the fact that they are pseudo-words presented in isolation, and the context-free format of their presentation in the recog- nition task. third, the corpus includes a limited set of emotions (i.e., happiness, sadness, fear, anger, and disgust) and a neutral tone. other emotions could have been included (i.e., surprise and contempt). we selected the emotions to be included in the corpus on the basis of the basic emotions identified by ekman ( ) and of whether they can have either a positive or a negative valence. therefore, surprise was not included, because it can have any valence (it can be neutral, positive, or nega- tive), and also because this emotion can be difficult to simulate in the laboratory (pell et al., ). researchers and clinicians should then consider this limitation when selecting this corpus for their work, as well as the fact that only one positive emotion (i.e., happiness) is included, which would impede systematic analyses of valence effects and the examination of dif- ferent positive emotions. fourth, resarchers should also consider that the cor- pus contains pseudo-words uttered by two females (i.e., it does not include male voices). finally, because the validation of the stimuli was based on a between- subjects design (i.e., each participant rated the pseudo- words from one actress only), it is hard to establish differences in the validation between the two speakers. although this design could be seen as a limitation, due to logistics, the information we provide in the corpus should enable researchers and clinicians to make in- formed decisions as to what stimuli to select for their work. potential applications of the hoosier vocal emotions corpus the hoosier vocal emotions corpus was specifically developed for the requirements of eeg research on emotion processing. the stimuli from this corpus were first used in a study on the neural responses (using eeg techniques) to vocal emotion processing and their asso- ciations with temperamental traits and behavioral prob- lems in young children (hoyniak et al., ). the cor- pus has unique characteristics that are useful for exper- imental paradigms requiring controlled stimuli (e.g., eeg or fmri studies)—namely, they are disyllabic a c c u r a c y . . . . . . emotion . . . . . . a c c u r a c y . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . s p e a k e r a g k m anger disgust fear happiness neutral sadness confidence r = . , p < . r = . , p < . r = . , p < . r = . , p < . r = . , n.s. r = . , p < . r = . , p < . r = . , p < . r = . , p < . r = . , p < . r = . , p < . r = . , p < . fig. correlations between the mean identification accuracy for each stimulus and the mean confidence ratings by speaker (top panels, a.g.; bottom panels, k.m.) behav res ( ) : – pseudo-words (i.e., short stimuli without a semantic meaning) that are overall similar in terms of duration and loudness, and that represent medium/normal emo- tional intensity. to the best of our knowledge, the magdeburger prosodie korpus (wendt et al., ; wendt & scheich, ) is the only other corpus that includes isolated disyllabic pseudo-words. however, this corpus is composed of stimuli that respect the phonotactic and phonetic rules of the german language. although there are data suggesting that emotions can be recognized across languages and cultures, there is still an in-group advantage in the processing of emotional vocalizations (sauter et al., ). we therefore developed new emo- tional vocalizations based on the phonology and pro- nunciation rules of north american english, for re- search and clinical work requiring english-based stimuli. the use of the corpus does not need to be limited to english speakers, however. for instance, studies of emo- tion or prosodic processing in monolingual or in multi- lingual individuals, or in nonnative english speakers, could be easily conducted using stimuli from this corpus (e.g., dewaele, ; min & schirmer, ; paulmann & uskul, ). stimuli from the corpus could also be used to inves- tigate emotion processing in individuals with certain temperamental or behavioral characteristics associated with difficulties in emotion recognition (e.g., individuals with psychopathic traits or alexithymia). in addition, the stimuli could be used to study the extent to which pa- tients with aphasia, schizophrenia, or other mental dis- orders (e.g., depression) are able to process prosodic/ vocal emotion information. the hoosier vocal emotions corpus’s short, disyllab- ic pseudo-words, which are acoustically more homoge- neous than longer sentences, can also be useful to re- searchers performing acoustic analyses. investigations that seek to characterize the prosodic and acoustic fea- tures of different emotions would benefit from this kind of tightly controlled and not exaggerated materials, since they can help isolate specific acoustic parameters for emotion recognition more precisely. also, the fact that our stimuli were produced with normal emotional intensity (as opposed to high, such as in affect bursts) contributes to creating more ambiguity in the corpus and makes emotion recognition not only less straightfor- ward, but possibly also more ecologically valid. ambiguous or subtle acoustic characteristics can be studied with a corpus like ours, that preserves this var- iability, and because we provide the full confusion ma- trix for each stimulus, researchers seeking to determine the acoustic parameters of various emotions will have a large range of clear, ambiguous, and misclassified stim- uli to choose from. this variability and the range of stimulus uncertainty could also be very useful for the field of automatic emotion recognition. training para- digms would thus be able first to use the nonambiguous stimuli (see brendel, zaccarelli, schuller, & devillers, ) and progressively to incorporate more subtle stimuli, ultimately leading to robust recognition scores. finally, the neutral-tone stimuli can be used on their own for research applications other than emotional processing. for instance, they could be used for pseudo-word or voice recog- nition tasks in investigations of individual differences in audi- tory, phonetic, or phonological processing or learning. conclusion in this article, we have presented the hoosier vocal emotions corpus, a set of controlled disyllabic pseudo-words spoken in five basic emotions and in a neutral tone. this corpus is one of the few databases of pseudo-word vocal expressions for north american english. the corpus consists of , high-definition audio recordings by two female speakers at a medium/ normal emotional intensity level. the validation of the corpus with a forced choice recognition paradigm re- vealed high rates of emotional validity. the recognition accuracy for each item and the full confusion matrix are provided in an accompanying database, which will al- low researchers to explore the full range of stimulus uncertainty. despite some of the limitations discussed above, this corpus presents a valuable resource for a wide variety of researchers and clinicians. acknowledgments we gratefully acknowledge the department of criminal justice and the office of women’s affairs (women in science program) at indiana university for their financial support (grant to n.m.g.f.), as well as the participants involved in this study. we also thank gabriela cepeda, franziska krüger, pyoung-hwa han, trisha thomas, chung-lin yang, and joshua lee for their assistance with the pseudo-word creation, acoustic analyses, participant testing, and data analysis. we are indebted to the actresses who took part in the recording sessions. we are further grateful to beate wendt for sharing some of her german stimuli, which were extremely useful in constructing our corpus. n.m.g.f. is a research scholar, junior , fonds de recherche du québec–santé. compliance with ethical standards conflict of interest none. open practices statement the validation study was not preregistered. the data and materials for all experiments are available at https:// psycholinguistics.indiana.edu/hoosiervocalemotions.htm. behav res ( ) : – https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm appendix: the following table outlines the structure of the corpus (see https://psycholinguistics.indiana.edu/hoosiervocalemotions. htm). row and row refer to the corresponding rows in the excel file (see the website link). each line is a column header in the excel file or in the comma-delimited spreadsheet (csv). explanation provides a brief outline of the column content. row row explanation ipa international phonetic alphabet transcription spelling item in english roman alphabet item item number token token number file_name audio file name with extension duration_ms file duration in milliseconds intensity_average_db average intensity in db intensity_min minimum intensity intensity_max maximum intensity voice speaker list list number n_listeners number of listeners who rated this list emotion emotion accuracy_mean mean accuracy over all trials confidence_mean mean confidence score over all trials confusion_matrix_cnt a confusion matrix: raw count of trials in which the emotion was chosen, over all trials d f h n s confusion_matrix_prct a confusion matrix:% of trials in which the emotion was chosen, over all trials d f h n s accuracy_mean_validrt mean accuracy over selected trials only (rt outliers removed) confidence_mean_validrt mean confidence score over selected trials only (rt outliers removed) confusion_matrix_cnt_validrt a confusion matrix: raw count of trials in which the emotion was chosen, over selected trials only d f h n s confusion_matrix_prct_validrt a confusion matrix: % of trials in which the emotion was chosen, over selected trials only d f h n s rt_mean mean rt over all trials rt_median median rt over all trials rt_mean_validrt mean rt over selected trials (rt outliers removed) rt_median_validrt median rt over selected trials (rt outliers removed) a, anger; d, disgust; f, fear; h, happiness; n, neutral; s, sadness. behav res ( ) : – https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm https://psycholinguistics.indiana.edu/hoosiervocalemotions.htm references baayen, r. h., piepenbrock, r., & gulikers, l. ( ). the celex lexical database (release , cd-rom). philadelphia, pa: linguistic data consortium, university of pennsylvania. bachorowski, j.-a., & owren, m. j. ( ). vocal expressions of emotion. in m. lewis, j. m. haviland-jones, & l. feldman barrett (eds.), handbook of emotions ( rd ed., pp. – ). new york, ny: guildford press. banse, r., & scherer, k. r. ( ). acoustic profiles in vocal emotion expression. journal of personality and social psychology, , – . bänziger, t., mortillaro, m., & scherer, k. r. ( ). introducing the geneva multimodal expression corpus for experimental research on emotion perception. emotion, , – . belin, p., fillion-bilodeau, s., & gosselin, f. ( ). the montreal affective voices: a validated set of nonverbal af- fect bursts for research on auditory affective processing. behavior research methods, , – . https://doi.org/ . /brm. . . boersma, p., & weenink, d. ( ). praat: doing phonetics by computer (version . . ) [computer program]. retrieved from www.praat. org brendel, m., zaccarelli, r., schuller, b., & devillers, l. ( ). towards measuring similarity between emotional corpora. in proceedings of the international conference on language resources and evaluation (pp. – ). luxembourg city: european language resources association. burkhardt, f., paeschke, a., rolfes, m., sendlmeier, w., & weiss, b. ( ). a database of german emotional speech. in proceedings of interspeech (pp. – ). lisbon, portugal. castro, s. l., & lima, c. f. ( ). recognizing emotions in spoken language: a validated set of portuguese sentences and pseudosentences for research on emotional prosody. behavior research methods, , – . clopper, c. g., & pisoni, d. b. ( ). some acoustic cues for the perceptual categorization of american english regional dialects. journal of phonetics, , – . costantini, g., iadarola, i., paoloni, a., & todisco, m. ( ). emovo corpus: an italian emotional speech database. in n. calzolari, k. choukri, t. declerck, h. loftsson, b. maegaard, j. mariani, . . . s. piperidis (eds.), proceedings of the ninth international conference on language resources and evaluation (pp. – ). luxembourg city: european language resources association. dewaele, j.-m. ( ). the emotional force of swearwords and taboo words in the speech of multilinguals. journal of multilingual and multicultural development, , – . ekman, p. ( ). an argument for basic emotions. cognition and emotion, , – . https://doi.org/ . / herba, c., & phillips, m. ( ). annotation: development of facial expression recognition from childhood to adolescence: behavioral and neurological perspectives. journal of child psychology and psychiatry, , – . hoyniak, c. p., bates, j. e., petersen, i. t., yang, c.-l., darcy, i., & fontaine, n. m. g. ( ). reduced neural responses to vocal fear: a potential biomarker for callous-uncaring traits in early childhood. developmental science, , e . juslin, p. n., & laukka, p. ( ). communication of emotions in vocal expression and music performance: different channels, same code? psychological bulletin, , – . https://doi.org/ . / - . . . laukka, p., elfenbein, h. a., chui, w., thingujam, n. s., iraki, f. k., rockstuhl, t., & althoff, j. ( ). presenting the venec corpus: development of a cross-cultural corpus of vocal emotion expressions and a novel method of annotating emotion appraisals. in proceedings of the international conference on language resources and evaluation (pp. – ). luxembourg city: european language resources association. lima, c.f., castro, s. l., & scott, s. k. ( ). when voices get emo- tional: a corpus of nonverbal vocalizations for research on emotion processing. behavior research methods, , – . https:// doi.org/ . /s - - - liu, p., & pell, m. d. ( ). recognizing vocal emotions in mandarin chinese: a validated database of chinese vocal emotional stimuli. behavior research methods, , – . https://doi.org/ . /s - - - livingstone, s. r., & russo, f. a. ( ). the ryerson audio– visual database of emotional speech and song (ravdess): a dynamic, multimodal set of facial and vocal expressions in north american english. plos one, , e . https:// doi.org/ . /journal.pone. min, c. s., & schirmer, a. ( ). perceiving verbal and vocal emotions in a second language. cognition and emotion, , – . parsons, c. e., young, k. s., craske, m. g., stein, a. l., & kringelbach, m. l. ( ). introducing the oxford vocal (oxvoc) sounds database: a validated set of non-acted affec- tive sounds from human infants, adults, and domestic ani- mals. frontiers in psychology, , . https://doi.org/ . /fpsyg. . paulmann, s., & uskul, a. k. ( ). cross-cultural emotional prosody recognition: evidence from chinese and british listeners. cognition and emotion, , – . pell, m. d., paulmann, s., dara, c., alasseri, a., & kotz, s. a. ( ). factors in the recognition of vocally expressed emotions: a com- parison of four languages. journal of phonetics, , – . pollak, s. d., & sinha, p. ( ). effects of early experience on children’s recognition of facial displays of emotion. developmental psychology, , – . rigoulot, s., wassiliwizky, e., & pell, m. d. ( ). feeling backwards? how temporal order in speech affects the time course of vocal emo- tion recognition. frontiers in psychology, , . https://doi.org/ . /fpsyg. . sauter, d. a., eisner, f., ekman, p., & scott, s. k. ( ). cross-cultural recognition of basic emotions through nonverbal emotional vocali- zations. proceedings of the national academy of sciences, , – . scherer, k. r., banse, r., & wallbott, h. g. ( ). emotion inferences from vocal expression correlate across languages and cultures. journal of cross-cultural psychology, , – . scherer, k. r., banse, r., wallbott, h. g., & goldbeck, t. ( ). vocal cues in emotion encoding and decoding. motivation and emotion, , – . scherer, k. r., clarke-polner, e., & mortillaro, m. ( ). in the eye of the beholder? universality and cultural specificity in the expression and perception of emotion. international journal of psychology, , – . tottenham, n., tanaka, j., leon, a. c., mccarry, t., nurse, m., hare, t. a., . . . nelson, c. a. ( ). the nimstim set of facial expressions: judgments from untrained research partic- ipants. psychiatry research, , – . van bezooijen, r. a. m. g. ( ) characteristics and recogniz- ability of vocal expressions of emotion. dordrecht, the netherlands: foris. retrieved on january , , from https://repository.ubn.ru.nl/handle/ / ververidis, d., & kotropoulos, c. ( ). emotional speech recognition: resources, features, and methods. speech communication, , – . wendt, b., hufnagel, k., brechmann, a., gaschler-markefski, b., tiedge, j., ackermann, h., & scheich, h. ( ). a method behav res ( ) : – https://doi.org/ . /brm. . . https://doi.org/ . /brm. . . http://www.praat.org http://www.praat.org https://doi.org/ . / https://doi.org/ . / - . . . https://doi.org/ . / - . . . https://doi.org/ . /s - - - https://doi.org/ . /s - - - https://doi.org/ . /s - - - https://doi.org/ . /s - - - https://doi.org/ . /journal.pone. https://doi.org/ . /journal.pone. https://doi.org/ . /fpsyg. . https://doi.org/ . /fpsyg. . https://doi.org/ . /fpsyg. . https://doi.org/ . /fpsyg. . https://repository.ubn.ru.nl/handle/ / for creation and validation of a natural spoken language cor- pus used for prosodic and speech perception. brain and language, , . https://doi.org/ . /s - x( ) - wendt, b., & scheich, h. ( ). the “magdeburger prosodie-korpus.” in proceedings of the speech prosody conference (pp. – ). aix-en-provence, france: isca. publisher’s note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. behav res ( ) : – https://doi.org/ . /s - x( ) - https://doi.org/ . /s - x( ) - 	the hoosier vocal emotions corpus: a validated set of north american english pseudo-words for evaluating emotion processing 	abstract 	method 	creation of the stimuli 	elicitation and recording procedures 	validation of the stimuli 	results 	discussion 	potential applications of the hoosier vocal emotions corpus 	conclusion 	appendix: 	references paclitaxel plus bevacizumab in patients with chemosensitive relapsed small cell lung cancer: a safety, feasibility, and efficacy study from the hoosier oncology group brief report paclitaxel plus bevacizumab in patients with chemosensitive relapsed small cell lung cancer a safety, feasibility, and efficacy study from the hoosier oncology group shadia jalal, md,* pablo bedano, md,† lawrence einhorn, md,* sumeet bhatia, md,‡ rafat ansari, md,§ naftali bechar, md,§� karuna koneru, md,¶ ramaswamy govindan, md,# jingwei wu, ms,** menggang yu, phd,** bryan schneider, md,* and nasser hanna,md* introduction: bevacizumab when combined with carboplatin and paclitaxel improves response rates (rrs) and overall survival in patients with advanced non-small cell lung cancer. paclitaxel has single-agent activity in relapsed small cell lung cancer (sclc). angiogenesis seems to play an important role in the pathogenesis of sclc. this study evaluated the safety and efficacy of paclitaxel plus bevacizumab in patients with chemosensitive relapsed sclc. methods: patients with relapsed chemosensitive sclc with an eastern cooperative oncology group performance status of to were eligible. they received paclitaxel mg/m intravenously on days , , and . bevacizumab was administered at mg/kg intravenously on days and . cycles were every days. the primary endpoint was progression-free survival (pfs). secondary endpoints included rrs, toxicity, and overall survival. correlative studies evaluated vascular endothelial growth factor polymorphisms. results: thirty-four patients were enrolled in the study. median age was . (range, – ) years, male:female: . %: . %, eastern cooperative oncology group performance status : . %: . %. median progression-free survival was . weeks (equivalent to historical controls). median survival time was weeks. the overall rr was . %. stable disease rate was . %, and . % of patients had progressive disease. no unexpected toxicities were noted, and grade / toxicities were limited to neutropenia, fatigue, and dyspnea. none of the vascular endothelial growth factor poly- morphisms evaluated were significantly associated with response. conclusions: the addition of bevacizumab to paclitaxel does not improve outcomes in relapsed chemosensitive sclc. key words: relapsed small cell, paclitaxel, bevacizumab. (j thorac oncol. ; : – ) small cell lung cancer (sclc) accounts for to % oflung cancers with the majority of patients ultimately succumbing to their disease. the majority of patients are diagnosed with extensive disease at presentation and are treated with platinum-based therapy. depending on the dura- tion of response to platinum-based therapy, patients are described as having chemosensitive or chemorefractory dis- ease. those with chemosensitive disease, typically defined as relapsing more than to days after completion of platinum-based therapy, frequently benefit from further che- motherapy at the time of disease progression. multiple agents have activity in the second-line setting, but responses are typically brief ( – months), and survival times are short (median, months). paclitaxel also has single-agent activity in chemotherapy-naive patients with sclc with a response rate (rr) of to %. , in addition, paclitaxel is active in a chemorefractory population with sclc. bevacizumab (avastin, genentech, south san fran- cisco, ca) is a recombinant vascular endothelial growth factor (vegf) monoclonal antibody that was included in the first-line treatment paradigm of advanced non-small cell lung cancer after its combination with carboplatin/paclitaxel in the eastern cooperative oncology group (ecog) study. this trial showed an improvement in rr and overall survival (os) compared with chemotherapy alone. sclc is a tumor with early hematogenous spread. tumor microvessel count and vegf expression were prog- *indiana university melvin and bren simon cancer center, indianapolis; †clarian hematology/oncology, indianapolis; ‡community regional cancer center, indianapolis; §northern indiana cancer research con- sortium, south bend; �community oncology center, kokomo; ¶cancer care center of southern indiana, bloomington, indiana; #washington university medical center, siteman cancer center, st. louis, missouri; and division of biostatistics, indiana university school of medicine, indianapolis, indiana. disclosure: dr. einhorn holds stock in amgen and glaxosmith kline. dr. bhatia was paid to be on the advisory board of genentech. dr. govindan has served as an advisor to genentech, lilly oncology, astrazeneca, gsk, and boehringer ingelheim pharma. dr. schneider is on the advi- sory board for genentech, has previously received research grant for biomarker discovery from genentech, and has a us patent for vegf single nucleotide polymorphisms the rights of which belong to genen- tech. dr. hanna has disclosed that his institution and the hoosier oncology group received a grant from genentech to conduct the study. address for correspondence: shadia jalal, md, iu simon cancer center, barnhill drive, rt , indianapolis, in . e-mail: sjalal@iupui.edu copyright © by the international association for the study of lung cancer issn: - / / - journal of thoracic oncology • volume , number , december nostic factors and correlated with worse survival (p � . and p � . , respectively) in patients with sclc who underwent pneumonectomies or lobectomies. vegf was also reported to be expressed in % of patients with sclc. therefore, we initiated this study evaluating the safety and efficacy of the combination of paclitaxel and bevaci- zumab in relapsed chemosensitive sclc. we used the dosing schedule evaluated as initial treatment in patients with met- astatic breast cancer in which the combination resulted in an improvement in rr ( . % versus . %, p � . ) and progression-free survival (pfs; . versus . months, haz- ard ratio � . , p � . ) over paclitaxel alone. patients and methods eligibility criteria included age � years, histo- logic or cytologic diagnosis of sclc, prior treatment with at least one platinum-containing regimen, ecog perfor- mance status to , and measurable disease by response evaluation criteria in solid tumor (recist) group re- sponse criteria. all patients had chemosensitive disease defined as relapse more than days after completing initial treatment. adequate hematologic, renal, and liver function at baseline were required. no therapies were allowed within weeks of registration. patients were excluded if they had evidence of brain metastases within days of registration. all patients provided written informed consent before enrolling in the study. local institutional review boards approved the protocol. treatment plan patients received mg/m of paclitaxel as a -hour intravenous infusion on days , , and and bevacizumab at mg/kg intravenously on days and . cycles were repeated every days. patients were premedicated before paclitaxel per package insert. bevacizumab was administered after the completion of paclitaxel on day of cycle . after to cycles of treatment with the combination, patients were allowed to continue with bevacizumab alone as maintenance therapy until disease progression or intolerable side effects. statistical analysis the primary endpoint of the study was pfs defined as the interval from the start of therapy until disease progression or death. secondary endpoints included rr as defined by the response evaluation criteria in solid tumor criteria and os time defined as time from initiation of first cycle to death. other secondary endpoints included determination of adverse events, their severity, and all grade and grade toxicities. the median pfs with historical controls (topotecan) is weeks. with a total number of patients, this study had % power and � � %, to detect an improvement in pfs to weeks. laboratory correlates correlative studies involved evaluation of vegf polymorphisms and their correlation with rr. fisher’s exact test was used to correlate vegf polymorphisms with response data. blood samples were collected in -ml lavender-top tubes from patients agreeable to laboratory correlatives. the following vegf polymorphisms were evaluated: � c/t, � g/c, � c/a, � g/a, and � c/t. these specific polymorphisms were selected because of their known impact on modulating angiogenesis. polymorphisms were assessed using standard polymerase chain-restriction frag- ment length polymorphism methods that are previously es- tablished and taqman-based assays. results between april , , and january , , patients were enrolled in the study. baseline characteristics are summarized in table . treatment administered (n � ) the median number of cycles received on trial was (range, – ). approximately % of patients were able to receive the study regimen without dose modifications. fifty percent of the patients required no dose delays. only two patients continued on the maintenance bevacizumab phase after six cycles of paclitaxel plus bevacizumab. efficacy response results are available for patients because one patient never received treatment. one complete response (cr) was observed, five patients achieved a partial response (pr) ( . %) for an overall rr of . %. fifteen patients progressed ( . %), and stable disease was noted in patients ( . %). the median pfs was . weeks ( % confidence interval � – . ), and the median survival time (mst) was weeks ( % confidence interval � – ). table . baseline characteristics (n � ) characteristics n � age (yr) median . range – sex, n (%) male ( . ) female ( . ) ecog ps ( . ) ( . ) smoking history, n (%) current ( . ) former ( . ) never ( . ) prior treatment, n (%) . chemotherapy ( . ) . chemotherapy ( . ) radiotherapy ( ) median number of days from previous regimen to registration (range) ( – ) ecog ps, eastern cooperative oncology group performance status. journal of thoracic oncology • volume , number , december paclitaxel plus bevacizumab in patients with sclc copyright © by the international association for the study of lung cancer safety/toxicity toxicities noted in � % of patients (n � ) are presented in tables and . no unexpected toxicities were observed with the combination. grade pulmonary hemor- rhage was noted in one patient. three patients had grade to grade proteinuria. there were no febrile neutropenia epi- sodes and no treatment-related deaths. laboratory correlative results data were available for patients to assess vegf polymorphisms and association with response. the frequen- cies of the various vegf polymorphisms in different patient response groups are presented in table . no associations were statistically significant, although the association of vegf ga with response approached statistical signifi- cance (p � . ). discussion the addition of bevacizumab to paclitaxel does not seem to improve pfs in patients with chemosensitive re- lapsed sclc. the median pfs reported with this combina- tion is similar to that reported with topotecan. in addition, the rr of . % is similar to that of other regimens, and the mst was . months. the regimen seems safe, and no unexpected toxicities were observed. our study yielded comparable results with another phase ii study, which combined topotecan with bevaci- zumab. that trial included both chemosensitive and che- morefractory patients, was associated with more grade to grade toxicities, and reported a mst and pfs of only and . weeks, respectively. other studies have evaluated bevacizumab combined with chemotherapy in previously untreated patients with sclc. ecog conducted a phase ii trial that evaluated the addition of bevacizumab to cisplatin and etoposide. the combination was tolerable with no unexpected toxicities, but efficacy was not substantially different than expected with cisplatin plus etoposide alone. the cancer and leukemia group b evaluated the combination of cisplatin and irinote- can plus bevacizumab in untreated patients with extensive stage sclc. median pfs was . months, and mst was . months. these results compared favorably with out- comes from other trials, but the study failed to reach its primary endpoint of improving survival times compared with historical controls. the sara cannon group evaluated bev- acizumab combined with carboplatin and irinotecan in a similar patient population. the median time to disease pro- gression was . months, and mst was . months. nevertheless, this regimen was associated with a significant incidence of grade or more diarrhea and fatigue. collectively, this data suggest that bevacizumab is unlikely to substantially improve outcomes in unselected patients with sclc. our study also investigated whether subgroups of patients with certain vegf polymorphisms may preferentially benefit from bevacizumab. schneider et al. were the first to describe vegf polymorphisms that correlate with efficacy and toxicity with bevacizumab in patients with metastatic breast cancer. in their analysis, table . hematologic toxicities (n � ) toxicity any grade, n (%) grades – , n (%) hematologic neutropenia ( ) ( . ) febrile neutropenia anemia ( ) thrombocytopenia ( . ) table . nonhematologic toxicities (n � ) toxicity any grade, n (%) grades – , n (%) fatigue ( ) ( ) dyspnea ( . ) ( . ) hypertension ( . ) neuropathy ( ) nausea ( ) vomiting ( . ) liver enzyme abnormalities elevation in alkaline phosphatase ( . ) elevation in ast ( . ) diarrhea ( ) constipation ( ) anorexia ( . ) mucositis ( . ) hyperglycemia ( . ) headache ( . ) ast, aspartate aminotransferase. table . frequencies of vegf polymorphisms and correlation with responses (n � ) vegf polymorphism cr/pr (%) pd (%) sd (%) vegf- aa ( . ) ( ) ( . ) ca ( ) ( . ) ( . ) cc ( ) ( . ) ( ) vegf- aa ( ) ( . ) ( ) ga ( . ) ( ) ( . ) gg ( ) ( ) ( . ) vegf cc ( ) ( . ) ( . ) ct ( . ) ( ) ( . ) tt ( . ) ( ) ( . ) vegf- cc ( ) ( . ) ( . ) gc ( ) ( . ) ( . ) gg ( . ) ( . ) ( ) vegf- cc ( . ) ( ) ( . ) ct ( ) ( ) ( . ) tt ( ) ( . ) ( ) vegf, vascular endothelial growth factor; sd, stable disease; pd, progressive disease; cr, complete response; pr, partial response. jalal et al. journal of thoracic oncology • volume , number , december copyright © by the international association for the study of lung cancer vegf- aa and vegf- aa genotypes correlated with a superior median os with a bevacizumab-containing regimen in advanced breast cancer. in this analysis, vegf- cc and vegf- tt genotypes were associated with a decreased risk of significant hypertension with bevaci- zumab. this study is the first to evaluate vegf polymor- phisms in sclc. we were unable to identify a vegf polymorphism that was associated with response. neverthe- less, interpretation of vegf polymorphisms and their impact on response in our study should be viewed with caution due to the small patient number, the even smaller patient number in different response subgroups, and the lack of grade to grade hypertension, a potential surrogate for benefit with vegf-directed therapy. furthermore, multiple association tests were performed; therefore, the traditional p value less than . for statistical significance may not be valid. patients with relapsed sclc continue to be in need of improved therapy options. continued understanding of the complex biology of this smoking-associated illness confirms the differences between sclc and non-small cell lung cancer where multiple agents of benefit in the latter are ineffective in sclc. chemotherapy remains the cornerstone of therapy for patients with sclc. novel therapies and strategies in the treatment of this disease continue to be explored. acknowledgments supported by genentech. references . govindan r, page n, morgensztern d, et al. changing epidemiology of small cell lung cancer in the united states over the last years: analysis of the surveillance, epidemiologic, and end results database. j clin oncol ; : – . . ettinger ds, finkelstein dm, sarma rp, et al. phase ii study of paclitaxel in patients with extensive-disease small-cell lung cancer: an eastern cooperative oncology group study. j clin oncol ; : – . . kirschling rj, grill jp, marks rs, et al. paclitaxel and g-csf in previously untreated patients with extensive stage small-cell lung can- cer: a phase ii study of the north central cancer treatment group. am j clin oncol ; : – . . smit ef, fokkema e, biesma b, et al. a phase ii study of paclitaxel in heavily pretreated patients with small-cell lung cancer. br j cancer ; : – . . sandler a, gray r, perry mc, et al. paclitaxel-carboplatin alone or with bevacizumab for non-small cell lung cancer. n engl j med ; : – . . lucci m, mussi a, fontanini g, et al. small cell lung carcinoma (sclc): the angiogenic phenomenon. eur j cardiothorac surg ; : – . . dowell je, amirkhan rh, lai ws, et al. survival in small cell lung cancer is independent of tumor expression of vegf and cox- . anticancer res ; : – . . miller k, wang m, gralow j, et al. paclitaxel plus bevacizumab versus paclitaxel alone for metastatic breast cancer. n engl j med ; : – . . schneider bp, radovich m, sledge gw, et al. association of polymor- phisms of angiogenesis genes with breast cancer. breast cancer res treat ; : – . . waterhouse dm, morgan sk, spigel dr, et al. phase ii study of oral topotecan plus bevacizumab (topo-bev) for second line treatment of small cell lung cancer (sclc). j clin oncol :abstract . . horn l, dahlberg se, sandler ab, et al. phase ii study of cisplatin plus etoposide and bevacizumab for previously untreated, extensive-stage small cell lung cancer: eastern cooperative oncology group study e . j clin oncol ; : – . . ready n, dudeck az, wan xf, et al. calgb :a phase ii study of cisplatin (c),irinotecan (i) and bevacizumab (b) for untreated extensive stage small cell lung cancer (es-sclc). j clin oncol ; :abstract . . noda k, nishiwaki y, kawahara m, et al. irinotecan plus cisplatin compared with etoposide plus cisplatin for extensive small cell lung cancer. n engl j med ; : – . . spigel dr, greco a, zubkus jd, et al. phase ii trial of irinotecan,car- boplatin, and bevacizumab in the treatment of patients with extensive- stage, small cell lung cancer. j thorac oncol ; : – . . schneider bp, wang m, radovich m, et al. association of vascular endothelial growth factor and vascular endothelial growth factor recep- tor genetic polymorphisms with outcome in a trial of paclitaxel compared with paclitaxel plus bevacizumab in advanced breast cancer: ecog . j clin oncol ; : – . journal of thoracic oncology • volume , number , december paclitaxel plus bevacizumab in patients with sclc copyright © by the international association for the study of lung cancer 	paclitaxel plus bevacizumab in patients with chemosensitive relapsed small cell lung cancer: a safety, feasibility, and efficacy study from the hoosier oncology group 	patients and methods 	treatment plan 	statistical analysis 	laboratory correlates 	laboratory correlates 	results 	treatment administered (n = ) 	efficacy 	safety/toxicity 	laboratory correlative results 	discussion 	acknowledgments 	references wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ aaas leshner fellows help confront climate impacts sciencemag.org s c i e n c e p h o t o : a r ia n a s u t t o n - g r ie r sep tember • vol issue by michaela jarvis climate scientists participating in the aaas leshner leadership institute are engaging communities across the united states in a way that could skirt—and possibly defuse—ideological opposition to acknowledging that the climate is changing. on the front lines of a trend and drawing on training they received after being selected as the aaas leshner fellows, the scientists are convening community leaders and policy-makers—in fields such as city planning, agriculture, disaster planning, and natural resource management—and asking them directly what kinds of scientific information they need to protect against ongoing changes in the climate, such as record flooding and heat waves. “for me, public engagement is about working from the bottom up—one audience at a time—building relationships and trust, sharing knowledge, and figuring out how to address the needs of stakeholders,” said fellow benjamin preston, senior research scientist at oak ridge national laboratory and deputy director of the labora- tory’s -person climate change science institute. preston is at work on an institution-wide plan to mobilize scientists to find out from city and county leaders “how our science, data, and tools can inform some of their challenging decisions.” a march gallup poll showed a sharp ideological divide in the acknowledgment of climate change, with just % of republicans worried “a great deal” or “a fair amount” about global warming, as opposed to % of democrats. according to research on the phenomenon, climate scientists presenting their data to the public has had little effect on public opinion. “more information or more basic knowledge doesn’t seem to overcome polarized differences on issues such as climate change,” matthew nisbet, associate professor of communication studies and affiliate associate professor of public policy and urban affairs at northeastern university, told the fellows at a training session. “the whole concept of public engagement,” said tiffany lohwater, aaas deputy chief communications officer, who helps run the leshner leadership institute program, “is recognizing that audiences have something to contribute to the conversation, that this can and should be a two-way dialogue. particularly on issues that are politi- cally polarized, this becomes even more important.” earlier this month, fellow kirstin dow helped organize the carolinas climate resilience conference to bring together represen- tatives of local communities, nongovernmental organizations, state and federal agencies, and researchers. with more than partici- pants, the conference was designed to promote scientific study that is responsive to people’s needs, said dow, professor of geography at the university of south carolina and principal investigator of the noaa-supported carolinas integrated sciences and assessments research team. in a region where lawmakers have suppressed reports of climate-change effects, dow said her team’s strategy is to “engage and support those responsible for public planning.” soliciting information from communities regarding exactly what kinds of measurements they need from scientists to deal with climate impacts is not always a straightforward process. “it has to be a con- versation,” said fellow jeffrey dukes, “and it isn’t an easy one.” in a meeting dukes helped convene in northwest indiana, he said community officials expressed a keen desire for information about flooding and record rains, but they were unclear whether they needed to know, for instance, how much rain fell in a half-hour or over two days in order to facilitate their planning. dukes, who is director of the purdue climate change research center, said a sort of interface had to be built between the scientists he works with and the community representatives, “to build foundational thinking on both sides in order to meet in the middle.” fellow melissa kenney, assistant research professor in environ- mental decision support science at the university of maryland, worked with students this sum- mer to synthesize the challenges of community representatives in the chesapeake bay by review- ing existing reports and other documents. “we wanted to do our homework so we didn’t ask a lot of questions that were already answered,” kenney said. she and her team are looking for “research gaps and science translation opportunities” to approach com- munity leaders with processes and tools that will “facilitate evidence- based decision-making.” “in a densely populated watershed, climate is amplifying existing impacts on our region,” added kenney. “so we don’t have a choice; we have to make decisions to consider current and future climate consequences.” as the director of the university of minnesota’s institute on the environment and the head of the university’s department of ecology, evolution and behavior, fellow jessica hellmann works with natural resource agencies and land managers to help them confront climate impacts in their conservation planning. at the same time, she and her colleagues are reaching out to sustainability directors at corpora- tions, who she said are “constantly seeking ways to make the sustain- ability case to corporate leadership.” dukes, for one, is hopeful that his team’s efforts will contribute to moving the public at large toward acknowledging and respond- ing to climate change. “climate impacts are happening, and it would be stupid not to consider them as we’re spending all kinds of money constructing infrastructure,” dukes said. “we’re going to save hoosiers money and strengthen our state, and hopefully that will encourage people to take climate change into account.” the aaas leshner leadership institute is now accepting applica- tions for from mid-career scientists working in infectious disease research. the application deadline is november . for more information, visit http://leshnerfellows.aaas.org. melissa kenney (center) on the shore of the chesapeake bay aaas leshner fellows help confront climate impacts a a a s n e ws & n o t e s published by aaas o n a p ril , h ttp ://scie n ce .scie n ce m a g .o rg / d o w n lo a d e d fro m http://science.sciencemag.org/ aaas leshner fellows help confront climate impacts michaela jarvis doi: . /science. . . ( ), . science article tools http://science.sciencemag.org/content/ / / permissions http://www.sciencemag.org/help/reprints-and-permissions terms of serviceuse of this article is subject to the is a registered trademark of aaas.sciencescience, new york avenue nw, washington, dc . the title (print issn - ; online issn - ) is published by the american association for the advancement ofscience copyright © , american association for the advancement of science o n a p ril , h ttp ://scie n ce .scie n ce m a g .o rg / d o w n lo a d e d fro m http://science.sciencemag.org/content/ / / http://www.sciencemag.org/help/reprints-and-permissions http://www.sciencemag.org/about/terms-service http://science.sciencemag.org/ original article phase ii trial of pembrolizumab in patients with platinum refractory germ cell tumors: a hoosier cancer research network study gu - nabil adra , lawrence h. einhorn , sandra k. althouse , natraj r. ammakkanavar , dana musapatika , costantine albany , david vaughn , nasser h. hanna melvin & bren simon cancer center, indiana university school of medicine, indianapolis, indiana. department of biostatistics - indiana university school of medicine, indianapolis, indiana. hoosier cancer research network, indianapolis, indiana university of pennsylvania abramson cancer center, philadelphia, pennsylvania running head: pembrolizumab in refractory germ cell tumors keywords: testicular cancer; germ cell tumor; pembrolizumab; immunotherapy; checkpoint inhibitors corresponding author: dr. nabil adra email: nadra@iu.edu address: barnhill drive, rt , indianapolis, in telephone: - - ; fax: - - number of manuscript pages: total number of words in abstract: total number of words in text: , number of references: number of tables: number of figures: number of tables online only: number of figures online only: ___________________________________________________________________ this is the author's manuscript of the article published in final edited form as: adra, n., einhorn, l. h., althouse, s. k., ammakkanavar, n. r., musapatika, d., albany, c., … hanna, n. h. ( ). phase ii trial of pembrolizumab in patients with platinum refractory germ-cell tumors: a hoosier cancer research network study gu - . annals of oncology, ( ), – . https://doi.org/ . /annonc/mdx https://doi.org/ . /annonc/mdx conflict of interest statement: the authors have no conflicts of interest to report funding: merck & co. - presented in part at the rd annual meeting of the american society of clinical oncology (asco), chicago, june . poster discussion session. abstract background despite remarkable results with salvage standard-dose or high-dose chemotherapy about % of patients with relapsed germ-cell tumors (gct) are incurable. immune checkpoint inhibitors have produced significant remission in multiple tumor types. we report the first study of immunotherapy in patients with gct. patients and methods single arm phase trial investigating pembrolizumab mg iv q weeks until disease progression in patients with relapsed gct and no curable options. patients age≥ with gct who progressed after first-line cisplatin-based chemotherapy and after at-least salvage regimen (high-dose or standard-dose chemotherapy) were eligible. centrally assessed programmed death-ligand (pd-l ) on tumor and infiltrating immune cells was scored. primary endpoint was overall response rate (orr) using immune-related response criteria. simon two-stage design with type i error % and power % was utilized. results male patients were enrolled. median age . all patients had non-seminoma. primary site was testis ( ) or mediastinum ( ). median afp (range, - , ) and hcg (range, . - , ). patients had late relapse (> years). median number of previous chemotherapy regimens was . patients received prior high-dose chemotherapy. patients had positive pd-l staining (h-score and ). median number of pembrolizumab doses was (range, - ). there were grade adverse events. no immune-related adverse events were reported. no partial or complete responses were observed. patients achieved radiographic stable disease for and weeks, respectively; both had continued rising afp level despite radiographic stability and had negative pd-l staining. conclusion this is the first reported trial evaluating immune checkpoint inhibitors in gct. pembrolizumab is well tolerated but does not appear to have clinically meaningful single-agent activity in refractory gct. clinical trial information:nct . key message in this phase ii trial, pembrolizumab was well tolerated but did not appear to have single agent activity in refractory germ cell tumors. two patients had stable disease while the rest had progressive disease as best response. programmed death-ligand (pd-l ) staining was positive in patients. introduction germ-cell tumors (gct) are remarkably chemosensitive and up to % of patients with metastatic disease will be cured with cisplatin-based combination chemotherapy.[ ] patients who relapse after initial chemotherapy can still be cured with salvage therapy including salvage surgery, standard dose chemotherapy, or high-dose chemotherapy plus peripheral-blood stem cell transplant (pbsct).[ - ] despite the high cure rates with frontline and salvage chemotherapy, there remains a - % cohort of patients with metastatic gct who are incurable with the current therapeutic options. novel therapeutic approaches are needed for these patients. monoclonal antibodies against programmed death (pd- ) and its ligands (pd-l and pd-l ) have demonstrated robust activity and manageable toxicity in many advanced tumors including melanoma, lung, kidney, urothelial, and other malignancies.[ - ] pembrolizumab is a highly selective, humanized monoclonal igg κ isotype antibody against pd- which can disrupt the engagement of pd- with its ligands and impede inhibitory signals in t cells. frequent pd-l expression has been reported in tumor samples from testicular cancer patients suggesting that these patients could potentially benefit from immunotherapy approaches with pd- or pd-l inhibition.[ ] in addition, a prognostic value has been suggested for pd-l expression on testicular gct indicating that patients with high pd- l expression were more likely to have poor clinical features and worse survival outcomes.[ ] case series have been reported regarding possible activity of pd- inhibitors in combination with chemotherapy in a small sample of patients with platinum refractory gct.[ ] immune therapy with pembrolizumab is a novel approach for salvage in patients with metastatic gct. in a multicenter single-arm open-label phase ii trial conducted within the hoosier cancer research network, we evaluated the efficacy and safety of pembrolizumab in patients with refractory gct with no further curative treatment options. patients and methods patients eligible patients had histologically confirmed metastatic gct (seminoma or non- seminoma histology) who progressed after first-line cisplatin-based chemotherapy and after at-least one salvage regimen: standard-dose chemotherapy or high-dose chemotherapy plus peripheral-blood stem-cell transplant. relapsed patients who are not eligible for salvage chemotherapy (e.g. late relapse of disease) were also included in this study. an eastern cooperative oncology group (ecog) performance status score of or was required (on a -point scale, with indicating no symptoms and higher numbers indicating greater disability). patients had measurable metastatic disease according to the response evaluation criteria in solid tumors (recist), version . and/or elevation of tumors markers (alpha-fetoprotein [afp] and/or β-human chorionic gonadotropin [β-hcg]).[ ] if a rising tumor marker was the only evidence of progressive disease, consecutive rising values at least one week apart were required. patients with treated brain metastases were eligible. treatment with systemic corticosteroids or immunosuppressants within days of first trial treatment was not permitted. all patients were required to provide archived tumor tissue from a biopsy or the original primary tumor for pd-l expression evaluation and other exploratory analyses. patients were eligible irrespective of pd-l staining result. patients were ineligible if they had received anti-pd- or anti-pd-l therapy previously. full eligibility criteria are listed in the trial protocol, available online with the full text of this article. pretreatment evaluation pretreatment evaluation included complete history and physical examination, performance status, complete blood count, comprehensive metabolic profile, serum tumor markers (afp and hcg), creatinine clearance, electrocardiogram, and computed tomography (ct) scan of the chest, abdomen, and pelvis. treatment program planned treatment consisted of pembrolizumab mg administered intravenously (iv) every weeks. pembrolizumab was provided by merck & co., the supporter of this investigator initiated phase ii trial. treatment was continued until documented disease progression, development of unacceptable level of toxic effects, withdrawal of consent, decision by investigator to discontinue treatment, or the completion of weeks of pembrolizumab therapy. evaluation of response and toxicity patients were assessed on day of each -week cycle. a complete blood count, complete metabolic panel, and tumor markers were determined on day of each cycle. toxicities were graded according to the national cancer institute common terminology criteria for adverse events, version . . tumor imaging was performed at baseline followed by every weeks for the initial weeks. subsequently, imaging was spaced to every weeks until discontinuation of study drug. the full assessment schedule is provided in the trial protocol. pd-l expression was assessed in formalin-fixed archived tumor samples at a central laboratory with the use of the commercially available pd-l ihc c assay. pd-l expression was scored as (no staining), (weak staining), (moderate staining), (intense staining) along with categorized pd-l combined positive score, defined as the percentage of pd-l -expressing tumor and infiltrating immune cells relative to the total number of tumor cells (h-score). endpoints and statistical analysis the primary endpoint was overall response rate (orr) using immune related response criteria (irrc).[ ] secondary endpoints were orr using recist version . , to assess toxicity and tolerability of pembrolizumab, and to estimate the duration of disease response.[ ] exploratory objectives included assessment of prevalence of pd- l expression in tumor tissue and correlate with tumor response to treatment. efficacy was assessed in the intention-to-treat population, which included all the patients who were assigned to the treatment group. safety was assessed in the as-treated population, which included all the patients who received at least one dose of study treatment. a simon two-stage optimal design with type i error rate of % and power of % was utilized.[ ] the null hypothesis was an orr of ≤ % and the alternate hypothesis was orr ≥ %. consequently, subjects were enrolled in the first stage. if no responses are determined in the initial stage, the study will be concluded. if at least patient achieved a partial or complete response, then the treatment will be considered worthy of further investigation and more subjects will be accrued in the second stage for a total sample size of subjects. if there are or more subjects with partial or complete response, then the treatment regimen would be considered a success. statistical analyses were performed using sas software, version . . the data cutoff was february . the study schema is depicted in figure . results patient and disease characteristics twelve patients were enrolled between march and october . patient characteristics are provided in table . median age was years (range, to years) and all patients were male. primary tumor sites was testis in patients ( %) and mediastinal non-seminoma in patient ( %). all patients had non-seminoma histology. tumor markers were elevated in all patients: afp only in , hcg only in , both afp and hcg were elevated in patients. metastatic sites included retroperitoneal lymph nodes in patients, pulmonary metastasis in patients, liver metastasis in patients, brain and bone metastasis in patient each. the median number of previous chemotherapy lines of treatment was (range, to ). six patients ( %) received prior high-dose chemotherapy and autologous stem-cell transplant as a previous line of therapy. ten patients had received prior ifosfamide or paclitaxel-containing salvage chemotherapy regimens. six patients ( %) had late relapse of disease defined as disease that relapsed more than years after first-line cisplatin-based combination chemotherapy. treatment administration all patients enrolled received at least one dose of pembrolizumab mg iv per study protocol. the median number of pembrolizumab doses was (range, - ). two patients received only dose of pembrolizumab and had clear clinical disease progression at the time of the scheduled second dose. eight patients received doses of pembrolizumab and had disease progression at the week interim evaluation. one patient received doses and another patient received doses of pembrolizumab. efficacy no partial or complete responses were observed. two patients achieved stable disease for and weeks, respectively. both these patients had continued rising afp values on treatment despite radiographic stability. one of these patients had late relapse with no response to previous salvage standard-dose chemotherapy (paclitaxel-ifosfamide- cisplatin [tip] and paclitaxel-gemcitabine) before enrolling on this clinical trial. the second patient with radiographic stability did not have late relapse of disease and had slow progression after salvage standard-dose chemotherapy with tip and gemcitabine- oxaliplatin. ten patients had progressive disease as their best response. none of the patients on study had tumor marker decline while on treatment with pembrolizumab. per study protocol, at least objective response was required to continue enrolling on the second stage of the study; therefore, the trial was closed at the completion of the first stage. at last follow-up, of the patients enrolled had died of disease progression. biomarker analysis archival formalin-fixed paraffin-embedded tissue samples were available for correlative studies in all patients. correlative studies were performed on samples from primary testicular tumor in patients, retroperitoneal lymph nodes in patients, bone metastasis biopsy sample in patient, and lung metastasis biopsy sample in patient. immunohistochemistry staining of archival tumor samples was positive for pd-l expression in patients: one patient with predominant embryonal carcinoma histology had + staining in % of tumor cells formulating an h-score of . another patient with predominant choriocarcinoma histology had + staining on %, + staining on %, and + staining on % of tumor cells formulating an h-score of . both these patients had progressive disease as their best response to therapy. toxicity pembrolizumab was tolerated by this patient population with patients experiencing grade adverse events. four adverse events were possibly related to disease or study treatment. grade or higher and most commonly occurring adverse events are listed in table . no grade or toxicities were observed. there were no immune-related adverse events observed. discussion in the current era, cisplatin-based combination chemotherapy will cure > % of patients with metastatic gct.[ ] % of patients with international germ-cell cancer collaborative group (igcccg) good-risk disease will achieve cure with front-line chemotherapy. patients with intermediate and poor-risk disease have less favorable outcomes and a significant proportion will relapse and require salvage therapy.[ ] high-dose chemotherapy plus pbsct achieves cures in % of patients with relapsed metastatic gct.[ , , ] patients who relapse after salvage standard-dose or high- dose chemotherapy have poor outcomes with no further curative options short of salvage surgery in patients with anatomically localized disease.[ , ] understanding the mechanism(s) of resistance to treatment are desperately needed. unfortunately, early studies with molecularly targeted therapies such as imatinib, sunitinib, thalidomide, and trastuzumab have yielded negative results.[ - ] more recently, studies with pazopanib have provided some signal of activity.[ ] immunotherapy with pd- and pd-l inhibitors has demonstrated remarkable results in a variety of advanced solid malignancies after progression on standard-of-care treatment.[ - ] preclinical investigations have shown that pd-l expression was present in % of all seminomas and % of non-seminomas.[ ] our study represents the first clinical trial investigating immunotherapy with checkpoint inhibitors as a novel approach in patients with platinum-refractory incurable gct. pembrolizumab was tolerated in this study with no immune-related adverse events reported. unfortunately, this clinical trial did not show clinical activity for pembrolizumab in patients with metastatic gct who progressed after salvage chemotherapy. no objective responses were observed among the patients enrolled; although patients had radiographic stable disease for and weeks respectively, both had continued rising afp level during treatment suggesting continued treatment resistance. pd-l expression was present on tumor samples from only patients in this study. both these patients had progressive disease as their best response. although this phase ii trial investigated a novel approach for salvage in refractory gct, there are limitations to report. the small cohort of patients limits the generalizability. other studies are underway investigating single agent or combination checkpoint inhibition in patients with metastatic gct. immunohistochemistry staining for pd-l was performed on archival tumor tissues which may not represent the current immune status or the tumor microenvironment at the time of treatment with checkpoint inhibitor. in addition, pd-l expression from tumor metastatic sites was not performed and this might also be informative. the expression of pd-l is known to be dynamic and is regulated by extrinsic signaling such as release of interferon-γ by immune cells, loss of expression of tumor suppressor genes, or activation of the akt-mtor pathway.[ , ] moreover, this study was not a biomarker driven study since all patients were eligible irrespective of pd-l expression. higher levels of pd-l expression have been associated with robust responses in certain advanced solid tumors such as melanoma and lung cancer but not in others.[ , ] in addition, the protocol for this study mandated response assessment at weeks, i.e. after infusions of pembrolizumab at which time most patients had progressive disease. this might have been a short interval given our understanding that the immune checkpoint inhibitors have a median time to response of ≥ months.[ , ] nevertheless, patients on this study had clinical unequivocal rapid progression while on study treatment. even though no immune- related adverse events were observed, the short duration of exposure to pembrolizumab with a median number of deliverable doses of might influence that analysis. in conclusion, this is the first reported clinical trial evaluating immune checkpoint inhibitors in testicular cancer. single agent pembrolizumab did not demonstrate clinical benefit in this cohort of patients with refractory gct. further investigation requires evaluation of the mechanism of resistance and possible combination therapy. acknowledgements: none references . hanna nh, einhorn lh. testicular cancer--discoveries and updates. n engl j med ; : - . . einhorn lh, williams sd, chamness a et al. high-dose chemotherapy and stem-cell rescue for metastatic germ-cell tumors. n engl j med ; : - . . feldman dr, sheinfeld j, bajorin df et al. ti-ce high-dose chemotherapy for patients with previously treated germ cell tumors: results and prognostic factor analysis. j clin oncol ; : - . . kondagunta gv, bacik j, donadio a et al. combination of paclitaxel, ifosfamide, and cisplatin is an effective second-line therapy for patients with relapsed testicular germ cell tumors. j clin oncol ; : - . . loehrer pj, sr., einhorn lh, williams sd. vp- plus ifosfamide plus cisplatin as salvage therapy in refractory germ cell cancer. j clin oncol ; : - . . loehrer pj, sr., gonin r, nichols cr et al. vinblastine plus ifosfamide plus cisplatin as initial salvage therapy in recurrent germ cell tumor. j clin oncol ; : - . . murphy br, breeden es, donohue jp et al. surgical salvage of chemorefractory germ cell tumors. j clin oncol ; : - . . bellmunt j, de wit r, vaughn dj et al. pembrolizumab as second-line therapy for advanced urothelial carcinoma. n engl j med . . motzer rj, escudier b, mcdermott df et al. nivolumab versus everolimus in advanced renal-cell carcinoma. n engl j med ; : - . . reck m, rodriguez-abreu d, robinson ag et al. pembrolizumab versus chemotherapy for pd-l - positive non-small-cell lung cancer. n engl j med ; : - . . robert c, long gv, brady b et al. nivolumab in previously untreated melanoma without braf mutation. n engl j med ; : - . . fankhauser cd, curioni-fontecedro a, allmann v et al. frequent pd-l expression in testicular germ cell tumors. br j cancer ; : - . . cierna z, mego m, miskovska v et al. prognostic value of programmed-death- receptor (pd- ) and its ligand (pd-l ) in testicular germ cell tumors. ann oncol ; : - . . zschabitz s, lasitschka f, jager d, grullich c. activity of immune checkpoint inhibition in platinum refractory germ-cell tumors. ann oncol ; : - . . eisenhauer ea, therasse p, bogaerts j et al. new response evaluation criteria in solid tumours: revised recist guideline (version . ). eur j cancer ; : - . . wolchok jd, hoos a, o'day s et al. guidelines for the evaluation of immune therapy activity in solid tumors: immune-related response criteria. clin cancer res ; : - . . simon r. optimal two-stage designs for phase ii clinical trials. control clin trials ; : - . . ko jj, bernard b, tran b et al. conditional survival of patients with metastatic testicular germ cell tumors treated with first-line curative therapy. j clin oncol ; : - . . adra n, ku k, kalra m et al. survival outcomes of patients with metastatic germ cell tumor (mgct) treated from to : the indiana university (iu) experience. in. j clin oncol , (suppl s; abstr ). . adra n, abonour r, althouse sk et al. high-dose chemotherapy and autologous peripheral- blood stem-cell transplantation for relapsed metastatic germ cell tumors: the indiana university experience. j clin oncol ; jco . . cary c, pedrosa ja, jacob j et al. outcomes of postchemotherapy retroperitoneal lymph node dissection following high-dose chemotherapy with stem cell transplantation. cancer ; : - . . kollmannsberger c, pressler h, mayer f et al. cisplatin-refractory, her /neu-expressing germ- cell cancer: induction of remission by the monoclonal antibody trastuzumab. ann oncol ; : - . . rick o, braun t, siegert w, beyer j. activity of thalidomide in patients with platinum-refractory germ-cell tumours. eur j cancer ; : - . . feldman dr, turkula s, ginsberg ms et al. phase ii trial of sunitinib in patients with relapsed or refractory germ cell tumors. invest new drugs ; : - . . einhorn lh, brames mj, heinrich mc et al. phase ii study of imatinib mesylate in chemotherapy refractory germ cell tumors expressing kit. am j clin oncol ; : - . . necchi a, lo vullo s, giannatempo p et al. pazopanib in advanced germ cell tumors after chemotherapy failure: results of the open-label, single-arm, phase pazotest trial. ann oncol ; : - . . song m, chen d, lu b et al. pten loss increases pd-l protein expression and affects the correlation between pd-l expression and clinical parameters in colorectal cancer. plos one ; : e . . lastwika kj, wilson w, rd, li qk et al. control of pd-l expression by oncogenic activation of the akt-mtor pathway in non-small cell lung cancer. cancer res ; : - . figure legends - fig : study schema table . patient and disease characteristics characteristic no. of patients % total patients median age (range) ( - ) location of primary tumor • testis • retroperitoneum • mediastinum % % % tumor histology • seminoma • non-seminoma % % predominant histology • choriocarcinoma • embryonal carcinoma • teratoma • yolk sac tumor % % % % metastatic site(s) • retroperitoneum • pulmonary • npvm -liver metastasis -brain metastasis* -bone metastasis* % % % % % % no. of previous chemotherapy regimens • • • • • • % % % % % % late relapse (> years) % elevated tumor markers • afp only • hcg only • afp and hcg % % % median serum afp ng/ml (range) ( - , ) median serum hcg miu/ml (range) ( . - , ) ecog performance status • • % % abbreviations: npvm, non-pulmonary visceral metastasis; afp, alpha fetoprotein; hcg, human chorionic gonadotropin; iu, international unit; ecog, eastern cooperative oncology group; *brain/bone imaging was not mandatory table . common and grade or higher adverse events observed with pembrolizumab therapy adverse event grade grade grade percent fatigue % nausea % vomiting % abdominal pain % diarrhea % skin and subcutaneous tissue disorders * % constipation % cough % dry skin % dyspnea % edema limbs % flank pain % flu-like symptoms % musculoskeletal and connective tissue disorder % non-cardiac chest pain % anemia % hyperglycemia % sciatic pain % *erythema at patch site fig . 	adra_ _phase 	table 	table 	figure i n d i a n a m a g a z i n e o f h i s t o r y sideration of american utopianism at its enigmatic word. on some level, the persuasiveness of storey’s argument depends on his scrupulously close reading: his eye catches everything, down to the me- chanical gait of a plow horse laboring under the combined metaphorical weight of both garden-variety pastoral- ism and incipient industrial capitalism. but the book also functions as well as it does because of storey’s innovative and unexpected take on the subject of american modernity, an approach that simultaneously challenges and corroborates much of what we already claim to know. on one hand, it is hardly obvious that novels about late nineteenth-century rural and small- town life would prove to be “a vital entry point to the themes, concerns, currents, and transformations of urban modernity,” as storey persua- sively demonstrates (p. ). but then again, how could any artifact from the gilded age have remained innocent of modernity? the answer is that none did, and rural fictions, urban realities demonstrates this beautifully. ida mckinley: the turn-of-the-century first lady through war, assassination, and secret disability by carl sferrazza anthony (kent, ohio: kent state university press, . pp. . notes, bibliography, index. $ . .) comparative analyses of first ladies routinely dismiss the influence of ida saxton mckinley ( - ), por- traying her as inconsequential during her husband william mckinley’s ten- ure as president from until his assassination in . while mystery shrouded her medical condition, the readers with a particular interest in indiana history will appreciate storey’s thoughtful engagement with a number of hoosier authors, including edward eggleston, booth tarkington, and maurice thompson, all of whom receive far less attention from serious critics than they deserve. but even readers who come to his book with a specific agenda will admire the sheer number of american authors storey is able to draw together by way of his adroit analysis. in sum, rural fic- tions, urban realities is a smart book, and one that is argued with an uncom- mon degree of care and sensitivity—a decidedly modern reconsideration of a misunderstood body of literature that has remained at the periphery of debates about american modernity. colin r. johnson is associate profes- sor of gender studies and adjunct associate professor of american studies, history and human biology at indiana university bloomington. he is the author of just queer folks: gender and sexuality in rural america ( ). reviews public seemed to accept an invalid as first lady. in producing the first full-length biography of ida mckin- ley, carl sferrazza anthony provides insight to this puzzling response in a riveting account of a victorian love affair, tragedy and scandals, medical misunderstandings, and presidential politics. above all, anthony depicts ida mckinley as a well-educated woman instilled with progressive views on gender and race. influ- enced by her work in banking and travel abroad, ida mckinley exhibited passion for the rights of working women, access to higher education, and woman suffrage. her intelligence and strong independence matched the ambitions of william mckinley, and a strong mutual dependency forged by marriage defined their political life. as anthony points out, both ida and william mckinley were at first fully synchronized in their motives and ambitions with a mutual under- standing of the importance of image in politics. as tragedies and scandal took their toll on ida’s mental and physical stamina, however, her enthusiasm for public life waned. william’s ambition ultimately superseded his wife’s, with dire consequences for how he man- aged her epilepsy. on the one hand, his devotion to his wife was deemed virtuous, if staged at times; yet, by giv- ing her a steady stream of bromides, in part to conceal the true nature of her illness, he inadvertently harmed her. piecing together ida mckinley’s long-term symptoms and treatments, anthony weaves a story within the biography about nineteenth-century misunderstandings and prejudices related to mental illnesses. anthony suggests that the treatments for her partial paralysis and mild epi- lepsy help to explain the personality changes and the escalation of her physical and mental impairment. much of this treatment was solicited and administered in secrecy, in fear of public revelation of a diagnosis of epilepsy which carried an immoral connotation. president mckinley desperately tried to contain the effect of his wife’s illness on his political career. the image of victorian mar- tyr often worked to his favor, yet at other times he went to great lengths to create the appearance of his wife as a more active first lady. in spite of her epilepsy, ida mckin- ley, as anthony shows, displayed ambition and a keen understanding of politics. she read and listened to all of her husband’s speeches; cen- tered herself prominently at official functions; influenced appointments and patronage; and used her love of knitting to bestow charitable aid. while surrogate hostesses, including nieces, filled in during her absences, the office of first lady was adapted to accommodate ida’s disability. on multiple occasions, her business sense provided valuable insight, even sparing her husband from the fallout of potential scandal. she also trav- eled with her husband, campaigned, and advocated for both suffrage and equal access to education for all. as first lady, ida mckinley also in- i n d i a n a m a g a z i n e o f h i s t o r y vited singers and actors to the white house, raising the legitimacy of the performing professions. anthony summons a wide variety of historical evidence to demon- strate the political influence of ida mckinley’s ambitions and ideas. her epilepsy egregiously mistreated, she became a victim of that same ambi- tion, as both she and her husband attempted to manipulate public opin- ion. appropriately, anthony raises even more questions about the true ida mckinley, but his work assures that she can no longer be dismissed as inconsequential. laura van assendelft is professor of political science at mary baldwin college where she researches and publishes in the field of women and politics, including first ladies. steel barrio: the great mexican migration to south chicago, - by michael innis-jiménez (new york: nyu press, . pp. . illustrations, notes, bibliography, index. clothbound, $ . ; paperbound, $ . .) histories of mexican migrants to the united states have proliferated since the s, but only in the last decade have we seen any that dealt with the chicago area. michael innis-jiménez’s steel bar- rio is only the third such monograph published on this topic. such paucity is lamentable—as early as almost , persons of mexican origin lived in the greater chicago area, includ- ing its industrial suburbs in indiana. those interested in chicago’s mexican community have had to rely on journal articles, dissertations, theses, or mono- graphs from other disciplines. steel barrio focuses on the early formation of chicago’s mexican immigrant community with an em- phasis on south chicago, an area that housed thousands of industrial workers occupied mainly in steel pro- duction. mexican workers shared the district with eastern and southern european immigrant workers who had arrived at least a decade earlier and did not welcome them. a ra- cial pecking order emerged, placing mexicans with african americans at the bottom. the author posits that in order to survive their hostile reception, south chicago’s mexicans resorted to a system of survival with interlocking institutions and cultural modes which “included actions that they might not have consciously con- sidered acts of resistance” (p. ). perhaps his most interesting assertion is that cultural survival efforts from the s immigrant generation were passed on to the second generation in the s, thereby assuring that assimilation did not occur. panamerican journal of trauma, critical care & emergency surgery, volume issue (may–august )vi dedication . /pajt- - -vi es con suma pena y tristeza recibir la noticia del fallecimiento de nuestro colega y gran amigo, gerardo “gerry” gómez ocurrido el pasado seis de agosto. pasado presidente de la sociedad panamericana del trauma en el y luego director ejecutivo por seis años, gerry siempre fue un pilar de nuestra sociedad desde sus comienzos y trabajó con gran esfuerzo y dedicación para su éxito. dr. gómez nació en la ciudad de barquisimeto en venezuela. Él siempre se caracterizó por ser una persona muy dinámica y emprendedora, atributos que lo encaminaron a su carrera en la medicina. gerry curso su escuela de medicina en la universidad de carabobo en venezuela para luego continuar y finalizar su residencia en cirugía en la universida de massachusetts en amherst. en el , se unió a la facultad de la universidad de miami donde ejerció un papel fundamental en el desarrollo del servicio de trauma y eventual creación del ryder trauma center, uno de los principales centros de trauma de la nación. en miami, dr. gómez fue parte del entrenamiento de una gran cantidad de profesionales, especialmente futuros cirujanos y cirujanas de trauma que actualmente ejercen su profesión a través de todo el país. siempre será recordado por todos por su pasión y dedicación a sus pacientes a los que siempre se refería como las damas y los caballeros del trauma. ante la oportunidad y desafío de desarrollar otro programa de trauma, gerry acepto la posición de jefe de trauma del wishard hospital como parte de la facultad de la universidad de indiana en indianápolis en . por años, dr. gómez no solo ayudo a desarrollar el centro de trauma si no que además fomento la educación profesional en la institución a través de los cursos como el atls y el atom. gerry será honrado y recordado con respeto y admiración por esta institución por sus logros y contribuciones a través de sus años de servicio. en el plano personal, gerr y era un afanado piloto que estuvo ejerciendo sus destrezas hasta hace unos meses atrás. era gran conocedor de la historia de la aviación, especialmente de la segunda guerra mundial. se caracterizaba por su entusiasmo del buen vino y siempre compartía un buen malbec argentino, su favorito, con la aromática delicadeza de un buen romeo y julieta, su cigarro preferido. al dr. gómez le sobreviven su esposa adrienne, cinco hijos, una hija, cinco nietos y nietas y dos bisnietos. aunque en su retiro ya no intervenía quirúrgicamente, siempre mantuvo su entusiasmo en la educación en trauma. tomaba con gran fervor y empeño su asistencia y participación en la enseñanza del curso del atls en kenya. además, a través de toda su carrera, siempre consideró como una misión y compromiso participar y trabajar activamente con los cirujanos de su amada venezuela en el desarrollo de sistemas y hospitales de trauma y en la educación de futuros cirujanos y cirujanas de trauma. luego de su retiro en el , dr. gómez trabajo en el restaurante de uno de sus hijos, gómez bbq, como manejador del negocio en indianápolis. fue un gran fanático de los deportes y siempre se le podía apreciar su asistencia en los juegos de futbol americano con los colts de indianápolis y del baloncesto con los hoosiers de la universidad de indiana. tampoco podía faltar su entusiasmo por las carreras de automóviles que presenciaba cada año en las millas de indianápolis. gerry y su esposa adrianne eran afamados viajeros del mundo, pero principalmente les encantaba visitar cada vez que tenían una oportunidad su segunda residencia en su amada ciudad de miami. gerry siempre será recordado por su alegría, jovialidad, energía y los gratos momentos que compartimos juntos. un gran cirujano, mentor y profesor, dr. gómez siempre se caracterizó por su entrega a la sociedad panamericana del trauma. se ha marchado un querido amigo y maestro, descansa en paz, gerry. te extrañaremos profundamente. dr. gómez junto al dr. feliciano y el dr. hoyt gerardo “gerry” gómez panamerican journal of trauma, critical care & emergency surgery, volume issue (may–august ) vii it is with great sadness that we receive the news of the death of our colleague and great friend, gerardo “gerry” gómez, which occurred on august th. past president of the pan american trauma society in and then executive director for six years, gerry was always a pillar of our society from its inception and worked with great effort and dedication for its success. dr. gómez was born in the city of barquisimeto in venezuela. he was always characterized by being a very dynamic and aspiring person, attributes that led him to his career in medicine. gerry went to the university of carabobo medical school in venezuela and then continued and finished his residency in surgery at the university of massachusetts in amherst. in , he joined the faculty of the university of miami where he played a pivotal role in the development of the trauma service and eventual creation of the ryder trauma center, one of the nation’s leading trauma centers. in miami, dr. gómez was part of the training of a large number of professionals, especially future trauma surgeons, who currently practice their profession throughout the country. he will always be remembered by all for his passion and dedication to his patients whom he always referred to as the ladies and gentlemen of trauma. faced with the opportunity and challenge to develop another trauma program, gerry accepted the position of chief of trauma at wishard hospital as part of the faculty at indiana university in indianapolis in . for years, dr. gomez not only helped to develop the trauma center, but also strongly promoted professional education at the institution through courses such as atls and atom. gerry will be honored and remembered with respect and admiration by this institution for his accomplishments and contributions throughout his years of service. on a personal level, gerry was an avid pilot who had been exercising his skills until a few months ago. he was very knowledgeable about the history of aviation, especially from the world war ii. he was characterized by his enthusiasm for good wine and always shared a good argentine malbec, his favorite, with the aromatic delicacy of a good romeo y julieta, his cigar of choice. dr. gomez is survived by his wife adrianne, five children, four sons and one daughter, five grandchildren and two great-grandchildren. although he was no longer actively involved in surgery during his retirement, he always maintained his enthusiasm for trauma education. with great fervor and commitment, he treasured his attendance and participation in teaching the atls course in kenya. in addition, throughout his career, he always considered it a mission to actively participate and work with the surgeons of his beloved venezuela in the development of trauma hospitals and systems and in the education of future trauma surgeons. after his retirement in , dr. gómez worked at the restaurant of one of his sons, gómez bbq, as a business manager in indianapolis. he was a huge sports fan and attending the indianapolis colts football games and the indiana university hoosiers basketball games was his passion. he was also a great fan of auto races and really enjoyed attending the indianapolis every year. gerry and his wife adrianne were world travelers, but mainly they really enjoyed visiting their second home in their beloved city of miami whenever they had a chance. gerry will always be remembered for his joy, joviality, energy and the great moments we shared together. a great surgeon, mentor, and teacher, dr. gómez was characterized by his dedication to the panamerican trauma society. a dear friend and teacher has left us, rest in peace, gerry! he will be deeply missed. manuel lorenzo dr gerry gomez dedication a phase ii study with cetuximab and radiation therapy for patients with surgically resectable esophageal and ge junction carcinomas: hoosier oncology group g - journal of thoracic oncology ®  •  volume  , number  , november  introduction: on the basis of the promising activity of cetuximab and radiation therapy for head and neck cancers, we evaluated the efficacy of this regimen followed by surgery in patients with resect- able esophageal cancer. this was a phase ii, open-label, single-arm, multicenter study of patients with potentially resectable esophageal cancer. methods: patients received two weekly doses of cetuximab followed by weekly cetuximab combined with radiation therapy for weeks. after a - to -week rest, patients’ primary tumor was resected. the main objective was to evaluate pathologic complete response (pcr) rate in the primary tumor after cetuximab and radiation therapy. results: thirty-nine patients completed the study. most patients were men ( %), median age was years, performance status was to ( %), patients had a histology of adenocarcinoma ( %), and tumors were located in the esophagus ( %). grade toxicities in more than % of patients included dysphagia ( %), anorexia and dehydration ( %), and dyspnea, fatigue, hypernatremia ( %). grade aspiration occurred in % ( patient). four patients died, two from disease progression, one from aspiration pneumonia postsurgery, and one from septic shock. thirty-one patients ( %) underwent esopha- gectomy. the pcr rate was . % by intention-to-treat and % for patients who underwent esophagectomy. the pcr by histology was of ( %) for squamous cell carcinomas and of ( %) for adenocarcinoma. earlier-stage disease was associated with increased pcr (iia %, iib %, iii %). conclusions: cetuximab and radiation therapy results in a pcr rate that seems at least comparable with that of chemotherapy and radia- tion therapy. this regimen may be better tolerated than preopera- tive chemotherapy and radiation therapy in patients with resectable esophageal cancers. (j thorac oncol. ; : – ) key words: esophageal, cetuximab, radiation therapy, resectable. esophageal carcinomas in affected , americans, resulting in an estimated , deaths. one standard approach for the management of patients with resectable esophageal cancer involves trimodality therapy that includes chemotherapy with concurrent radiation therapy followed by surgical resection of the primary lesion with improved long- term survival but marginal tolerance to treatment. , patients who cannot have surgical resection of the esophagus cancer can be treated with combined chemotherapy and radiation therapy with superior survival outcomes similar to those with radiation alone. the epidermal growth factor receptor (egfr) signal- ing pathway is an important target in esophageal cancers, and high egfr expression by immunohistochemical score in the tumor of patients with resected esophageal adenocar- cinomas after trimodality therapy predicts a poor outcome. cetuximab is a monoclonal antibody that binds to the egfr and competitively inhibits the binding of epidermal growth factor and other ligands blocking phosphorylation and activa- tion of receptor-associated kinases. egfr inhibition results in gap cell-cycle arrest, and radiation results in gap cell- cycle arrest, gap and gap being part of the interphase segment of cell growth. in preclinical models, egfr inhi- bition enhances radiation-induced apoptosis, inhibits repair of radiation-induced dna damage, and inhibits formation of new blood vessels. the radio-sensitizing effect of cetux- imab has already been tested in patients with squamous cell copyright © by the international association for the study of lung cancer issn: - / / - a phase ii study with cetuximab and radiation therapy  for patients with surgically resectable esophageal and  ge junction carcinomas hoosier oncology group g - carlos r. becerra, md,* nasser hanna, md,† andrew david mccollum, md,* naftali becharm, md,‡ robert d. timmerman, md,§ michael dimaio, md,§ kenneth a. kesler, md,║ menggang yu, phd,║ tong yan, phd, ms,║ and hak choy, md§ *us oncology research, mckesson specialty health, the woodlands, texas, and texas oncology-baylor sammons cancer center, dallas, texas; †indiana university melvin and bren simon cancer center, indianapolis, indiana; ‡community oncology center, kokomo, indiana; §university of texas southwestern medical center at dallas, dallas, texas; and ║indiana university school of medicine, indianapolis, indiana. disclosure: carlos r. becerra is on the speakers’ bureau for eli lilly and company. a.d. mccollum is a consultant for genentech. hak choy is chairman of radiation oncology at ut southwestern medical center and is a consultant for bristol-meyers-squibb and eli lilly. all the other authors declare no conflict of interest.abstract part of this study (abstract no. ) was published at the american society of clinical oncology annual meeting (j clin oncol ; : – ). address for correspondence: carlos r. becerra, md, texas oncology, baylor sammons cancer center, worth street, dallas texas . e-mail: carlos.becerra@usoncology.com original article mailto:carlos.becerra@usoncology.com copyright © by the international association for the study of lung cancer becerra et al. journal of thoracic oncology ®  •  volume  , number  , november  carcinomas of the head and neck, with improvement in local control, progression-free and overall survival over radiation therapy alone. herein, we present the results of a single-arm, open- label, phase ii study with cetuximab and external beam radia- tion therapy for patients with newly diagnosed, potentially resectable esophageal and gastroesophageal (ge) junction carcinomas, with the intent of improving efficacy and decreas- ing toxicity over current standard treatment approaches. materials and methods this was a phase ii, open-label, single-arm, multicenter study that accrued patients with potentially resectable esopha- geal cancer. the study was conducted in accordance with the declaration of helsinki and is consistent with international conference on harmonisation good clinical practice guide- lines. the protocol was approved by the local institutional review boards/ethics committees, and all patients provided a written informed consent. the study was coordinated by the hoosier oncology group clinical trials working group (indianapolis, in). the primary objective of the study was to evaluate the pathologic complete response (pcr) rate in the primary tumor after cetuximab and radiation therapy. secondary objectives were to evaluate the clinical complete and partial response rates, and overall toxicities. patients key inclusion criteria were the following: age more than years; eastern cooperative oncology group (ecog) performance status of to ; pathologic diagnosis of squa- mous cell carcinoma or adenocarcinoma of the esophagus or ge junction; clinical stages iia, iib, iii, or iva with celiac node involvement; surgical candidates as determined by a surgical consult; patient being agreeable to surgical resec- tion of the primary tumor; no prior use of radiation or che- motherapy for cancer of the esophagus or ge junction; no prior therapy directed against the egfr pathway; no major surgeries within days of registration for protocol therapy; no other active malignancies; no history of uncontrolled car- diac disease; no history of interstitial pneumonitis or pul- monary fibrosis; and adequate marrow and organ function including absolute neutrophil count more than mm ; platelet count more than , mm , hemoglobin more than g/dl, creatinine less than two times the upper limit of normal (uln), bilirubin less than . times uln, aspar- tate transaminase (sgot), and/or alanine aminotransferase (sgpt) less than . times uln; no prior severe infusion reaction to a monoclonal antibody; and use of an effective method of contraception. treatment all patients received premedication with dyphenhydr- amine hydrochloride at an initial dose of mg by intrave- nous infusion given to minutes before the first dose of cetuximab. premedication with dyphenhydramine hydrochlo- ride for subsequent doses of cetuximab was modified at the investigators’ discretion. the initial dose of cetuximab was administered at mg/m intravenously over minutes followed by weekly infusions of cetuximab at mg/m over min- utes. cetuximab was administered weekly for two doses. radiation therapy was started after the second dose of cetuximab, and the combination of once-weekly cetuximab and radiation therapy was continued until the completion of radiation therapy. total weekly cetuximab doses equaled eight. external beam radiation therapy was started on week of cetuximab infusion to a total dose of . gy divided into . gy per fraction per day. megavoltage equipment was required with effective photon energies at more than mv. twice weekly, the verification films of orthogonal views were reviewed by the treating physicians. all study patients had a planned surgical resection of the primary tumor and adjacent mediastinal and/or celiac lymph nodes after satisfactory hematologic and functional recovery within weeks of completion of radiation therapy. lymph node staging performed at the time of laparoscopy or thora- coscopy was repeated at the time of resection. rather than simply sampling a site-representative node from each level, we removed all technically accessible lymph nodes. assessment efficacy patients had a baseline computed tomography scan of the chest, abdomen, and pelvis with repeat imaging performed after cetuximab and radiation therapy to exclude interim development of metastatic disease. patients had a baseline upper endoscopy with ultrasound, pulmonary function tests, and tumor sampling. clinical staging was reported according to the american joint commission on cancer criteria ver- sion . at the time of surgery, an evaluation of the primary sample was performed to determine complete pathologic response (defined as the absence of tumor cells in the resected specimen). safety and tolerability a safety evaluation was performed at baseline and then weekly for the duration of treatment. patients were followed for weeks after surgical resection or until resolution of treatment-related toxicities. patients underwent a physical examination, vital signs, ecog performance status, complete blood count, and chemistries as part of the safety evaluation. toxicities were assessed according to the national cancer institute common terminology criteria for adverse events version . , and this included assessment of dysphagia as follows: grade : none grade : mild dysphagia but can eat a regular diet grade : dysphagia requiring predominantly liquids, pureed foods, or soft diet grade : dysphagia requiring a feeding tube, intravenous hydration, or parenteral hyperalimentation grade : complete obstruction (cannot swallow saliva); ulceration with bleeding not induced by minor trauma or abrasion or perforation. copyright © by the international association for the study of lung cancer journal of thoracic oncology ®  •  volume  , number  , november  esophageal and ge junction carcinomas adverse events not covered by the national cancer institute common terminology criteria for adverse events criteria were graded according to a three-point system: mild, moderate, and severe. statistical methods the efficacy of the treatment was evaluated using simon’s two-stage minimax design where the endpoint vari- able was pcr. the null hypothesis was that the probability of pcr was . . the alternative hypothesis was that the pcr was . . a type i error of . and a type ii error of . were selected. a total of patients were enrolled in the first stage, and at least five pcrs were required to enroll an additional patients in the second stage. if the number of pcrs exceeded , then the alternative hypothesis would be accepted. during the first stage of the efficacy analysis, the impact of any nonhematologic toxicities exceeding grade would be assessed with two sets of stopping rules, one for grade toxici- ties and another for grade or grade toxicities. if the boundar- ies of grade or toxicities were reached (i.e., of patients) in the first phase of the trial, the study would be terminated. results the characteristics of the patients enrolled in the study are delineated in table . the majority of the patients were men ( %), median age was years, and they had an ecog performance status of to ( %); patients had a histology of adenocarcinoma ( %), and tumors were located in the esophagus ( %). drug exposure cetuximab was given at day − (loading dose), day − , chemoradiation therapy week to week , all together weeks. thirty-seven patients ( %) received weeks of cetuximab. radiation therapy was given during chemoradiation therapy week to week , all together weeks. thirty-nine patients ( %) received weeks of radiation therapy. patients who completed treatment per protocol were either those who com- pleted weeks of cetuximab and weeks of radiation therapy or those who missed part of the per protocol treatment but underwent the surgery. thirty-nine patients ( %) completed the study treatment (table ). dose modification and dose delay please refer to table . cetuximab required dose modi- fication because of toxicity ( %), planned per protocol ( %), or because of intercurrent illness ( %). cetuximab was delayed because of toxicity ( %), intercurrent illness ( %), or scheduling ( %). the dose of radiation therapy was modi- fied because of toxicity ( %) or scheduling ( %). table .  patient characteristics characteristics category/statistics n/value % sex female male race white black or african american unknown ethnicity hispanic or latino non-hispanic not reported age median minimum maximum mean . sd . ps tumor site esophagus gastroesophageal junction disease stage iia iib iii iva histology squamous cell carcinoma adenocarcinoma itt population (n = ). ps, performance status; itt, intent-to-treat. table .  study treatment compliance patients treated cetuximab radiation therapy n = (%) n = (%) number of weeks receiving treatment ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) statistical summary of weeks protocol treatment received median minimum maximum mean . . sd . . total number of patients who completed protocol treatment ( ) table .  dose delay drug dose delay n % cetuximab toxicity intercurrent illness scheduling (patient or site) itt population (n = ). itt, intent-to-treat. copyright © by the international association for the study of lung cancer becerra et al. journal of thoracic oncology ®  •  volume  , number  , november  patients came off study if protocol-defined follow-up was completed ( %), if there was symptomatic deteriora- tion ( %), or if they were lost to follow-up ( %), had died because of disease progression ( %) or other causes ( %), or for other reasons, that is, disease progression after restaging evaluation and before surgery ( %). toxicity table shows grade toxicities seen in more than % of patients. no grade toxicities occurred in more than % of patients, and one grade toxicity occurred in % of patients. each occurrence of toxicity is the highest level each patient reported, irrespective of causality attributed. the most fre- quent toxicity for all grades was dysphagia ( %) followed by acneiform rash ( %), nausea ( %), and fatigue ( %). the most frequent grade toxicities occurring in more than % of patients were dysphagia ( %), dehydration ( %), and anorexia ( %). grade toxicities included aspiration and infection, each reported in one patient ( %). four patients died during the study, two because of dis- ease progression, one because of aspiration pneumonia after surgery, and one because of septic shock. pathological response thirty-one patients ( %) underwent esophagectomy. of these, ( %) had a pathologic complete response (pcr; % confidence interval [ci], %– %), and ( %) had no pcr ; % ci, %– %). the pcr percentage rate was higher in earlier-stage disease and in patients with squamous histology (table ). there was no significant difference in pcr for tumors of the esophagus ( %) versus ge junction ( %). ten patients ( %) patients did not have surgery. seven had metastatic disease on restaging evaluation before surgery and were taken off study. one patient was taken off study because of treatment-related complications, one patient devel- oped bilateral lower extremity critical lower limb ischemia, and one declined surgery after a restaging positron emission tomography scan revealed no evidence of active disease. thirty-six patients ( %) reported dysphagia while on the study. twenty-six patients ( %) reported dysphagia relief. discussion in this phase ii study, the addition of cetuximab to exter- nal beam radiation therapy in patients with potentially resect- able esophageal cancers resulted in a pcr rate of . % by intention-to-treat analysis and % in patients who had esoph- agectomy. patients with earlier-stage of disease and squamous histology had a higher percentage of pcr. the combina- tion regimen was well tolerated; the most frequent toxicities included rash, nausea, and fatigue, with more than % of patients completing the scheduled treatment before surgery. pcr as a surrogate marker of improved clinical out- come has been reported in several studies and has been the subject of a recent meta-analysis. scheer et al. reported a meta-analysis on the effect of pcr on overall survival. the patients with tumors that went into a pcr had a median over- all survival of . months versus . months (p = . ) for patients with persistent disease after combined chemotherapy and radiation therapy. thus, the pcr rate seems to be a valid clinical surrogate endpoint for studies with chemotherapy and radiation therapy in esophageal cancers. the role of neoadjuvant chemotherapy and radiation therapy in patients with potentially resectable esophageal adenocarcinomas and squamous cell carcinomas has been the subject of large phase iii randomized studies with mixed results. walsh et al. randomized patients with resectable esophageal adenocarcinoma to surgery or cisplatin/ fu and radiation therapy to gy followed by surgery. the pcr rate was %. surgical mortality was reported as % and % in the experimental and control arm, respectively with esti- mated survival of % versus % at years. in the cancer and leukemia group b study, eligible patients were planned to be randomized to esophagectomy with lymph node dissection or trimodality therapy followed by esophagectomy. the study was closed prematurely because of poor accrual after patients were enrolled ( in the trimodality therapy arm and in the surgery-only arm). the pcr rate in the tri- modality arm of the study was % in patients with available data and . % by intention-to-treat analysis. two patients ( . %) in the trimodality arm did not have surgery, because of evidence of metastatic disease. the median survival ( . versus . years) and -year survival ( % versus %) favored the trimodality arm of the study over surgery alone. despite the early termination of the study, small sample size, and overlapping cis, the study provides additional evidence of the benefit of neoadjuvant trimodality therapy for patients with potentially resectable esophageal cancer. table .  grade   toxicities ≥ % irrespective of causality  attributed ctcae grade n % anorexia dehydration dysphagia dyspnea fatigue hypernatremia itt population (n = ). grade and toxicities did not occur in ≥ % of patients. ctcae, common terminology criteria for adverse events; itt, intent-to-treat. table .  pathologic complete remission by initial clinical  stage and histology stage or histology pcr % iia / iib / iii / iva / adenocarcinoma / squamous cell / pcr, pathologic complete response. copyright © by the international association for the study of lung cancer journal of thoracic oncology ®  •  volume  , number  , november  esophageal and ge junction carcinomas safran et al. reported on a phase ii clinical trial of cetuximab in combination with weekly paclitaxel, carboplatin, and radiation therapy for patients with esophageal adenocarci- nomas. the complete clinical response rate was %. grade and toxicity included cetuximab-related rash ( %) and esophagitis ( %). in a phase ib/ii study, patients with resectable esophageal adenocarcinoma and squamous cell car- cinomas received two -week cycles of cisplatin, docetaxel, and cetuximab followed by radiation therapy, weekly cis- platin, and cetuximab. no limiting toxicity occurred, and the three main toxicities included esophagitis, anorexia, and fatigue. surgery was performed in patients. anastomotic leak occurred in three patients. the rate of complete patho- logic response was %, and the event-free and estimated sur- vival at months was % and %, respectively. leichman et al. reported on a phase ii clinical trial with oxaliplatin, -fluorouracil and external beam radiation therapy on patients with resectable esophageal and ge junction adenocarcinomas. seventy-three patients ( . %) had esophagectomy, the r resection rate was . %, and the pcr rate was %. the death rate resulting from chemora- diation therapy or surgery complications while on study was . %. in a study by spigel et al. on neoadjuvant docetaxel, oxaliplatin, capecitabine, and radiation therapy, % of the patients never made it to surgery because of declining perfor- mance status, patient request, death, physician decision, and disease progression. in our study, % ( of ) of patients with resectable disease did not undergo surgery, % ( of ) because of development of metastatic disease on restag- ing evaluation. although multiagent chemotherapy seems to control micrometastatic disease, the toxicities of the regi- men result in poor patient tolerance and an increased rate of complications. although we can only speculate among studies, our study compares favorably, the drug regimen was very well tol- erated, and it had a very high rate of pcrs, raising the question of the need for concomitant chemotherapy, radiation therapy, and an egfr-blocking agent. sequencing of the agents might be more appropriate to incorporate chemotherapy into the schedule and possibly reduce the rate of systemic recurrence. our study has limitations. we did not collect informa- tion on tumor location and did not perform a complete assess- ment of dysphagia, which may be helpful information in the management of esophageal cancer. acknowledgments this study was supported by a grant from bristol- meyers squibb. references . cancer facts & figures . atlanta, ga: american cancer society, : no. . . walsh tn, noonan n, hollywood d, kelly a, keeling n, hennessy tp. a comparison of multimodal therapy and surgery for esophageal adenocar- cinoma. n engl j med ; : – . . tepper j, krasna mj, niedzwiecki d, et al. phase iii trial of trimodality therapy with cisplatin, fluorouracil, radiotherapy, and surgery compared with surgery alone for esophageal cancer: calgb . j clin oncol ; : – . . herskovic a, martz k, al-sarraf m, et al. combined chemotherapy and radiotherapy compared with radiotherapy alone in patients with cancer of the esophagus. n engl j med ; : – . . gibson mk, abraham sc, wu tt, et al. epidermal growth factor recep- tor, p mutation, and pathological response predict survival in patients with locally advanced esophageal cancer treated with preoperative chemoradiotherapy. clin cancer res ; : – . . pueyo g, mesia r, figueras a, et al. cetuximab may inhibit tumor growth and angiogenesis induced by ionizing radiation: a preclinical rationale for maintenance treatment after radiotherapy. oncologist ; : – . . bonner ja, harari pm, giralt j, et al. radiotherapy plus cetuximab for squamous-cell carcinoma of the head and neck. n engl j med ; : – . . ajcc cancer staging handbook, th ed. new york, ny: springer- verlag, . . national cancer institute common terminology criteria for adverse events v . (ctcae). washington, dc: national cancer institute, . available at: http://ctep.cancer.gov/protocoldevelopment/electronic_ applications/docs/ctcaev .pdf. accessed march , . . scheer rv, fakiris aj, johnstone pa. quantifying the benefit of a pathologic complete response after neoadjuvant chemoradiotherapy in the treatment of esophageal cancer. int j radiat oncol biol phys ; : – . . safran h, suntharalingam m, dipetrillo t, et al. cetuximab with concur- rent chemoradiation for esophagogastric cancer: assessment of toxicity. int j radiat oncol biol phys ; : – . . ruhstaller t, pless m, dietrich d, et al. cetuximab in combination with chemoradiotherapy before surgery in patients with resectable, locally advanced esophageal carcinoma: a prospective, multicenter phase ib/ii trial (sakk / ). j clin oncol ; : – . . leichman lp, goldman bh, bohanes po, et al. s : a phase ii clinical and prospective molecular trial with oxaliplatin, fluorouracil, and exter- nal-beam radiation therapy before surgery for patients with esophageal adenocarcinoma. j clin oncol ; : – . . spigel dr, greco fa, meluch aa, et al. phase i/ii trial of preoperative oxaliplatin, docetaxel, and capecitabine with concurrent radiation therapy in localized carcinoma of the esophagus or gastroesophageal junction. j clin oncol ; : – . http://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/ctcaev .pdf http://ctep.cancer.gov/protocoldevelopment/electronic_applications/docs/ctcaev .pdf 	a phase ii study with cetuximab and radiation therapy for patients with surgically resectable esophageal and ge junction carcinomas: hoosier oncology group g - 	materials and methods 	patients 	treatment 	assessment 	efficacy 	safety and tolerability 	statistical methods 	results 	drug exposure 	dose modification and dose delay 	toxicity 	pathological response 	discussion 	acknowledgments 	references overstory species response to clearcut harvest across environmental gradients in hardwood forests contents lists available at sciencedirect forest ecology and management journal homepage: www.elsevier.com/locate/foreco overstory species response to clearcut harvest across environmental gradients in hardwood forests j. travis swaima,c, daniel c. deyb, michael r. saundersa, dale r. weigelb, christopher d. thorntonc, john m. kabrickb, michael a. jenkinsa,⁎ a department of forestry and natural resources and hardwood tree improvement and regeneration center, purdue university, west state street, west lafayette, in , united states b usda forest service, northern research station, abnr building, columbia, mo , united states c tell city ranger district, hoosier national forest, th street, tell city, in , united states a r t i c l e i n f o keywords: long-term data stand development even-aged species composition central hardwood region ordination analysis edaphic variables disturbance history a b s t r a c t despite their long history as a forest dominant, the importance of quercus (oak) species is declining under contemporary disturbance regimes in many parts of the world. this is cause for concern considering the great economic and ecological value of this genus. while many chronosequence studies have shown that clearcutting has accelerated the loss of quercus species in forests of eastern north america, long-term repeated measures studies are needed to understand how topo-edaphic variables and disturbance history influence the persistence of the genus in post-harvest stands. in , a study was implemented on the hoosier national forest (hnf) in southern indiana, usa to examine the fate of quercus species and their competitors within developing stands following clearcut harvests. permanent plots were established in six harvest units in quercus-carya (oak-hickory) forests using a stratified design to capture a variety of physiographic and edaphic conditions. pre-harvest plot data were collected in and plots were resampled in , allowing documentation of shifts in species composition over a -year period ( – ). aerial photos from the s were used to determine canopy cover and likely historic land-use within each stand prior to incorporation into the hnf. to characterize edaphic conditions, soil samples were collected and analyzed for chemical characteristics in . non-metric multi- dimensional scaling (nms) and multiple linear regression using fixed and mixed-effect models were used to examine species composition along topo-edaphic and historic canopy cover gradients. we observed drastic declines in the importance of quercus alba (white oak), quercus velutina (black oak), and quercus prinus (chestnut oak) across all stands following harvest. during the same time period, we observed large increases in the im- portance of other species, with acer rubrum (red maple) showing large increases on the driest sites and liriodendron tulipifera (tulip-poplar) displaying the greatest increase across all sites. in pre-harvest stands, q. prinus was confined to the poorest sites and displayed the strongest association of all species with historically closed canopies in both pre and post-harvest stands. in post-harvest stands, the diminished importance of q. alba was associated with low soil nitrogen levels and historically open canopies. l. tulipifera and prunus serotina (black cherry) were associated with more nutrient-rich mesic sites in post-harvest stands. populus grandidentata (big- tooth aspen) in post-harvest stands was associated with historically closed canopies and low cation exchange capacity. . introduction quercus (oak) is an ecologically important genus in temperate for- ests throughout the world, with overstory species in eastern north america, in eastern asia, in europe, and in pacific north america (latham and ricklefs, ). however, the sustainability of quercus species across forest landscapes is threatened by climate-related mortality of mature trees (jenkins and pallardy, ; demchik and sharpe, ; allen et al., ; levanič et al., ), and wide-spread regeneration failure resulting from shifting disturbance regimes (shrestha, ; zavaleta et al., ; plieninger et al., ; mcewan et al., ; dey, ). observed and predicted declines in quercus species importance have caused great concern because significant re- ductions in oak abundance will likely have profound effects on forest https://doi.org/ . /j.foreco. . . received january ; received in revised form june ; accepted june ⁎ corresponding author. e-mail address: jenkinma@purdue.edu (m.a. jenkins). forest ecology and management ( ) – available online june - / © elsevier b.v. all rights reserved. t http://www.sciencedirect.com/science/journal/ https://www.elsevier.com/locate/foreco https://doi.org/ . /j.foreco. . . https://doi.org/ . /j.foreco. . . mailto:jenkinma@purdue.edu https://doi.org/ . /j.foreco. . . http://crossmark.crossref.org/dialog/?doi= . /j.foreco. . . &domain=pdf ecosystems throughout the global range of the genus (fei et al., ; dey, ). quercus is often considered a foundation genus because member species drive trophic interactions and influence population and community dynamics by providing hard mast and shelter for numerous vertebrate species (ellison et al., ) and serving as habitat for large numbers of insect species (opler, ; futuyma and gould, ; vodka et al., ), often several times more than other co-occurring hardwood species (southwood, ; ranius and jansson, ). many presettlement forests of eastern north american were pyro- genic as structure and composition was dependent on the recurrent use of fire by native americans (nowacki and abrams, ). during early europeans settlement, fire frequencies increased and forests were cleared for row agriculture and livestock grazing, perpetuating dis- turbance-adapted quercus species across the landscape (guyette et al., ). in many eastern hardwood forests, the loss of castanea dentata (american chestnut) also resulted in increased importance of quercus species, particularly in the southern appalachian mountains (nelson, ; woods and shanks, ). starting in the early th century, fire suppression created drastic reductions in quercus regeneration, espe- cially on more productive sites (dey, ). although quercus species remain dominant in the overstory of many forests, the lack of fire or surrogate disturbances have allowed them to be replaced by shade- tolerant and fire-sensitive species. the desire to perpetuate quercus-dominated forests fostered ex- tensive research to identify silvicultural systems best suited to re- generating and sustaining the genus. in the eastern united states, even- aged reproduction systems in the form of clearcutting were deemed suitable and widely implemented throughout the range of quercus species (mcgee, ). clearcut harvests were considered the most efficient and economically advantageous form of silviculture due to greater yields per hectare, lower harvesting costs, and simplified management (clark and watt, ; sander and clark, ). this technique was frequently viewed as a panacea for regenerating quercus species and was implemented across much of the eastern united states from the s through the s (roach and gingrich, ; sander and clark, ). subsequent observations in post-harvest stands often found com- position shifting away from quercus domination towards a mix of hardwoods species, raising concerns about the efficacy of clearcutting (george, ). early research in clearcuts suggested that site condi- tions determined the distribution and importance of quercus species, with quercus reproduction in greatest abundance on the poorest site, but rarely occurring on more productive sites where l. tulipifera, acer spp., p. serotina, populus spp., and fraxinus spp. (ash species) dominated (trimble and hart, ; hilt, ). understanding the post-harvest persistence of quercus species re- quires a fine-scale examination of how the silvics of the genus interact with disturbance and environmental gradients. several physiological and morphological characteristics give quercus species an advantage under xeric conditions including the rapid development of a long tap- root, the ability to photosynthesize and conduct water under high drought stress, and the maintenance of high root:shoot ratios through recurrent shoot dieback (abrams, ; pallardy and rhoads, ; parker and dey, ; johnson et al., ). however, these char- acteristics are associated with greater investment in belowground growth and may create a competitive disadvantage on more mesic sites with greater water availability. the central hardwood region (chr) of the united states is the archetype of a region whose forests have experienced declines in quercus species importance in response to shifting disturbance regimes. the hoosier national forest (hnf) in southern indiana used clearcut- ting as its dominant harvesting prescription for approximately years, ending in the late s. as with much of the southern portion of the chr, forests of hnf are underlain by bedrock (nonglaciated) substrate that creates gradients of moisture availability, microclimate, and soil fertility (van kley et al., ). throughout the chr, the forests that developed after clearcutting are now entering the understory re-in- itiation stage (oliver and larson, ), and more research is needed to understand how these forests developed across physiographic gradients as they enter later stages of stand development. in a chronosequence study of – year-old clearcuts, jenkins and parker ( ) found that clearcuts on mesic slopes contained greater importance of l. tulipifera than dry-mesic slopes where quercus species competed better. while such chronosequence studies have provided valuable information, they may fail to capture ecological patterns and processes across environ- mental gradients that are better assessed by the resampling of long-term monitoring plots. consequently, long-term datasets are critical to un- derstanding forest development in post-harvest stands where a diverse array of overstory species and high inter-site variability may limit the use of a chronsequence (walker et al., ). in , a long-term, repeated measure study was implemented on the hnf by the usda forest service north-central experiment station to examine the competitive ability of quercus species after clearcutting and the post-harvest development of forest stands. in , we re- sampled plots that were established prior to harvesting in and used these data to examine the importance of quercus species and their competitors in pre- and post-harvest stands. we also examined how species were distributed across gradients created by historical land use and topo-edaphic characteristics. based upon these examinations, we address three primary questions: . how do the distributions of species differ between pre- and post- harvest stands in relation to underlying environmental gradients? . how have land-use practices prior to the creation of the hnf (re- presented by canopy cover in – aerial photos) influenced the pre- and post-harvest composition of sampled stands? . where across contemporary environmental and disturbance gra- dients are quercus species best able to persist in post-harvest stands? . methods . . harvest treatments in , six mature quercus-carya dominated stands ranging in size from . to . ha were harvested on the hnf. three of the stands were located on the brownstown ranger district and three stands were located on the tell city ranger district (table ). all stands were clearcut for merchantable timber between the months of april and august . after removal of merchantable timber, all remaining live and cull trees ≥ . cm dbh (diameter at breast height) were felled. no herbicide treatments or additional site preparation treatments were performed. . . study sites stand is located in the brown county hills section of the highland rim natural region (homoya et al., ). this section is characterized by deeply dissected uplands underlain by siltstone, sandstone, and shale. soils are brownstone channery silt loams, which are moderately deep and well drained soils formed in residuum from siltstone. steep slopes and narrow hollows are typical topographic features of this section (van kley et al., ). there is little disparity in forest com- position, with quercus-carya dominated uplands often consisting of nearly pure stands of q. prinus on upper slopes and ridge tops. mesic ravines harbor species such as fagus grandifolia (american beech), quercus rubra (northern red oak), acer saccharum (sugar maple), and fraxinus americana (white ash). stands , , and are located in the crawford upland section of the shawnee hills natural region (homoya et al., ). the broad ridge tops and flats of this section contrast with the narrow ridges and steep terrain of the brown county hills section. stand is underlain by wellston-tipsaw-adyeville complex (ultic hapludalfs, typic j.t. swaim et al. forest ecology and management ( ) – dystrudepts, and typic hapludults, respectively) soils, which are moderately deep, well to excessively well-drained and were formed in loamy residuum from sandstone interbedded with shale and siltstone (soil survey staff, ). soils in stand include the apalona-zanes- ville complex (oxyaquic fragiudults) and the adyeville-wellston complex (typic hapludults and ultic hapludalfs), which formed in residuum from interbedded sandstone, siltstone, and shale (soil survey staff, ). these soils are moderately deep to deep and vary from moderately well to excessively drained. the apalona and zanesville soils have a fragipan that restricts rooting and impedes drainage. stand is underlain by adyeville (typic hapludults) and wellston (ultic hapludalfs) silt loams, which are moderately deep, well drained to excessively well drained, and formed in residuum from interbedded sandstone, siltstone, and shale (soil survey staff, ). species com- position on upper slopes often consists of mixed quercus and carya species including q. velutina, q. alba, q. prinus, quercus coccinea (scarlet oak), carya glabra (pignut hickory), and carya ovata (shagbark hickory). cove forests of the section typically consist of mesophytic species such as f. grandifolia, l. tulipifera, q. rubra, a. saccharum, and juglans nigra (black walnut). stands and are located in the escarpment section of the shawnee hills natural region (homoya et al., ). the rugged hills of this section are situated along the eastern boundary of the region and are characterized by soils from the wellston-adyeville complex (typic hapludults and ultic hapludalfs), which are moderately deep, well drained to excessively drained, and formed in residuum from inter- bedded sandstone, siltstone, and shale (soil survey staff, ). dry- mesic and mesic forest types are common and similar to those of the crawford upland section, although q. prinus is often absent and re- placed by q. velutina and quercus stellata (post oak). . . plot establishment and sampling prior to harvest, between april and june , , data were collected from sixty-one m fixed circular plots across the six stands. within each stand, between and permanent overstory plots were located using a stratified random design based upon topographic positions. plots centers were located a minimum of . m from the edge of the clearcut, in order to reduce edge effects from surrounding mature stands, and situated within each stand to encompass as many aspect-slope combinations as possible. study plots were established by personnel from the usda forest service, north-central research station and the hnf. during the summer of , of the original plots were relocated and remeasured. two plots in stand were never harvested and, therefore, not resampled in . four plots in stand could not be relocated. all trees ≥ . cm dbh were measured by species in each m overstory plot. advanced reproduction at the time of harvest was as- sessed in three . m circular subplots within each overstory plot. each overstory plot was divided into thirds using a randomly chosen azimuth to locate the first dividing line. one hundred twenty degrees were added to this azimuth to locate the second line, and another degrees added to locate the third line. within each third, a . m circular plot was randomly located by adding a random number from between and to the azimuth of the dividing line. these plots were permanently marked with a length of metal conduit. the . m sub- plots were not allowed to overlap, so they were spaced at least . m apart. subplots were also spaced so they were at least . m from the center and outside edge of the main plot. all woody stems in these plots were tallied into nine diameter classes ranging from to . cm dbh and nine height classes ranging from < . cm to > . cm. in , overstory plots and regeneration subplots were both re- sampled. we collapsed height and diameter classes from the original sampling protocol to define seedlings as woody stems ≤ . m height and saplings as stems > . m height but less than ≤ . cm dbh. site variables measured, at the scale of the overstory plot, included aspectta b le lo ca ti on , si te , an d so il ch em ic al d es cr ip ti on s of si x st u d y st an d s on th e h oo si er n at io n al fo re st in so u th er n in d ia n a. * t ot al ac ro ss al l sp ec ie s (m h a ); m ea n ± st an d ar d er ro r. st an d lo ca ti on la t/ lo n g (n /w ) si ze (h a) # of p lo ts p re -h ar ve st ba sa l ar ea * p os t- h ar ve st ba sa l ar ea * h is t. ca n op y co ve r (% ) t ra n s. as p ec t ra n ge sl op e (% ) ra n ge so il p h ra n ge c e c ra n ge c a (% ) ra n ge k (% ) ra n ge m g (% ) ra n ge t ot al n (% ) ra n ge t ot al c (% ) ra n ge . – . . . ± . . ± . . – . – . – . . – . . – . . – . . – . . – . . – . . – . . . ± . . ± . . – . – . – . . – . . – . . – . . – . . – . . – . . – . . . ± . . ± . . – . – . – . . – . . – . . – . . – . . – . . – . . – . . . ± . . ± . . – . – . – . . – . . – . . – . . – . . – . . – . . – . . . ± . . ± . . – . – . – . . – . . – . . – . . – . . – . . – . . – . . . ± . . ± . . – . – . – . . – . . – . . – . . – . . – . . – . j.t. swaim et al. forest ecology and management ( ) – (with compass), percent slope (with clinometer), and slope position, which was visually assessed as a relative proportion in relation to a ridge summit (= ) or drainage (= ). soil samples were also col- lected from each plot to examine how soil characteristics are correlated with species composition and the importance of quercus species. subsamples were collected from the top cm of the upper horizon at ∼ m from plot center in each cardinal direction. subsamples were pooled and analyzed by overstory plot. . . data preparation summary statistics calculated for each overstory plot included density (stems/ha), basal area (m /ha), relative density, relative basal area, and importance value [iv; (relative basal area + relative density)/ ] of each species present in (pre-harvest) and (post-harvest) stands. pre and post-harvest regeneration subplot data were used to calculate relative density of seedlings and saplings by species. overstory plot and regeneration subplot data were not combined for analysis as some stems were counted in both plot types (i.e., stems . – . cm dbh). soil samples were frozen immediately after field collection until they were processed for analysis. samples were analyzed for ph (in a : slurry of deionized water using a ph meter), concentrations of melich-iii extractable cations (p, k, mg, and ca; ppm), organic matter (%; ignition loss method), and cation exchange capacity (cec; barium chloride method) by a&l analytical laboratories (memphis, tn). percent saturation of k, mg, and ca was calculated as the percentage of respective cation that contributed to total cec. a subset of all samples were dried at °c for h and then ground to a fine texture with mortar and pestle. carbon (mg absolute), nitrogen (mg absolute), % carbon, and % nitrogen were determined from these samples using an ecs chnso analyzer (costech analytical technologies, inc). to more efficiently relate aspect to site productivity, aspect was transformed to a linear scale ranging from to , with zero value at southwest using the following formula: transformed aspect = cos( - aspect) + (beers et al., ). percent historical canopy cover for each study site was estimated to the nearest % using a stereoscope and dot grid on : , aerial photos of the hnf flown between and . this allowed assessment of likely historic land-use in each stand prior to incorporation into the hnf. . . data analysis non-metric multidimensional scaling (nms) was conducted using pc-ord version (mccune and mefford, ) to examine species composition along topo-edaphic and historical canopy cover gradients. we used data from both and to examine compositional changes in the overstory following harvest and data to examine the distribution of seedling and saplings in contemporary stands years after harvest. nms was utilized due to the relaxed assumption of normality and because it does not assume a linear response of species to site-related gradients (mccune et al., ). the main matrices contained overstory iv, sapling relative density, and seedling relative density data, while the secondary matrix for all analyses included environmental variables consisting of all previously described soil chemical analyses, in addition to transformed aspect, slope position, percent slope, and percent historical canopy cover. autopilot mode (slow and thorough) was selected using the sorensen (bray-curtis) distance measurement, with . × − stability criterion, runs with real data, runs with randomized data, and maximum iterations for each. according to clarke ( ), nms ordination ana- lyses with a final stress > are likely to yield plots which could be dangerous to interpret. our nms analysis of sapling data yielded a stress = . because of this high stress value and the resulting proble- matic interpretation, the results of this analysis are not presented. correlations between environmental variables and ordination axes scores were evaluated based on pearson's correlation coefficients. overstory a. saccharum and a. rubrum were not separated in stands , , and during pre-harvest sampling, but were instead grouped by genus. therefore, we combined acer species in the data prior to running nms ordination for overstory iv data from all six stands. successional vectors were created for the overstory nms ordination to allow visual comparison of temporal shifts in species composition. vectors of environmental gradients were assessed with pearson’s cor- relation analysis and overlaid onto species biplots to examine species/ site relationships. because topography and edaphic characteristics in- fluence land use, we also used pearson’s correlation analysis to examine relationships between historic canopy cover and all other variables. bubble plots were created to illustrate overstory species importance on individual plots and shifts in distribution along environmental gra- dients following harvest. to confirm the results of the nms analysis and more directly ex- amine the relationships between the importance of individual species and environmental variables, we used multiple linear regression with fixed and mixed-effects models to analyze the relationship between overstory species ivs (response variables) and environmental variables (independent variables; table ). square root and natural log trans- formations of response variables were used as needed to homogenize error variance. full models of all environmental variables were fit to the data with multiple linear regression, and then reduced to include only significant environmental variables with a backwards stepwise proce- dure. criteria for a variable to stay in the model was set at p < . . variance inflation factors (vif) were then used to detect whether in- dependent variables in the final models had strong linear associations. models with mean vifs considerably greater than were considered to contain significantly collinear variables (neter et al., ); variables were removed as needed to reduce overall vif to less than . these models were then used in a mixed-effects framework to account for the sample design of plots within stands. stand was added as a random effect that would adjust the overall y-intercept of the model. sig- nificance of random effects was assessed using a maximum likelihood test of the mixed effects model against the fixed effects model. in many cases, random effects could not be estimated (i.e., variance equal to ), and were subsequently dropped from the model. for all final models, plots of studentized residuals versus fitted values were used to assess the assumption of constant variance, and residual plots and standard techniques were used to evaluate influence of potential outliers (neter et al., ). all predictors in the final models, except in some cases the intercept and random effects, were significant at p < . . regression analysis was performed in r . . (r core team ) using packages stat and lme (bates et al., ). the package lmertest (kuznetsova et al., ) was used to assess significance of all models and provide summary statistics such as r . . results . . overstory ordination analysis the species data from all six stands were best described by a - dimensional solution in nms, which displayed a final stress of . and instability criterion of < . . overall, the ordination explained % of variance in overstory community composition, with the domi- nant axes and explaining % and % of variation, respectively. overstory (trees ≥ . cm dbh) species compositions of pre- and post- harvest stands were strongly separated in ordination space and shifts in composition were largely unidirectional (fig. a). in ordination space, pre-harvest plots were strongly associated with greater importance of cornus florida (flowering dogwood), q. alba, q. prinus, q. rubra, and q. velutina, while post-harvest plots were associated with acer spp., l. tulipifera, oystra virginiana (eastern hophornbeam), p. grandidentata, and p. serotina (fig. b). nms results for all stands indicated that slope position, percent j.t. swaim et al. forest ecology and management ( ) – slope, soil ph, soil potassium, percent soil calcium saturation, percent soil potassium saturation, percent soil hydrogen, and percent historical canopy cover were the environmental variables most correlated with the distribution of overstory species across axes and (table ). variables showing strong correlations with axis included: percent soil calcium (r = . ), percent historical canopy cover (r = − . ), per- cent soil hydrogen (r = − . ), ph (r = . ), and soil calcium con- centration (r = . ). variables showing strong correlations with axis included: transformed aspect (r = − . ), slope position (r = − . ), percent slope (r = − . ), soil potassium (r = . ), percent soil po- tassium saturation (r = − . ), and percent historical canopy cover (r = − . ). percent historical canopy cover displayed the strongest correlation of any variable with axis and the second strongest cor- relation with axis (table ). however, plots with high historic canopy cover also had steep slopes, low ph, and low cec, making it difficult to distinguish gradients resulting for past disturbance from gradients re- sulting from topo-edaphic conditions. generally, edaphic variables were most strongly correlated with the gradients represented by axis , although the variables we measured displayed poor correlations com- pared to axes and (max r = . for ph). percent historic canopy cover was significantly correlated with nine out of nineteen topo- graphic and edaphic variables, and was most strongly correlated with percent slope (r = . ) and percent soil calcium saturation (r = − . ; data not shown). . . site conditions and compositional shifts the overstory ordinations illustrate that clearcutting has largely reduced the importance of quercus and carya species across study sites (table , fig. ). in pre-harvest stands, quercus alba showed a strong correlation with axis , which was also correlated with historically open canopies, less steep slopes, and reduced k and ca availability (table , fig. ). while regression analysis displayed poor fits between the im- portance of q. alba and environmental variables (table ; adj. table pearson's correlation coefficients between nms ordination axes scores and environmental variables and individual overstory and seedling species. overstory seedlings axis axis axis axis axis axis variable symbol r r r r r r transformed aspect asp . − . . − . − . − . slope position sp . − . − . − . . − . percent slope %sl − . − . − . . . − . soil ph ph . . . − . − . . soil phosphorus (ppm) p . − . − . − . . − . soil potassium (ppm) k . − . . − . . . soil calcium (ppm) ca . . . − . − . . soil magnesium (ppm) mg . − . . − . − . . percent organic matter %om . − . . − . . . cation exchange capacity (meq/ g) cec . . . − . − . . percent soil potassium saturation %k − . − . − . . . − . percent soil calcium saturation %ca . − . − . − . − . − . percent soil magnesium saturation %mg . − . − . . − . . percent soil hydrogen saturation %h − . − . . . . − . soil nitrogen (mg) n . − . . − . − . − . soil carbon (mg) c − . − . . − . . . percent soil nitrogen %n . − . . − . − . − . percent soil carbon %c − . − . . − . . . percent historical canopy closure %cc − . − . . . . − . species acer rubrum ar – – – . . − . a. saccharum as – – – − . − . − . acer spp. asp . − . . – – – ailanthus altissima aa – – – − . − . . amelanchier arborea aar – – – . . . asimina triloba at – – – − . − . . carpinus caroliniana cc . . − . − . − . − . carya spp. cs − . . . . . . celtis occidentalis co – – – − . − . . cercis canadensis cec . . − . . − . . cornus florida cf − . . − . − . − . . fagus grandifolia fg . − . . − . − . . fraxinus americana fa . − . − . − . − . . liriodendron tulipifera lt . − . − . − . − . . morus rubra mr – – – . − . − . nyssa sylvatica ns − . . − . . − . − . ostrya virginiana ov . . − . − . . − . populus grandidentata pg . − . − . − . . − . prunus sylvatica ps . − . − . − . . . qurcus alba qa − . . . . . . q. coccinea qc − . . − . − . . . q. muehlenbergii qm . − . − . − . − . − . q. prinus qp − . − . . . . . q. rubra qr − . . . − . . . q. velutina qv − . . . . . . rhus spp. rs – – – . . . sassafras albidum sa . . − . . − . . ulmus spp. us . − . − . − . − . − . j.t. swaim et al. forest ecology and management ( ) – r = . ), pre-harvest importance of the species was most strongly, and negatively, related to percent slope and organic matter content. quercus prinus displayed a strong negative correlation with axis , which was associated with historically closed canopies, steep slopes, and acidic low calcium soils (table , fig. b). regression analysis re- vealed that q. prinus iv in pre and post-harvest stands was strongly associated (adj. r = . and . , respectively) with historically closed canopies on xeric sites with poor soils (high c content, low ph, low soil mg and p content). in post-harvest stands, q. alba nearly dis- appeared altogether on poor sites (steeper slopes and low ph; fig. b, table ), and was associated with historically open canopies on sites with lower soil nitrogen content and less steep slopes (table ). in post- harvest stands, q. prinus retained the greatest importance of any quercus species on plots were it occurred (table ). quercus velutina and q. rubra largely disappeared across all plots following harvest (fig. ) and ivs of these species were weakly related to environmental variables in pre and post-harvest stands (table ; max adj. r = . ). carya species displayed an ordination pattern similar to q. alba in post-harvest stands; their importance was strongly corre- lated with historically open canopies, less steep slopes, and reduced k and ca availability in the nms ordination (table ) and the genus largely disappeared from steeper more nutrient poor sites after harvest (fig. ). according to regression analysis, the limited importance of this genus in post-harvest stands was negatively associated with historically closed canopies and positively associated soil hydrogen saturation (table ). as expected, early-successional species, including l. tulipifera, p. serotina, and p. grandidentata, established a strong presence in post- harvest stands (table , fig. ). prior to harvest, l. tulipifera was mostly found on mesic plots with high nutrient availability, but in postharvest stands its importance increased across environmental gradients. for example, plots in stand did not contain a single l. tulipifera prior to harvest, but in the post-harvest stand, it was the single most important species (iv = . ), more than double that of the most important quercus species (q. prinus, iv = . ). according to regression analysis, the importance of l. tulipifera in both pre and post-harvest stands was most strongly related to low levels of h saturation. populus grand- identata and p. serotina were largely absent prior to harvest, but greatly increased in importance after harvest. p. grandidentata established on poor sites (most strongly associated with low cec; table ) and p. serotina importance was associated with better sites (table ; figs. b and ). according to regression analysis, p. serotina iv was most asso- ciated with soils with low carbon content in post-harvest stands (table ). stand saw the largest increase in p. grandidentata im- portance where the species was absent prior to harvest but was the single most important species (iv = . ) after harvest displaying an importance three times greater than any quercus species (q. prinus, iv = . ; table ). prunus serotina was most successful in stand , in- creasing from an iv less than . to the second most important species (iv = . ) behind a. saccharum (iv = . ), and sixteen times more important than the most important quercus species (black oak, iv = . ) in the postharvest stand (table ). . . contemporary seedling and sapling species composition and abundance the seedling species data (stems ≤ . m in height) from all six stands were best described by a -dimensional solution in nms, resulting in a final stress of . and instability criterion of < . . overall, the ordination explained % of the variance in seedlings species composition with axis explaining the greatest proportion of variance in the ordination ( %), followed by axis ( %) and axis ( %). soil calcium (ppm; r = − . ), soil phosphorus (ppm; r = − . ), cec (r = − . ), percent organic matter (r = − . ), and percent nitrogen (r = − . ) were most strongly correlated with axis (table ). transformed aspect displayed the strongest correlation with axis of any non-edaphic variable (r = − . ). soil ph (r = − . , percent hydrogen saturation (r = . ), and percent soil calcium (r = − . ) were most strongly associated with axis . per- cent historical canopy cover displayed the strongest correlation with axis of any non-edaphic variable (p = . ). two topographic variables, percent slope and slope position, were the variables most strongly associated with axis (r = − . and r = − . , respec- tively). overall, axis was poorly correlated with the variables we measured. the relative densities of c. caroliniana, f. grandifolia, ulmus spp., f. americana, and a. saccharum were associated with better sites represented by greater ph, percent calcium saturation, soil nitrogen and fig. . nms ordination of overstory species importance across all six stands. (a) successional vectors indicating change in composition from (pre-harvest) to (post-harvest). (b) biplot of most important overstory species in and dominant environmental variables (r ≥ . ); %ca (percent soil calcium saturation), ph (soil ph), %k (percent soil potassium saturation), %sl (percent slope), %cc (percent historical canopy closure), and %h (percent soil hydrogen saturation) are shown as vectors. explanations of species abbreviations are provided in table . j.t. swaim et al. forest ecology and management ( ) – cec, while q. prinus, q. rubra, and a. rubrum were associated with poorer sites that had lower ph, cec, soil calcium saturation, and soil nitrogen (fig. ). twenty-three years post-harvest, a. rubrum seedlings had the highest relative density in each stand with the exception of stand , where p. serotina was most abundant (table ). between and , the relative density of a. rubrum increased greatly (six to -fold) in the three stands where acer species were separated in the survey. cornus florida was a dominant seedling species in , but was a minor component of the seedling layer in (table ). viburnum acerifolium (mapleleaf viburnum) was an important component of the seedling layer in most stands in , but declined greatly in the inventory. in , a. saccharum dominated the woody sapling (> . m height, ≤ . cm dbh) layer in four out of six stands (table ). cornus florida was an important sapling species in with relative densities ranging from . to . (table ). in , the relative density of this species declined in all stands, ranging from to . . three species, fagus grandifolia, f. americana, and and o. virginiana increased in re- lative density on five out of six stands between and . overall, seedlings and saplings of quercus species were relatively scarce in both and (tables and ). . discussion . . pre-harvest conditions and historic disturbance the sites sampled in this study covered a variety of physiographic conditions, including mesic slopes on northern and eastern aspects, dry- mesic slopes on southern and western aspects, and more exposed, xeric ridge tops. the fact that quercus species were once dominant across this range of site conditions, but were unable to regenerate under con- temporary disturbance regimes, suggests that the genus was perpe- tuated by exogenous factors in pre-harvest stands. evidence shows that widespread burning in this region created an historical mismatch be- tween the physiographic limits set by climate and the expression of vegetation on the landscape (nowacki and abrams, ). native americans were the primary ignition source in presettlement indiana, but as europeans began settlement around , fire frequencies quickly escalated (guyette et al., ; ponder, ). during this post- settlement period, large scale clearing and burning for cropland and open forest grazing by livestock further altered the landscape and helped perpetuate quercus species. starting in the s, these agri- cultural lands were largely abandoned and fire was excluded from the system, allowing late-seral species to accumulate in the understories of maturing forests. in our study, we observed high densities of small diameter (sapling) acer species prior to harvest, setting the stage for these forests to transition from early-mid seral quercus-dominated communities to late-seral communities after harvest. historical aerial photographs flown between and dis- played evidence of past disturbances in the stands we studied. in gen- eral, stands were much more open compared to the closed canopies that existed in prior to harvest. while the resolution of the photo- graphy did not allow fine-scale identification of individual disturbance agents, the openness of these stands suggests they were subjected to a well-documented array of disturbances (burning, grazing, logging and agriculture) that occurred across southern indiana in the late s and early s (sieber and munson, ; jenkins, ). in our study, q. prinus and q. alba showed contrasting affinities for historic canopy cover. as reflected in the nms axis correlations, quercus prinus was most dominant on plots with the highest historical canopy cover ( – % pre-harvest cover), while q. alba importance was positively fig. . bubble plots of quercus species importance overlain onto the nms sample ordination of all six stands. importance value of a species on a given plot is proportional to symbol size. (a) pre-harvest ( ) species importance. (b) species importance years after harvest ( ). j.t. swaim et al. forest ecology and management ( ) – related to more historically open canopies in post-harvest stands. his- toric canopy cover was positively correlated with percent slope and negatively correlated with percent soil calcium saturation, which sup- ports the intuitive conclusion that less steep areas with better soils were used for agriculture. however, of all the environmental variables we included in our analyses, historic canopy cover was most strongly correlated with the distribution of species across nms axes and was a significant term in regressions of q. alba, q. prinus, q. velutina, acer spp. carya spp., and p. grandidentata importance, suggesting that its re- lationship with species distributions extends beyond being a covariate with topo-edaphic variables. . . post-harvest shifts in composition results from this long-term study illustrate a dramatic shift in spe- cies composition years after clearcutting on the hnf. following overstory removal, all stands experienced decreased importance of quercus and carya species and increased importance of other species including a. rubrum, l. tulipifera, p. grandidentata, and p. serotina. this corresponds with other reports of poor competition by quercus species following clearcut harvests in southern indiana (standiford and fischer, ; george and fischer, ; jenkins and parker, ) and else- where in the chr (sander and graney, ; lorimer, ; campione et al., ). prior to harvest, l. tulipifera was a common canopy tree in those stands where it maintained or increased its importance years after harvest. prolific seed production, wide dispersal, seed-banking ability, and rapid growth allowed the successful establishment of this species following clearcut harvests and its longevity may allow it to dominate these stands for many decades (boring et al., ; burns and honkala, ; lambers and clark, ). observed increases on higher quality sites where the species held less importance prior to harvest supports other observations in young clearcut stands in southern indiana (george and fischer, ; jenkins and parker, ; morrissey et al., ). we observed some increase in l. tulipifera importance on dry- mesic sites, but it is questionable whether stems established post-har- vest will persist under more xeric conditions. studies have shown that the importance of l. tulipifera decreases through time following clearcut harvest on dry sites (hilt, , jenkins and parker, ), particularly after drought events. for example, morrissey et al. ( ) speculated that an exceptional drought occurring from to reduced the density of l. tulipifera stems in their study of southern indiana clearcut sites and we have seen similar declines on nearby sites after the drought (meier and saunders ). in the current study, l. tulipifera was well established in all post-harvest stands and persisted for years, but these stands initiated after the – drought oc- curred, likely permitting l. tulipifera to remain highly competitive across most sites. future drought events may reduce l. tulipifera im- portance and eventually allow less drought-sensitive, shade-tolerant species, such as a. rubrum or a. saccharum, to secure growing space. further research is needed to understand the competitive ability of l. tulipifera as stands on dry and dry-mesic sites transition into later de- velopmental stages. the increased importance of p. serotina and ruderal p. grandidentata we observed after harvest illustrates how these species can outcompete quercus and carya species in young stands (swaim et al., ). largely absent in our stands prior to harvest, p. grandidentata grew rapidly on drier sites where it occupied growing space that might otherwise be table linear fixed and mixed-effects regression equations relating importance value (iv) of common overstory species in and to environmental variables. no model = analyses not run; six or fewer non-zero data points in sample year. fixed effects regression analyses: vif = mean of variance inflation factor values for all independent variables. mixed-effects regression analyses: rstand std = standard deviation of random effect stand; rstand p = p-value of random effect stand. environmental variable abbreviations are provided in table . fixed-effects regression: species f adj. r p vif quercus alba : square root (iv) = . – . (%sl) + . (k) − . (%om) + . (%h) . . . . : ln(iv) = . – . (%cc) − . (n) − . (%sl) . . < . . quercus prinus : iv = − . + . (%cc) − . (as) − . (mg) + . (%c) . . < . . : square root = . + . (%cc) + . (%c) − . (ph) − . (p) . . < . . quercus rubra : square root (iv) = . – . (%sl) + . (k) − . (%mg) . . . . : square root (iv) = − . + . (k) − . (cec) + . (%ca) . . . . quercus velutina : square root (iv) = . – . (%sl) − . (%n) . . . . : square root (iv) = . + . (%c) − . (p) − . (%cc) . . . . acer spp. : iv = . + . (sp) + . (p) − . (%cc) . . . . : iv = − . + . (%cc) + . (%om) . . < . . carya spp. : square root (iv) = . – . (k) − . (sp) + . (mg) − . (%cc) . . < . . : ln (iv) = . – . (%cc) + . (%h) . . < . . fagus grandifolia : square root (iv) = − . + . (%sl) − . . (%c) + . (mgn) . . . . : square root (iv) = . + . (sp) − . (p) . . < . . liriodendron tulipifera : ln (iv) = . – . (mg) − . (%h) . . < . . mixed-effects linear regression: species rstand std rstand p marg. r cond. r liriodendron tulipifera : ln (iv) = . + . (ca) − . (%h) + restand . . . . populus grandidentata : no model : ln (iv) = . + . (k) − . (cec) + . (%cc) + restand . . . . prunus serotina : no model : square root (iv) = . – . (c) + . (%om) + restand . < . . . j.t. swaim et al. forest ecology and management ( ) – t ab le im p or ta n ce va lu e of ov er st or y tr ee s (m ea n ± se ) by sp ec ie s in p re -h ar ve st ( ) an d p os t- h ar ve st ( ) st an d s on th e h oo si er n at io n al fo re st . p os t- h ar ve st d at a w er e co ll ec te d ye ar s af te r cl ea rc u tt in g. iv = (r d + r b a )/ , w h er e r d = re la ti ve d en si ty an d r b a = re la ti ve ba sa l ar ea . * a ce r ru br um an d a . sa cc ha ru m w er e n ot d el in ea te d in st an d s , , an d d u ri n g th e sa m p li n g an d ar e co m bi n ed as a ce r sp p . fo r th es e st an d s. st an d st an d st an d st an d st an d st an d sp ec ie s a ce r ru br um . ± . . ± . . ± . . ± . . ± . . ± . – . ± . – . ± . – . ± . a . sa cc ha ru m . ± . . ± . . ± . . ± . . ± . . ± . – . ± . – . ± . – . ± . a ce r sp p .* – – – – – – . ± . – . ± . – . ± . – c ar pi nu s ca ro lin ia na . ± . . ± . . ± . . ± . . ± . c ar ya sp p . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . c er ci s ca na de ns is . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . c or nu s fl or id a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . fa gu s gr an di fo lia . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . fr ax in us am er ic an a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . li ri od en dr on tu lip if er a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . n ys sa sy lv at ic a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . o st ry a vi rg in ia na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . p op ul us gr an di de nt at a . ± . . ± . . ± . . ± . . ± . . ± . . ± . p ru nu s se ro ti na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q ue rc us al ba . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q . co cc in ea . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q . m ue hl en be rg ii . ± . . ± . . ± . . ± . q . pr in us . ± . . ± . . ± . . ± . q . ru bu s . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q . ve lu ti na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . sa ss af ra s al bi du m . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . u lm us sp p . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . o th er sp ec ie s . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . j.t. swaim et al. forest ecology and management ( ) – utilized by dry-site hardwoods such as quercus species (kabrick et al., ). barring future large canopy disturbances, the relatively short life span of p. grandidentata ensures it will eventually be replaced by other species, such as a. rubrum, that can establish under a closed canopy. in our survey, a. rubrum had the highest relative seedling density of any species in five of six study sites, highlighting the ability of this species to establish during stem exclusion. on mesic and dry-mesic sites, p. serotina has filled a role similar to that of p. grandidentata on dry sites. delayed germination, frequent seed crops, and wide dispersal by birds and mammals promote the estab- lishment of p. serotina in many early-successional habitats without the presence of parent trees (pairon et al., ; vanhellemont et al., ). as an early-seral species, p. serotina replaced quercus species on better sites in -year-old stands, and its ability to maintain a growth advantage over associated species for – years may safeguard against replacement on these more productive sites (burns and honkala, ). acer saccharum was an important species across all sites before treatment, and maintained similar importance years after harvest. the high densities of small-diameter, advanced regeneration of a. saccharum forebode its dominance during stand development. although a large proportion of these midstory a. saccharum stems were felled during harvest, smaller suppressed a. saccharum advance reproduction was released by the harvest. also, since herbicides were not used on cut stems, dormant buds located on lower boles of cut a. saccharums likely sprouted prolifically (church and godman, ), further perpetuating a. saccharum in developing stands. traditionally, disturbance was thought to set succession back to an earlier stage. however, if the dis- turbance only affects canopy trees, accelerated succession may occur as suppressed stems are released (abrams and scott, ; oliver and larson, ). although the -year-old stands in this study are fig. . bubble plots of non-quercus species importance overlain onto the nms sample ordination of all six stands. importance value of a species on a given plot is proportional to symbol size. (a) pre-harvest ( ) species importance. (b) species importance years after harvest ( ). fig. . nms ordination of seedling species relative density across all six stands: biplot of most important species in and dominant environmental variables (r ≥ . ); asp (transformed aspect), cec (cation exchange capacity), %ca (percent soil calcium saturation), %cc (percent historical canopy closure), %h (percent soil hydrogen), %k (percent potassium saturation), ppmp (soil phos- phorus in ppm), ph (soil ph), %om (percent organic matter), and %n (percent soil nitrogen) are shown as vectors. explanations of species abbreviations are provided in table . j.t. swaim et al. forest ecology and management ( ) – t ab le r el at iv e d en si ty (m ea n ± se ) of w oo d y se ed li n gs (≤ . m h ei gh t) by sp ec ie s in p re -h ar ve st ( ) an d p os t- h ar ve st ( ) st an d s on th e h oo si er n at io n al fo re st .* a ce r ru br um an d a .s ac ch ar um w er e n ot d el in ea te d in st an d s , , an d d u ri n g th e sa m p li n g an d ar e co m bi n ed as a ce r sp p . fo r th es e st an d s. st an d st an d st an d st an d st an d st an d sp ec ie s a ce r ru br um . ± . . ± . . ± . . ± . . ± . . ± . – . ± . – . ± . – . ± . a . sa cc ha ru m . ± . . ± . . ± . . ± . . ± . . ± . – . ± . – . ± . – . ± . a ce r sp p .* – – – – – – . ± . – . ± . – . ± . – a ila nt hu s al ti ss im a . ± . a si m in a tr ilo ba . ± . . ± . . ± . . ± . c ar pi nu s ca ro lin ia na . ± . . ± . . ± . . ± . . ± . c ar ya sp p. . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . c el ti s oc ci de nt al is . ± . . ± . . ± . . ± . . ± . c er ci s ca na de ns is . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . c or nu s fl or id a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . fa gu s gr an di fo lia . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . fr ax in us am er ic an a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . li ri od en dr on tu lip if er a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . n ys sa sy lv at ic a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . o st ry a vi rg in ia na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . p op ul us gr an di de nt at a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . p ru nu s se ro ti na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q ue rc us al ba . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q . co cc in ea . ± . . ± . . ± . . ± . . ± . q . m ue hl en be rg ii . ± . . ± . q . pr in us . ± . . ± . . ± . . ± . . ± . q . ru br a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q . ve lu ti na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . r hu s sp p. . ± . . ± . . ± . . ± . sa ss af ra s al bi du m . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . u lm us sp p . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . v ib ur nu m ac er if ol iu m . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . o th er sp ec ie s . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . j.t. swaim et al. forest ecology and management ( ) – t ab le r el at iv e d en si ty (m ea n ± se ) of w oo d y sa p li n gs (> . m h ei gh t; < . cm d bh ) by sp ec ie s in p re -h ar ve st ( ) an d p os t- h ar ve st ( ) st an d s on th e h oo si er n at io n al fo re st . * a ce r ru br um an d a . sa cc ha ru m w er e n ot d el in ea te d in st an d s , , an d d u ri n g th e sa m p li n g an d ar e co m bi n ed as a ce r sp p . fo r th es e st an d s. st an d st an d st an d st an d st an d st an d sp ec ie s a ce r ru br um . ± . . ± . . ± . . ± . . ± . . ± . – . ± . – . ± . – . ± . a . sa cc ha ru m . ± . . ± . . ± . . ± . . ± . . ± . – . ± . – . ± . – . ± . a ce r sp p .* – – – – – – . ± . – . ± . – . ± . – a ila nt hu s al ti ss im a . ± . a m el an ch ie r ar bo re a . ± . . ± . . ± . a si m in a tr ilo ba . ± . . ± . . ± . . ± . . ± . c ar pi nu s ca ro lin ia na . ± . . ± . . ± . c ar ya sp p. . ± . . ± . . ± . . ± . . ± . . ± . . ± . c er ci s ca na de ns is . ± . . ± . . ± . c or nu s fl or id a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . fa gu s gr an di fo lia . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . fr ax in us am er ic an a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . li ri od en dr on tu lip if er a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . n ys sa sy lv at ic a . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . o st ry a vi rg in ia na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . p op ul us gr an di de nt at a . ± . . ± . . ± . . ± . . ± . p ru nu s se ro ti na . ± . . ± . . ± . . ± . . ± . . ± . . ± . q ue rc us al ba . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . q . co cc in ea . ± . . ± . q . pr in us . ± . . ± . . ± . . ± . . ± . q . ru br a . ± . . ± . . ± . . ± . . ± . . ± . q . ve lu ti na . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . r hu s sp p. . ± . . ± . sa ss af ra s al bi du m . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . u lm us sp p . . ± . . ± . . ± . . ± . . ± . . ± . v ib ur nu m ac er if ol iu m . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . o th er sp ec ie s . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . . ± . j.t. swaim et al. forest ecology and management ( ) – currently dominated by early-seral species, sapling layers are largely dominated by a. saccharum, suggesting a likely future shift towards dominance by late-seral, self-replacing species. in the seedling layer, limited quercus species reproduction is largely occurring on poor sites with low soil ph and limited nutrient availability. . . management implications maintaining quercus in post-harvest stands located on more pro- ductive sites often depends on adequate densities of large advance re- production at time of harvest (sander and clark, ). in some cases, large numbers of smaller quercus seedlings are present, but these stems cannot compete with fast growing early-succession species (e.g., p. serotina and s. albidum) and the dense cover of ruderal herbaceous species (e.g., rubus spp.). in the current study, quercus importance drastically decreased following clearcut harvests, likely due to a lack of large advance reproduction that would have competed better in the post-harvest environment. the exception occurred on dry sites with historically closed canopies, where species like q. prinus successfully established from stump sprouts originating from harvested trees (swain, ), largely due to their higher propensity to sprout from sawlog-sized parents (weigel and peng, ). in a decade-long study of clearcut harvests on the hnf, weigel and peng ( ) found q. prinus was the only species to maintain competitive success prob- abilities above . in year , year , and year . although q. prinus importance was relatively low years after harvest, the high propor- tion of sprout origin trees may allow it to increase in relative im- portance as the stand matures and shorter-lived species such as p. grandidentata and s. albidum begin to die out. other studies have also observed greater importance of quercus species relative to mesophytic competitors following clearcutting on xeric sites (hilt, ; jenkins and parker, ), and a prevalence of stump sprouts in post-harvest stands (arthur et al., ; morrissey et al., ). management strategies and practices following clearcutting may have produced more desirable species composition in our post-harvest stands. pre-commercial crop tree thinnings that release desirable quercus species during stem exclusion often improve quercus stocking and dominance (ward, ; dey et al., ). this technique not only helps control undesirable overstory species, but helps reduce highly competitive midstory species such as o. virginiana and s. albidum, which were abundant in our post-harvest stands. periodic burning favors the establishment of quercus species by increasing light levels on the forest floor and reducing competition (brose et al., ; iverson et al., ). under this regime, high root:shoot ratio and the ability to resprout makes quercus a strong competitor compared to mesophytic competitors that invest resources in aboveground growth. however, this trait makes quercus a poor competitor in large canopy openings in the absence of fire where slower rates of shoot growth create a competitive disadvantage. prior to overstory treatment, prescribed burning may be used as a site pre- paration tool to increase the relative abundance of quercus seedlings by capitalizing on their superior ability to resprout after repeated burning compared to many mesophytic competitors (brose et al. ). post-harvest burning is most effective during the earlier stages of stem exclusion (brose and van lear, ). as stems reach larger dia- meters (> – cm) during later stages, and hence greater bark thicknesses, low intensity fire is less likely to top-kill mesophytic competitors (brose et al., ; schafer et al., ). as an alternative to clearcutting, two- or three-stage shelterwood treatments would also have offered a more viable option for regenerating quercus in these stands. these methods provide adequate light to foster root develop- ment of shade-intolerant reproduction prior to final release while lim- iting establishment of faster growing species such as p. serotina and l. tulipifera (loftis, ; iverson et al., ). when combined with prescribed burning and herbicide treatments that control competitors and allow the well-established root systems of quercus species to resprout and establish as large advance reproduction, these techniques have proven successful and may offer the best hope for regenerating quercus across much of the chr (dey et al., ). once established, thinning should be used to release these crop trees before they lose vigor (dey, ). . conclusions clearcutting upland hardwoods on the hnf without additional control of competing vegetation has created a considerable shift in overstory species importance years after harvest. our results confirm that clearcutting alone is not an adequate disturbance for regenerating or maintaining historically quercus-dominated forests in southern indiana (george, ; jenkins and parker, ; morrissey et al., ). however, on dry sites with historically closed canopies, q. prinus persisted in the canopy of post-harvest stands, although a much reduced importance. the composition of pre-harvest stands was largely de- termined by more intense anthropogenic disturbances that often ac- companied historic land use practices; and forest succession varied along edaphic and topographic gradients. the canopy dominance of quercus in pre-harvest stands suggests that quercus benefited from a historic regime that was dominated by anthropogenic disturbance (i.e., fire, grazing and heavy cutting) that allowed the genus to become dominant across environmental gradients. without management to limit competitors, quercus species will continue to disappear from all but the poorest sites, while species such as a. rubrum and l. tulipifera will continue to dominate across much of the environmental gradient. if clearcut harvesting continued throughout the chr, then it is likely that quercus would continue to decrease in importance, and with each forest rotation there would be fewer and fewer quercus stems until the genus is lost or relegated to the status of a minor associate in the new forest types. as the number of seed-bearing quercus trees decrease in the overstory from one rotation to the next, the likelihood of abundant oak advance reproduction decreases. therefore, relying on quercus stump sprouts to perpetuate quercus in new stands is a losing strategy in the long-term because not all large, older overstory quercus are vigorous sprouters (weigel and peng ). if maintaining quercus species is a primary goal, alternatives to clearcutting should be considered, such as shelterwood prescriptions that include prescribed burning or herbicide control of competing vegetation to promote quercus regeneration. in- termediate stand treatments, such as precommercial crop tree release, offer other viable options for sustaining desired levels of quercus stocking in new stands. acknowledgments charlie jackson provided invaluable assistance with fieldwork. we also thank the personnel of the hoosier national forest for their as- sistance and logistical support. funding for this study was provided by the usda forest service northern research station and hoosier national forest under joint venture agreement -jv- - and the usda national institute of food and agriculture, mcintire stennis cooperative forestry research program (project ind ms). appendix a. supplementary material supplementary data associated with this article can be found, in the online version, at https://doi.org/ . /j.foreco. . . . references abrams, m.d., . adaptations and responses to drought in quercus species of north america. tree physiol. , – . abrams, m.d., scott, m.l., . disturbance-mediated accelerated succession in two michigan forest types. for. sci. , – . j.t. swaim et al. forest ecology and management ( ) – https://doi.org/ . /j.foreco. . . http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h allen, c.d., macalady, a.k., chenchouni, h., bachelet, d., mcdowell, n., vennetier, kitzberger, t., rigling, a., breshears, d.d., hogg, e.h., gonzalez, p., fensham, r., zhang, z., castro, j., demidova, n., limp, j.-h., allard, g., running, s.w., semerci, a., cobb, n., . a global overview of drought and heat-induced tree mortality reveals emerging climate change risks for forests. for. ecol. manage. , – . arthur, m.a., muller, r.n., costello, s., . species composition in a central hardwood forest in kentucky years after clearcutting. am. midland naturalist , – . bates, d., mächler, m., bolker, b., walker, s., . fitting linear mixed-effects models using lme . j. stat. soft. , – . beers, t.w., dress, p.e., wensel, l.c., . aspect transformation in site productivity research. j. for. , – . boring, l.r., monk, c.d., swank, w.t., . early regeneration of a clearcut southern appalachian forest. ecology , – . brose, p.h., van lear, d.h., . responses of hardwood advance regeneration to sea- sonal prescribed fires in oak-dominated shelterwood stands. can. j. for. res. , – . brose, p.h., dey, d.c., phillips, r.j., waldrop, t.a., . a meta-analysis of the fire-oak hypothesis: does prescribed burning promote oak reproduction in eastern north america? for. sci. , – . burns, r.m., honkala, b.h., . silvics of north america: volume . hardwoods. united states department of agriculture, forest service. campione, m.a., nagel, l.m., webster, c.r., . herbaceous-layer community dy- namics along a harvest-intensity gradient after years of consistent management. open j. for. , – . church, t.w., godman, r.m., . the formation and development of dormant buds in sugar maple. for. sci. , – . clark, f.b., watt, r.f., . silvicultural methods for regenerating oaks. in: white, d.e., roach, b.a. (eds.), the silviculture of oaks and associated species. usda forest service, northeastern forest experiment station, upper darby, pa, pp. – . clarke, k.r., . non-parametric multivariate analysis of changes in community structure. aust. j. ecol. , – . demchik, m., sharpe, w.e., . the effect of soil nutrition, soil acidity, and drought on northern red oak (quercus rubra l.) growth and nutrition on pennsylvania sites with high and low red oak mortality. for. ecol. manage. , – . dey, d.c., . the ecological basis for oak silviculture in eastern north america. in: mcshea, w.j., healy, w.m. (eds.), oak forest ecosystems: ecology and management for wildlife. johns hopkins university press, baltimore, md, pp. – . dey, d.c., royo, a.a., brose, p.h., hutchinson, t.f., spetich, m.a., stoleson, s.h., . an ecologically based approach to oak silviculture: a synthesis of years of oak ecosystem research in north america. revista colombia forestal , – . dey, d.c., . sustaining oak forests in eastern north america: regeneration and re- cruitment, the pillars of sustainability. forest sci. , – . ellison, a.m., banks, m.s., clinton, b.d., colburn, e.a., elliot, k., ford, c.r., foster, d.r., kloeppel, b.d., knoepp, j.d., lovett, g.m., mohan, j., orwig, d.a., rodenhouse, n.l., sobczak, w.v., stinson, k.a., stone, j.k., swan, c.m., thompson, j., houe, b.v., webster, j.r., . loss of foundation species: consequences for the structure and dynamics of forested ecosystems. front. ecol. environ. , – . fei, s., kong, n., steiner, k.c., moser, w.k., steiner, e.b., . change in oak abundance in the eastern united states from to . for. ecol. manage. , – . futuyma, d.j., gould, f., . associations of plants and insects in a deciduous forest. ecol. monogr. , – . george, d.w., . tree regeneration response to clearcut harvesting on the hoosier national forest. ms thesis, purdue university, west lafayette, in. george, d.w., fischer, b.c., . the effects of site and age in tree reproduction in young upland hardwood clearcuts. in: rink, g., budelsky, c.a. (eds.), proceedings of the seventh central hardwood conference. nc- . u.s. department of agriculture, forest service, carbondale, il, pp. – . guyette, r.p., dey, d.c., stambaugh, m.c., . fire and human history of a barren- forest mosaic in southern indiana. am. midl. nat. , – . guyette, r.p., dey, d.c., stambaugh, m.c., muzika, r.-m., . fire scars reveal variability and dynamics of eastern fire regimes. in: fire in eastern oak forests: delivering science to land managers, proceedings of a conference. usda forest service, ohio state university, columbus, ohio, pp. – . hilt, d.e., . species composition of young central hardwood stands that developed after clearcutting. in: dawson, j.o., majerus, k.a. (eds.), proceedings of the th central hardwood forest conference, - april . department of forestry, university of illinois urbana, champaign, ill, champaign, illinois, pp. – . homoya, m.a., abrell, d.b., aldrich, j.r., post, t.w., . the natural regions of indiana. proc. indiana acad. sci. , – . iverson, l.r., hutchinson, t.f., peters, m.p., yaussy, d.a., . long-term response of oak-hickory regeneration to partial harvest and repeated fires: influence of light and moisture. ecosphere :e . . /ecs . . jenkins, m.a., . the history of human disturbance in forest ecosystems of southern indiana. in: swihart, r.k., saunders, m.r., kalb, r.a., haulton, g.s., michler, c.h. (eds.), the hardwood ecosystem experiment: a framework for studying responses to forest management. u.s. department of agriculture, forest service, northern forest experiment station gen. tech. rep. nrs-p- , newtown square, pa, usa, pp. – . jenkins, m.a., pallardy, s.g., . the influence of drought on red oak group species growth and mortality in the missouri ozarks. can. j. for. res. , – . jenkins, m.a., parker, g.r., . composition and diversity of woody vegetation in silvicultural openings of southern indiana forests. for. ecol. manage. , – . johnson, p., shifley, s., rogers, r., . the ecology and silviculture of oaks. cabi publishing, new york. kabrick, j.m., shifley, s.r., jensen, r.g., larsen, d.r., grabner, j.k., . oak forest composition, site quality, and dynamics in relation to site factors in the southeastern missouri ozarks. in: spetich, m.a. (ed.), upland oak ecology symposium: history, current conditions, and sustainability. gen. tech. rep. srs- . usda forest service, southern research station, asheville, nc, pp. . kuznetsova, a., brockhoff, p.b., christensen, r.h.b., . lmertest package: tests in linear mixed effects models. j. stat. soft. , – . lambers, j.h.r., clark, j.s., . effects of dispersal, shrubs, and density-dependent mortality on seed and seedling distributions in temperate forests. can. j. for. res. , – . latham, r.e., ricklefs, r.e., . continental comparisons of temperate-zone tree spe- cies diversity. in: ricklefs, r.e., schluter, d. (eds.), species diversity in ecological communities: historical and geographical perspectives. university of chicago press, chicago, pp. – . levanič, t., Čater, m., mcdowell, n.g., . associations between growth, wood anatomy, c isotope discrimination and mortality in a quercus robur forest. tree physiol. , – . loftis, d.l., . a shelterwood method for regenerating red oak in the southern appalachians. for. sci. , – . lorimer, c.g., . causes of the oak regeneration problem. in: oak regeneration: serious problems, practical recommendations, usda forest service general technical report se- . southeastern forest experiment station, asheville, north carolina, pp. – . mccune, b., grace, j.b., urban, d.l., . analysis of ecological communities. mjm software design gleneden beach, oregon, usa. mccune, b., mefford, m.j., . pc-ord: multivariate analysis of ecological data. mjm software design, gleneden beach, oregon, usa. mcewan, r.w., dyer, j.m., pederson, n., . multiple interacting ecosystem drivers: toward an encompassing hypothesis of oak forest dynamics across eastern north america. ecography , – . mcgee, c.e., . clearcutting in upland hardwoods: panacea or anathema? in: hay, r.l., woods, f.w., deselm, h. (eds.), proceedings of the th central hardwood forest conference. university of tennessee, knoxville, pp. – . meier, a.r., saunders, m.r., . initial observations on tree mortality following a se- vere drought in in two indiana state forests and implications for long-term compositional dynamics. in: groninger, j.w., holzmueller, e.j., nielsen, c.k., dey, d.c. (eds.), proceedings of the th central hardwood forest conference. gen. tech. rep. nrs-p- . u.s. department of agriculture, forest service, northern research station, newtown square, pa, pp. – . morrissey, r.c., jacobs, d.f., seifert, j.r., fischer, b.c., kershaw, j.a., . competitive success of natural oak regeneration in clearcuts during the stem exclusion stage. can. j. for. res. , – . nelson, t.c., . chestnut replacement in the southern highlands. ecology , – . neter, j., kutner, m.h., nachtsheim, c.j., wasserman, w., . applied linear regression models. irwin, chicago. nowacki, g.j., abrams, m.d., . the demise of fire and “mesophication” of forests in the eastern united states. bioscience , – . oliver, c.d., larson, b.c., . forest stand dynamics, update edition. john wiley & sons inc. opler, p.a., . oaks as evolutionary islands for leaf- mining insects. am. sci. , – . pairon, m., chabrerie, o., casado, c.m., jacquemart, a.-l., . sexual regeneration traits link to black cherry (prunus serotina ehrh.) invasiveness. acta oecol. , – . pallardy, s.g., rhoads, j.l., . morphological adaptations to drought in seedlings of deciduous angiosperms. can. j. for. res. , – . parker, w.c., dey, d.c., . influence of overstory density on ecophysiology of red oak (quercus rubra) and sugar maple (acer saccharum) seedlings in central ontario shel- terwoods. tree physiol. , – . plieninger, t., rolo, v., moreno, g., . large-scale patterns of quercus ilex, quercus suber, and quercus pyrenaica regeneration in central-western spain. ecosystems , – . ponder jr, f., . ecological regions and soil conditions in the hoosier-shawnee eco- logical assessment area. the hoosier-shawnee ecological assessment. gen. tech. rep. nc . st. paul, mn: us department of agriculture, forest service, north central research station, pp. – . r core team, . r: a language and environment for statistical computing. r foundation for statistical computing, vienna, austria. https:/www.r-project.org/. last accessed on may , . ranius, t., jansson, n., . the influence of forest regrowth, original canopy cover and tree size on saproxylic beetles associated with old oaks. biol. conserv. , – . roach, b.a., gingrich, s.f., . even-aged silviculture for upland central hardwoods. forest service, usda forest service, agricultural handbook, p. sander, i.l., clark, f.b., . reproduction of upland hardwood forests in the central states. agriculture handbook no. . u.s. department of agriculture, forest service. sander, i.l., graney, d.l., . regenerating oaks in the central states. in: oak regeneration: serious problems practical recommendations. usda forest service gtr se- , pp. – . schafer, j.l., breslow, b.p., hohmann, m.g., hoffmann, w.a., . relative bark thickness is correlated with tree species distributions along a fire frequency gradient. fire ecol. , – . shrestha, b.b., . quercus semecarpifolia sm. in the himalayan region: ecology, ex- ploitation and threats. j. himalayan sci. , – . sieber, e., munson, c.a., . looking at history: indiana's hoosier national forest region, to . us dept. of agriculture, forest service. soil survey staff, natural resources conservation service, usda, . web soil survey j.t. swaim et al. forest ecology and management ( ) – http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://websoilsurvey.nrcs.usda.gov/. accessed october , . southwood, t.r.e., . the number of species of insect associated with various trees. j. anim. ecol. , – . standiford, r.b., fischer, b.c., . fifteen year results of three harvesting methods on composition and development of regeneration in southern indiana upland hard- woods. in: garrett, h.e., cox, g.s. (eds.), proceedings of the rd central hardwood forest conference, - september , columbia, missouri. university of missouri press, columbia, missouri, pp. – . swaim, j.t., . stand development and the competitive ability of oak (quercus spp.) following silvicultural clearcutting on the hoosier national forest. m.s. thesis. purdue university, west lafayette, in, p. swaim, j.t., dey, d.c., saunders, m.r., weigel, d.r., thornton, c.d., kabrick, j.m., jenkins, m.a., . predicting the height growth of oak species (quercus) re- production over a -year period following clearcutting. for. ecol. manage. , – . trimble, g.r., hart, g., . an appraisal of early reproduction after cutting in northern appalachian hardwood stands. northeastern forest experiment station, forest service, us department of agriculture. van kley, j., parker, g., franzmeier, d., randolph, j., . field guide: ecological classification of the hoosier national forest and surrounding areas of indiana. usda forest service, hoosier national forest. vanhellemont, m., wauters, l., baeten, l., bijlsma, r.-j., de frenne, p., hermy, m., verheyen, k., . prunus serotina unleashed: invader dominance after years of forest development. biol. invasions , – . vodka, s., konvicka, m., cizek, l., . habitat preferences of oak-feeding xylophagous beetles in a temperate woodland: implications for forest history and management. j. insect conserv. , – . walker, l.r., wardle, d.a., bardgett, r.d., clarkson, b.d., . the use of chron- osequences in studies of ecological succession and soil development. j. ecol. , – . ward, j.s., . intensity of precommercial crop tree release increases diameter growth and survival of upland oaks. can. j. for. res. , – . weigel, d.r., peng, c.-y.j., . predicting stump sprouting and competitive success of five oak species in southern indiana. can. j. for. res. , – . woods, f.w., shanks, r.e., . natural re- placement of chestnut by other species in the great smoky mountains national park. ecology , – . zavaleta, e.s., hulvey, k.b., fulfrost, b., . regional patterns of recruitment success andfailure in two endemic california oaks. divers. distributions , – . j.t. swaim et al. forest ecology and management ( ) – http://websoilsurvey.nrcs.usda.gov/ http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h http://refhub.elsevier.com/s - ( ) - /h 	overstory species response to clearcut harvest across environmental gradients in hardwood forests 	introduction 	methods 	harvest treatments 	study sites 	plot establishment and sampling 	data preparation 	data analysis 	results 	overstory ordination analysis 	site conditions and compositional shifts 	contemporary seedling and sapling species composition and abundance 	discussion 	pre-harvest conditions and historic disturbance 	post-harvest shifts in composition 	management implications 	conclusions 	acknowledgments 	supplementary material 	references sunitinib plus paclitaxel in patients with advanced esophageal cancer: a phase ii study from the hoosier oncology groupe journal of thoracic oncology • volume , number , april brief report abstract: the combination of sunitinb ( . mg orally daily) + pacli- taxel ( mg/m intravenously on days , , every weeks) was examined in patients with advanced esophageal or gastroesophageal junction cancer, and progression-free survival (pfs) was compared to that of historical controls. the end points included response rate, overall survival, and toxicities. twenty-eight patients were enrolled at six centers. median age was . years. the -week pfs rate was % ( % confidence interval [ci], – %). three ( %) of evaluable patients had a response ( complete response and partial response) ( % ci, − %). median overall survival was days ( % ci, – days). grade / toxicities included leukopenia/ neutropenia ( %), anemia ( %), fatigue ( %), and hemorrhage ( %). there were four grade toxicities including upper gastroin- testinal hemorrhage (n = ), gastrointestinal/esophageal fistula (n = ), and unexplained death (n = ). in our study, we found that suni- tinib + paclitaxel in patients with advanced esophageal or gastroe- sophageal junction cancer had a -week pfs no better than the pfs of historical controls. the combination also had a high rate of serious toxicities and will not be pursued. key words: sunitinib, paclitaxel, esophageal, gastroesophageal junction, angiogenesis, tyrosine kinase inhibitors. (j thorac oncol. ; : – ) despite advances in staging and treatment for patients with esophageal cancer, the relative -year survival rate for all stages remained at an estimated % between and . response rates (rrs) to classic chemotherapeutic regimens range from to % and are typically short in duration. , angiogenesis is a critical factor in tumor growth and meta- static spread. in patients with esophageal cancer, to % of them have overexpression of vascular endothelial growth fac- tor (vegf), which has been correlated with greater microves- sel density, advanced disease, and poor prognosis. inhibition of angiogenesis with compounds like sunitinib maleate, a well-tolerated oral inhibitor of tyrosine kinase with activity against vegf receptors (types – ), platelet-derived growth factor a and platelet-derived growth factor , c-kit, and fetal liver tyrosine kinase receptor , is a novel therapeutic approach that has demonstrated activity in a variety of malig- nancies. in preclinical studies, gastric and esophageal cancer cell lines overexpressed vegf, proliferated in the presence of recombinant vegf, and were inhibited in a dose-dependent manner by sunitinib. sunitinib has been studied for locally advanced esophageal cancer and studies in patients with relapsed/refractory metastatic disease are ongoing. on the basis of this preclinical and clinical work, the known overex- pression of vegf in esophageal cancer, the hypothesized syn- ergy between vegf inhibition, and the antiangiogenic effects of taxane therapy, we conducted this multi-institutional phase ii study to evaluate the efficacy of sunitinib + paclitaxel in patients with advanced esophageal or gastroesophageal (ge) junction carcinoma. patients and methods patients enrolled on study included adults with histo- logically proven recurrent or metastatic esophageal or ge junction squamous cell or adenocarcinoma, who had received less than two chemotherapy regimens for locally advanced or metastatic disease, and had measurable and/or evaluable dis- ease as per the response evaluation criteria in solid tumors (recist) criteria. all patients had an eastern cooperative oncology group performance status of to . key exclusion criteria included prior exposure to an epidermal growth factor receptor inhibitor or any other antiangiogenic agents. each treatment cycle was defined as days and included paclitaxel mg/m intravenous infusion over minutes on *indiana university simon cancer center, indianapolis, indiana; †indiana university medical center, indianapolis, indiana; ‡south carolina oncology associates pa, columbia, south carolina; §hoosier oncology group,  indianapolis,  indiana;  •medical  consultants,  pc,  muncie,  indiana; ¶northern indiana cancer research consortium, south bend, indiana; #community hospital, munster, indiana; **cancer care center of southern indiana, bloomington, indiana; ††medical and surgical specialists, llc, galesburg, illinois; and ‡‡indiana university division of biostatistics indianapolis, indiana. disclosure: this work was supported by pfizer. address for correspondence: jordan schmitt, md, barnhill drive, rt , indianapolis, in . e-mail: jmw @iupui.edu copyright © by the international association for the study of lung cancer issn: - / / - sunitinib plus paclitaxel in patients with advanced esophageal cancer a phase ii study from the hoosier oncology group jordan m. schmitt, md,*† scott r. sommers, md,‡ william fisher, md,§• rafat ansari, md,§¶ erwin robin, md,§# karuna koneru, md,§** john mcclean, md,§†† ziyue liu, phd,‡‡ yan tong, ms,‡‡ and nasser hanna, md*†§ copyright © by the international association for the study of lung cancer journal of thoracic oncology • volume , number , april sunitinib plus paclitaxel in esophageal cancer patients days , , and and sunitinib maleate . mg orally daily. participants were scheduled to receive four cycles of the com- bination therapy. treatment was discontinued upon disease progression or unacceptable toxicity. after four cycles, pacli- taxel was discontinued and each patient was allowed to con- tinue on sunitinib maleate for two more cycles provided he/ she did not have disease progression, intolerable side effects, or adverse reactions. patients were evaluated at the beginning of each treat- ment cycle. before odd-numbered cycles, a disease evalu- ation per recist criteria was performed. at the end of the study, patients, including those who discontinued pro- tocol therapy early, were evaluated days after the last dose. patients were then serially followed up for duration of response and for survival. all toxicities were measured according to the national cancer institute common toxicity criteria version . . the primary end point was the rate of nonprogressive disease at months. nonprogressive disease was defined as a complete response, partial response, or stable disease. a pre- determined improvement from to % in nonprogressive disease rate was deemed reasonable for the proposed regimen as two prior phase ii trials of paclitaxel therapy alone dem- onstrated a median duration of response ranging from to days , and we anticipated prolonged duration of response with the addition of antiangiogenic therapy. with subjects, the study had more than % power based on one-sided test with error rate controlled at %. results twenty-eight patients were enrolled at six participating centers. patient demographics and disease characteristics are summarized in table . over % of the participants were men, with a median age at enrollment of . years (age range of – years). all tumors were located in the esophagus ( / or ge junction [ / ]). ninety-three percent of the tumors (n = ) were adenocarcinoma and % (n = ) were squamous cell carcinoma. nine of patients had received prior chemotherapy. the median follow-up was days, with a range of to days. six patients ( %) completed less than one cycle of therapy. seven patients ( %) finished all four cycles and went on to receive maintenance treatment (median length of maintenance treatment was days). at the end of the study, patients ( %) were evaluable. five patients were not evaluable owing to no postbase- line assessment. reasons for coming off study included death caused by progressive disease (n = [ %]), death caused by other reasons (n = [ %]), and refusal to fol- low-up (n = [ %]). among the patients, ( %) had dose modification and ( %) had dose delay of sunitinib. fifteen ( %) had dose modification of paclitaxel, and ( %) had dose delay of paclitaxel. toxicity data for all patients are summarized in tables through . the most common toxicities reported included fatigue, anemia, and leukopenia/neutropenia. grade and grade toxicities included leukopenia/neutropenia ( %), anemia ( %), fatigue ( %), and hemorrhage ( %). there were four grade toxicities including two deaths caused by upper gastrointestinal hemorrhage and one caused by gastro- intestinal/esophageal fistula (table ). three of the four events were considered treatment-related deaths. sunitinib-related toxicities including diarrhea, rash, hemorrhage, thrombosis, and hypertension were also observed. the -week progression-free survival (pfs) rate was % ( / ) ( % ci, – %) for the combined regimen. of the evaluable patients, ( %) had a measurable response by recist criteria ( % ci, − %). a complete response was observed in one patient and partial response observed in two patients. for the three patients who responded, the dura- tions of response were , , and days, respectively. for the intention-to-treat patients, the -year overall survival (os) rate was % ( % ci, − %). median os was days ( % ci, – days). the median pfs was days ( % ci, – ). table . demographics (intention to treat n = ) characteristics category/ statistics n/value % sex female male race caucasian african american ethnicity non-hispanic not reported age median . yr range – yr ecog ps tumor site esophagus gastroesophageal junction histology adenocarcinoma squamous cell carcinoma prior treatment yes ( with prior chemo) no ecog ps, eastern cooperative oncology group performance status. table . selected hematologic toxicities irrespective of causality attributed (n = ) grade grade grade / leukopenia/neutropenia % % % anemia % % % thrombocytopenia % — % febrile neutropenia (fever without clinically or microbiologically documented infection) (anc < . , fever ≥ . °c) % — % anc, absolute neutrophil count. copyright © by the international association for the study of lung cancer schmitt et al. journal of thoracic oncology • volume , number , april discussion the results of this multi-institutional phase ii trial are, to our knowledge, the first reported study of sunitinib in advanced esophageal/ge junction cancer. this combination did not meet our predetermined goal of improved nonprogres- sive disease from to % at weeks. the majority of the patients completed less than two cycles and toxicities were significant with at least three treatment-related deaths. paclitaxel alone as first-line therapy for metastatic disease has demonstrated an rr ranging from to % in two prior phase ii studies of patients with advanced esophageal cancer. , this is inferior to rrs reported with use of regimens such as epirubicin + cisplatin + fluorouracil and epirubicin + oxaliplatin + capecitabine or paclitaxel containing combination regimens that have yielded higher rrs ( – %) at the expense of higher toxicity rates, including docetaxel + cisplatin + -fluorouracil, and oxaliplatin + paclitaxel. our study failed to show similar rrs with a reported overall rr of only % and a -week pfs rate estimated at % with increased toxicities. vegf inhibitors have been studied in patients with advanced esophageal cancer. in a phase ii study combining irino- tecan, cisplatin and bevacizumab, a % overall rr was noted with a median time to progression of . months and a median survival of . months. a second study looking at docetaxel + cisplatin + -fu + bevacizumab demonstrated an improvement in pfs at months from a historic control of % for docetaxel + cisplatin + fluorouracil alone to % with the addition of beva- cizumab. grade or toxicities in both of these studies were fewer than appreciated in our study. another study evaluating bevacizumab + cisplatin and capecitabine in advanced gastric cancer reported fewer serious toxicities than observed in our study, and although this study showed significant improvements in the overall rr and pfs, it failed to meet its primary end point of improved os ( . versus . months [p = . ]). sunitinib has been evaluated in esophageal cancer in the adjuvant setting, looking at neoadjuvant irinotecan + cisplatin combined with radiation followed by esophagectomy followed by adjuvant sunitinib therapy. in this phase ii study, patients were enrolled; following esophagectomy only patients were fit enough to receive sunitinib therapy. median os in this locally advanced population was months with a -year os of around %. specific toxicity data are not yet available. currently, there is an ongoing phase ii study comparing suni- tinib single agent to placebo therapy in relapsed or refractory esophageal or ge junction carcinoma (clincaltrials.gov id #nct ). the sunitinib-specific toxicities observed in our study were similar to those previously reported; however, the increased incidence of grade / / hemorrhage was con- cerning with this combination and was not previously reported in therapy with taxanes alone, combination regimens contain- ing taxanes, or with other vegf inhibitors. in conclusion, our study of the combination of sunitinib + paclitaxel in advanced esophageal and ge junction cancer was negative as it did not meet our primary end point of % pfs at weeks, and with a high rate of serious toxicities, including four grade- events, this combination will not be pursued further. references . jemal a, siegel r, xu j, ward e. cancer statistics, . ca cancer j clin ; : – . . shah ma, schwartz gk. treatment of metastatic esophagus and gastric cancer. semin oncol ; : – . table . selected nonhematologic toxicities irrespective of causality attributed (n = ) grade grade grade / grade hemorrhage/bleeding % % % % thrombosis/thrombus/embolism — % % — fistula, gi/esophagus % — % % rash/dermatologic complaints — % % — dyspnea — % % — neuropathy: motor % — % — allergic reaction/hypersensitivity (including drug fever) % — % — pulmonary (effusion, pneumonitis nos) % % % — fatigue (asthenia, lethargy, and malaise) % — % — diarrhea % — % — dysphagia % — % — hyponatremia % — % — hyperglycemia % — % — death nos — % — % gi, gastrointestinal; nos, not otherwise specified. table . grade toxicity and treatment-related death evaluable population (n = ) n % cycle consequence attribution hemorrhage, gi/ upper gi nos maintenance death possible fistula, gi/ esophagus death unrelated death nos death probably hemorrhage, gi/ stomach death possible gi, gastrointestinal. copyright © by the international association for the study of lung cancer journal of thoracic oncology • volume , number , april sunitinib plus paclitaxel in esophageal cancer patients . cunningham d, starling n, rao s, et al.; upper gastrointestinal clinical studies group of the national cancer research institute of the united kingdom. capecitabine and oxaliplatin for advanced esophagogastric cancer. n engl j med ; : – . . kleespies a, guba m, jauch kw, bruns cj. vascular endothelial growth factor in esophageal cancer. j surg oncol ; : – . . chow lq, eckhardt sg. sunitinib: from rational design to clinical efficacy. j clin oncol ; : – . . orestis l, annett m, florian h, et al. analysis of anti-proliferative and chemosensitizing effects of sunitinib on human esophagogastric cancer cells: synergistic interaction with vandetanib via inhibition of multi-receptor tyrosine kinase pathways. int j cancer ; : – . . knox jj, wong r, darling ge, et al. adjuvant sunitinib for locally advanced esophageal cancer: results of aphase ii trial. j clin oncol ; (suppl). . george s, blay jy, casali pg, et al. clinical evaluation of continuous daily dosing of sunitinib malate in patients with advanced gastrointestinal stromal tumour after imatinib failure. eur j cancer ; : – . . ajani ja, ilson dh, daugherty k, pazdur r, lynch pm, kelsen dp. activity of taxol in patients with squamous cell carcinoma and adenocarcinoma of the esophagus. j natl cancer inst ; : – . . ilson dh, wadleigh rg, leichman lp, kelsen dp. paclitaxel given by a weekly -h infusion in advanced esophageal cancer. ann oncol ; : – . . van cutsem e, moiseyenko vm, tjulandin s, et al. v study group. phase iii study of docetaxel and cisplatin plus fluorouracil compared with cisplatin and fluorouracil as first-line therapy for advanced gastric cancer: a report of the v study group. j clin oncol ; : – . . kaechele v, moehler m, lutz mp, et al. a phase i/ii study of oxaliplatin and paclitaxel in patients with non-resectable cancer of the oesophagus and adenocarcinoma of the gastro-oesophageal junction: a study of the arbeitsgemeinschaft internistische onkologie. cancer chemother pharmacol ; : – . . shah ma, ramanathan rk, ilson dh, et al. multicenter phase ii study of irinotecan, cisplatin, and bevacizumab in patients with metastatic gastric or gastroesophageal junction adenocarcinoma. j clin oncol ; : – . . shah ma, jhawer m, ilson dh, et al. phase ii study of modified docetaxel, cisplatin, and fluorouracil with bevacizumab in patients with metastatic gastroesophageal adenocarcinoma. j clin oncol ; : – . . ohtsu a, shah ma, van cutsem e , et al. bevacizumab in combination with chemotherapy as first-line therapy in advanced gastric cancer: a randomized, double-blind, placebo-controlled phase iii study. j clin oncol ; : – . 	sunitinib plus paclitaxel in patients with advanced esophageal cancer: a phase ii study from the hoosier oncology group 	patients and methods 	results 	discussion 	references pei regimen: a therapeutic option in small cell lung cancer? a retrospective monoinstitutional analysis of consecutive cases boni et al. journal of translational medicine ( ) : doi . /s - - - protocol open access pei regimen: a therapeutic option in small cell lung cancer? a retrospective monoinstitutional analysis of consecutive cases corrado boni *, maria pagano , licia baldi , roberta gnoni , luca braglia , luisa savoldi and francesca zanelli abstract objectives: combination chemotherapy is very active in small cell lung cancer (sclc), although no improvement in overall survival (os) has been done in the last years, with cisplatin-etoposide (pe) still considered the world-wide standard, with an average median survival of about – months in patients with extended disease (ed). in , a randomized trial of the hoosier group in ed patients showed a significant advantage in overall survival in patients treated with pei (cisplatin, etoposide and ifosfamide), compared to pe. despite that, pei regimen has not become a commonly used regimen in sclc. materials and methods: here we present a retrospective analysis of consecutive patients ( males and females) with sclc that were treated at our institution with pei regimen: cisplatin mg/m , etoposide mg/m and ifosfamide mg/m , day to , every weeks. patients received a total of cycles of chemotherapy, with a mean of , cycles per patient. median age was (range – ); performance status (ps) was in patients ( %), in patients ( %) and in patients ( %). results: in limited disease (ld) patients partial response (pr) rate was %, and complete response (cr) was %. in ed patients we observed % of pr and % of cr. median time to progression (ttp) was . months in ld and . months in ed with median overall survival (os) of . and . months, respectively. toxicity was manageable, with a high dose intensity. conclusions: pei regimen, in our opinion, may be a possible therapeutic option, with high activity and an acceptable toxicity profile. trial registration: clinicaltrials.gov identifier: nct . institutional review board that approved the study: institutional review board of reggio emilia, azienda ospedaliera s.maria nuova/irccs. keywords: small cell lung cancer, extended disease, limited disease, chemotherapy, cisplatin, etoposide, ifosfamide * correspondence: boni.corrado@asmn.re.it department of oncology and advanced technologies, oncology unit, azienda ospedaliera s.maria nuova/irccs of reggio emilia, viale risorgimento, , reggio emilia, (italy) full list of author information is available at the end of the article © boni et al.; licensee biomed central. this is an open access article distributed under the terms of the creative commons attribution license (http://creativecommons.org/licenses/by/ . ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. the creative commons public domain dedication waiver (http://creativecommons.org/publicdomain/zero/ . /) applies to the data made available in this article, unless otherwise stated. https://clinicaltrials.gov/ct /show/nct ?term=nct &rank= mailto:boni.corrado@asmn.re.it http://creativecommons.org/licenses/by/ . http://creativecommons.org/publicdomain/zero/ . / boni et al. journal of translational medicine ( ) : page of introduction neuroendocrine tumors account for approximately % of lung cancers; most of them (≈ %) are sclc [ ]. in , about , new cases of sclc have occurred in the united states, in the majority attributable to cigarette smoking [ ]. sclc tends to disseminate early in the course of its natural history and to grow quickly. approximately % to % of patients present with brain metastases (bm) at the time of initial diagnosis, and in an additional % to % will develop bm some time during course of their disease [ ]. although the incidence of sclc is decreased in recent years, it remains a therapeutic challenge, as survival in patients with limited disease has not changed markedly over the past years, reaching approximately % to % at years in the best published series of patients treated with a multimodality approach [ ] sclc has a propensity for early hematogenous spread and for association with paraneoplastic syndromes and is characterized by its rapid doubling time, high growth fraction and high, but short-lasting, chemosensitivity. paraneoplastic syndromes are more frequently seen in patients with limited-stage sclc. than in those with extensive-stage disease, but their presence is not un- equivocally prognostically favourable [ , ]. chemotherapy is the cornerstone of treatment and cisplatin-etoposide (pe) is still considered the world- wide standard since over years. approximately one-third of patients diagnosed with sclc present with ld, with a median survival time of – months, compared to – months for patients with ed. an overall response to combination chemo- therapy is achieved in - % of ld patients and in - % of patients with ed [ ]. also the complete response rate is influenced by the extension of the disease, and is significantly higher in pa- tients with limited disease ( - %) compared to pa- tients with extended disease ( - %) [ ]. since s, several studies have been conducted adding ifosfamide, an analogue of cyclophosphamide, to the pe combination, based on his single-agent activity and low toxicity [ ]. the randomized trial of the hoosier oncology group in ed pts showed a significant advantage in overall survival in the arm treated with pei, compared to pe [ – ]. despite that, pei has not become a commonly used regimen in sclc, maybe because its less conveni- ent schedule, requiring consecutive days of chemother- apy. relying on its efficacy and manageable toxicity, pei was the reference schedule for the treatment of sclc, in our centre since years. primary objective of the present retrospective study is to evaluate the activity of pei in the sclc in terms of: response rate, time to progression and overall survival, in patient with ld or ed, treated in first line. secondary objective were: the analysis of the dose intensity of chemotherapy and the evaluation of the tolerability in terms of toxicity and adverse events. material and methods between december and december , con- secutive patients, older than years, performans status (ps) < , with localized (ld) and extensive stage (ed) sclc presenting to the medical oncology unit of reggio emilia (italy) were treated, in first line, with pei regimen. all eligible patients had histologically or cytologically proven sclc, with measurable disease defined by recist criteria, and received at least one cycle of chemotherapy. patients with central nervous system (cns) metastases were included in the study. pretreatment evaluation included physical examination, bronchoscopy, total body computed tomography (ct), and blood tests including a complete blood count, platelet count, serum creatinine and liver enzyme determination. the study was approved by our local ethics commit- tee. all patients gave written informed consensus before starting chemotherapy with pei regimen, and at the time of data analysis all living patients gave another written informed consent for the use of their personal data in the study analysis. treatment regimen the pei regimen consisted of cisplatin (p) mg/m intravenously (iv) infused in minute, on days to with pre hydration with ml of % dextrose and ml of % normal saline over hours and mannitol- induced diuresis, etoposide (e) mg/m iv infused in minute, on days to and ifosfamide (i) mg/m iv infused in minute, on days to . mesna was ad- ministered at the dose of mg total by iv bolus before the first dose of ifosfamide and then after and hours at the same dose. urinalysis was performed daily during the ifosfamide treatment. courses were repeated every week for – cycles unless demonstration of disease progression or inacceptable toxicity. all patients received intravenous -ht antagonists and dexamethasone as antiemetic prophylaxis, for the days of treatment. all drugs were reduced by % in case of g haem- atological toxicity. in case of febrile neutropenia or thrombocytopenia, requiring platelet transfusion, all drugs were reduced by % and, in such cases, no re- escalation was performed. the use of colony stimulating factors were allowed but not included as part of the standard treatment regimen. drug administration was postponed by week if there was no full haematological recovery (granulocyte count > boni et al. journal of translational medicine ( ) : page of /mm , platelets count > . /mm ) from the prior course of chemotherapy. after week of delay, if the granulocyte count was between /mm and /mm and/or platelets were between . /mm and . /mm , the doses of all drugs were reduced by %. sequential chest irradiation ( gy in fractions) was also administered to patients with ld achieving cr or partial response (pr). in patient with ld who achieved complete response (cr) or partial response (pr) prophylactic cranial irradi- ation (pci) was administered; the radiation dose to the whole brain was . gy in fractions. in selected cases of ed patients the whole brain radio- therapy ( . gy in fractions) was allowed according to clinical evaluation. response and toxicity evaluation clinical response was evaluated according to recist criteria (version . ). clinical response was evaluated by repeating ct scan and all previously abnormal tests. after three and six cycles. toxicity, based on ctcae toxicity criteria (version . ), was evaluated weekly and at the end of treatment. two patients that received only course of chemo- therapy were not considered evaluable for response. table study population: patients treated with pei n° patients (%) sex male female age (yrs) median range - stage ld ed ps (who) statistical analysis overall survival (os) time was measured from the date of diagnosis until death. survival time of patients who did not experienced the event considered during follow up observation was cen- sored at the time of the last follow-up. os analysis was performed in all the patients. time to progression (ttp) was calculated from the start of treatment to the date of first disease progression or relapse. ttp of patients who did not experienced the event considered during follow up period was censored at the time of the last follow-up or at death, if occurred. ttp analysis was performed in all the patients con- sidered evaluable for response. main statistical analysis has included: descriptive statistics of the main patients characteristics (gender, age, stage, per- formance status, metastatic site), estimation (with kaplan- meier method) and plot of survival probabilities for os and ttp (stratified according to stage). we used the statistical packages r . . , sas . and pasw statistics . (spss, chicago, il) for data ana- lysis and visualization. dose intensity (di is the drug dose delivered per time unit and is expressed as mg/m per week) was calcu- lated for each drug of pei regimen. results between december and december , consecu- tive sclc patients were treated in first line of chemother- apy with pei regimen. the study population was composed of male ( %) and female ( %) with a median age of years (range – ) (table ). twenty-seven patients ( %) had ed and ( %) had ld; the ps, graded according to the world health organization (who), was in cases ( %); in ( %) and in patients ( %). metastatic sites were: lung ( % of the patients), mediastinal node ( %), liver ( %) and bone %. patients had brain metastases at diagnosis. there was no substantial significant difference in meta- static sites between genders: on average, in females group there were a mean of . metastatic sites, compared to . in the males (p-value . , two tailed poisson test). patients received a total of cycles of chemother- apy, with a mean of , cycles per patient. dose intensity (di) was calculated for each drug of pei regimen. based on di, the patients were divided in two groups, if they received < % or ≥ %. the median di of cisplatin was %; ifosfamide was administred with a me- dian di of % and the median di of etoposide was %. response and survival in ld patients an objective response (or), defined as complete response (cr) or partial response (pr), was observed in patients ( %), with % of pr and % of cr. in ed patients an or was observed in cases ( %), with % of pr and % of cr (table ). patients have died, / in ed ( %), / in ld ( %). table response ( pts): limited and extended disease limited disease nv pd no (%) cr no (%) pr no (%) sd no (%) ( ) ( ) ( ) ( ) ( ) extended disease nv pd no (%) cr no (%) pr no (%) sd no (%) ( ) ( ) ( ) ( ) ( ) boni et al. journal of translational medicine ( ) : page of figures and show the survival probabilities for os and ttp up to the fifth year of follow up, stratified ac- cording to stage. in the patients, the median time to progression was . months ( %. ci . - ): . months in ld and . months in ed. the median os in the patients was months ( % ci – . ): . months in ld and . months in ed patients. toxicity despite the use of antiemetic treatment, nausea was very common on the days of cisplatin infusion; however vomiting occurred only in % of patients. myelosuppression was the most frequent adverse event, with g - neutropenia in %, anemia in % and thrombocytopenia in % of the patients. % of patients had febrile neutropenia. three toxic deaths were observed, all in patients with ps ≥ : two patients with sepsis and febrile neutropenia and one for heart failure in a cardiopatic patient. figure overall survival in limited and extended disease. nonhematologic toxicity consisted mainly of asthe- nia ( %), mucositis ( %), nausea and vomiting ( %) (table ). discussion small cell lung carcinoma (sclc) which constitutes about % of all lung cancer, is a very aggressive tumor with a tendency for early metastasis [ ]. targeted therapy agents are widely used in nsclc, but no one is approved for use in sclc; in fact all the clinical trials investigating biological agents in sclc, yielded negative results. imatinib did not demonstrate clinical efficacy even in patients with c-kit expression [ ]. vandetanib failed to demonstrate efficacy as a main- tenance therapy and cediranib failed to demonstrate ob- jective responses in recurrent or refractory sclc [ ]. despite its peculiar chemo and radio-sensitivity, long term survival is achieved in a small minority of patients and combination of cisplatin and etoposide (pe) still represent the standard of care after years; with this figure time to progression in the limited and extended disease. boni et al. journal of translational medicine ( ) : page of treatment objective response is achieved in - % of patients but median survival of – months in ed [ ]. two randomized phase trials evaluated doxorubicin- based chemotherapy compared with platinum-based chemotherapy in sclc. baka et al. enrolled patients with either ld or ed- sclc who were randomized to doxo- rubicin, cyclophosphamide and etoposide (ace) or pe for table grade and toxicities (%) leukopenia granulocytopenia anemia thrombocytopenia febrile neutropenia asthenia mucositis vomiting/nausea diarrhea alopecia neurotoxicity anorexia venous thrombosis pulmonary embolism heart failure six cycles. there were no differences in response rates or in median survival time for ld ( . vs . months) and ed ( . vs . months) [ ]. a cochrane systemic review of platinum vs non- platinum chemotherapy regimens evaluated trials with patients and concluded that platinum-based chemo- therapy regimens did not offer a statistically significant benefit in either overall response or survival compared to non-platinum chemotherapy. subsequentely, a japanase study, in , reported that irinotecan and cisplatin combination resulted in a sub- stantial improvement in survival compared with treatment with pe (median survival . vs . months) in ed [ ]. the authors concludes that the combination of irino- teca and cisplatin was an attractive option for patients with metastatic small-cell lung cancer who have a good performance status. numerous other drugs have been studied in combin- ation with platinum, without substantial improvemet in median survival or response rate. combination chemotherapy of belotecan (is a new camptothecin-derivative antitumor agent that belongs to the topoisomerase inhibitors) and cisplatin showed promis- ing efficacy comparable to currently available standard regi- men with favorable non-hematologic toxicity profile [ ]. however, its high hematologic toxicity profile seems to preclude general acceptance of this regimen as option for treatment of sclc. we are wainting for the result of boni et al. journal of translational medicine ( ) : page of the phase iii combat study which is comparing the ef- ficacy of belotcan plus cisplatin to etoposide plus cis- platin in patients with ed sclc. recent data with newer agents as thalidomide and bevacizumab, allow some cautious optimism for future advances in the treatment of this complex and challen- ging disease. strategies that have evaluated an increase of total dose, of dose intensity, number of courses, or number of drugs, or alternation of non-cross-resistant drugs, have been un- successful. these approaches are not recommended out- side clinical trials [ ]. some evidence suggests that adding thoracic radiother- apy to chemotherapy improves survival in patients with extensive-stage sclc who have a complete response out- side the thorax and at least a partial response within the thorax after three cycles of etoposide and cisplatin [ ]. immediate whole-brain radiotherapy is indicated in pa- tients with brain metastases and intracranial hypertension or other neurological emergencies. in some series of pa- tients with sclc or nsclc and brain metastases, whole brain radiotherapy combined with different chemotherapy regimens seemed to increase the risk of neurological toxic effects, but also to increase response rates and lengthen the time to progression of brain metastasis [ , ]. this increase in toxic effects was probably related to the use of anthracyclines and high doses of radiation per fraction. on the basis of this evidence whole-brain radiotherapy should be started after the completion of chemotherapy in patients with brain metastases, with or without symptoms, but not delivered concomitantly with cytotoxic treatment. loehrer pj et al., compared pei regimen (cisplatin- etoposide-ifosfamide) with pe regimen in previously un- treated ed-sclc. the results indicated that pei regimen was associated with a significant improvement in os and pfs, without significant difference in toxicity [ – ]. the hoosier oncology group study showed an im- proved time to progression (statistically different) and over- all survival associated with pei combination chemotherapy ( . month versus . with pe regimen), in patients randomized from and ; in the previous study, reported in patients from to , pei regimen produced a high complete remission rate in patients with extensive disease. ifosfamide combination chemotherapy demonstrated a durable complete remissions also in heavily pretreated patients with recurrent germ cell tumors [ ]. in our study, pei regimen gave interesting results both in ld and ed sclc, with a remarkable median overall sur- vival of . months in ld, and of . months in ed. in both stages, the os and ttp appear higher than expected, comparing to the results obtained with pe regimen published in the literature. conclusions sclc remains a disease with disappointing results and no good news coming from the research. in extended- stage disease new drug combinations and approaches did not result in any improvement in overall survival, that remains the ultimate goal as, unlike in other che- mosensitive cancers, second-line treatment is not an op- tion for most patients. many other strategies, including maintenance therapy, dose-intense, dose-dense chemotherapy and alternating regimens have failed to demonstrate consistent benefits, often with unacceptable toxicity. in the absence of a firm evidence of any activity of bio- logical agents, pei regimen, in our opinion, may be a possible therapeutic option, with high activity and an ac- ceptable toxicity profile. competing interests the authors declare that they have no competing interests. authors’ contributions bc, pm, bl and zf conceived and designed the study, collected the clinical data and wrote the manuscript. gr, bl and sl participated in the design of the study and performed the statistical analysis. all authors read and approved the final manuscript. acknowledgement all authors contributed towards the study by making substantial contributions to conception, design, acquisition of data, or analysis and interpretation of data, drafting the manuscript or revising it critically. author details department of oncology and advanced technologies, oncology unit, azienda ospedaliera s.maria nuova/irccs of reggio emilia, viale risorgimento, , reggio emilia, (italy). department infrastructure research and statistics, azienda ospedaliera s.maria nuova/irccs of reggio emilia, reggio emilia, (italy). received: january accepted: april references . american cancer society. cancer facts and figures . . . govindan r, page n, morgensztern d, read w, tierney r, vlahiotis a, et al. changing epidemiology of small-cell lung cancer in the united states over the last years : analysis of the surveillance, epidemiologic, and end results database. j clin oncol. ; : – . . hochstenbag mm, twijnstra a, wilmink jt, woutes ef, velde gp. asymptomatic brain metastases (bm) in small lung cancer (sclc): mr-imaging is useful at initial diagnosis. j neurooncol. ; : – . . sorensen m, pijls-johannesma m, felip e, for the esmo guidelines working group. small- cell lung cancer: esmo clinical practice guidelines for diagnosis, treatment and follow-up-up. ann oncol. ; suppl :v – . . maddison p, newsom-davis j, mills kr, souhami rl. favourable prognosis in lambert-eaton myasthenic syndrome and small-cell lung carcinoma. lancet. ; : – . . maddison p, lang b. paraneoplastic neurological autoimmunity and serviva in small-cell lung cancer. j neuroimmunol. ; : – . . hanna nh., einhorn lh. small cell lung cancer: treatment for extensive stage disease. malignant tumors of the lung. ; - . . kurup a, hanna nh. treatment of small cell lung cancer. crit rev in oncol/ hematol. ; : – . . lad t, piantadosi s, thomas p, payne d, ruckdeschel j, giaccone g. a prospective randomized trial to determine the benefit of surgical resection of residual disease following response of small cell lung cancer to combination chemotherapy. chest. ; : s– . boni et al. journal of translational medicine ( ) : page of . loehrer sr pj, rynard s, ansari r, songer j, pennington k, einhorn l. etoposide, ifosfamide, and cisplatin in extensive small cell lung cancer. cancer. ; : – . . rosti g, bevilacqua g, bidoli p, portalone l, santo a. e genestreti g. small cell lung cancer. ann oncol. ; (suppl ):ii – . . loehrer sr pj, ansari r, gonin r, monaco f, fisher w, sandler a, et al. cisplatin plus etoposide with and without ifosfamide in extensive small cell lung cancer: a hoosier oncology group study. j clin oncol. ; : – . . johnson b, fisher b, fisher t, dunlop d, rischin d, silberman s, et al. phase ii study of sti (gleevectm) for patients with small cell lung cancer. proc am soc clin oncol. ; : a. abstract . . dabir s, babakoohi s, kluge a, morrow jj, kresak a, yang m, et al. ret mutation and expression in small-cell lung cancer. j thorac oncol. ; ( ): – . . baka s, califano r, ferraldeschi r, aschroft l, thatcher n, taylor p, et al. phase iii randomized trial of doxorubicin-based chemotherapy compared with platinum-based chemotherapy in small-cell lung cancer. br j cancer. ; : – . . noda k, nishiwaki y, kawahara m, negoro s, sugiura t, yokoyama a, et al. irinotecan plus cisplatin compared qith etoposide plus cisplatin for extensive small-cell lungcancer. n engl j med. ; : – . . lim s, cho bc, jung jy, kim gm, kim sh, kim hr, et al. phase ii study of camtobell inj. (belotecan) in combination with cisplatin in patients with previously untreated, extensive stage small cell lung cancer. lung cancer. ; ( ): – . . demedts i, vermaelen k, van meerbeeck jp. treatment of es-sclc: current status and future prospects. eur respir j. ; : – . . jeremic b, shibamoto y, nikolic n, milicic b, milisavljevic s, dagovic a, et al. role of radiation therapy in the combined-modality treatment of patients with extensive disease small-cell lung cancer: a randomized study. j clin oncol. ; : – . . postmus pe, haaxma-reiche h, smit ef, groen hj, karnicka h, lewinski t, et al. treatment of brain metastases of small-cell lung cancer: comparing teniposide and teniposide with whole-brain radiotherapy- a phase iii study of the european organization for the research and treatment of cancer lung cancer cooperative group. j clin oncol. ; : – . . neuhaus t, ko y, muller rp, grabenbauer gg, hedde jp, schueller h, et al. a phase iii trial of topotecan and whole brain radiation therapy for patients with cns-metastases due to lung cancer. br j cancer. ; : – . . loehrer pj, lauer r, roth bj, williams sd, kalasinski la, einhorn lh. salvage therapy vp or vinblastine plus ifosfamide plus cisplatin in recurrent germ cell tumors. ann intern med. ; : – . submit your next manuscript to biomed central and take full advantage of: • convenient online submission • thorough peer review • no space constraints or color figure charges • immediate publication on acceptance • inclusion in pubmed, cas, scopus and google scholar • research which is freely available for redistribution submit your manuscript at www.biomedcentral.com/submit 	abstract 	objectives 	materials and methods 	results 	conclusions 	trial registration 	institutional review board that approved the study 	introduction 	material and methods 	treatment regimen 	response and toxicity evaluation 	statistical analysis 	results 	response and survival 	toxicity 	discussion 	conclusions 	competing interests 	authors’ contributions 	acknowledgement 	author details 	references pemetrexed in second line and beyond small cell lung cancer: a hoosier oncology group phase ii study original article pemetrexed in second line and beyond small cell lung cancer a hoosier oncology group phase ii study shadia jalal, md,* rafat ansari, md,† ramaswamy govindan, md,‡ sumeet bhatia, md,§ daniel bruetman, md,� william fisher, md,¶ gregory masters, md,# angela white, mba,** daniel stover, mt (ascp),** menggang yu, phd,†† and nasser hanna, md* introduction: small cell lung cancer (sclc) is initially a chemo- therapy-sensitive disease. nevertheless, drug-resistance results in disease recurrence in most patients. many drugs, including antime- tabolites, are active, but only minimal progress has been made in improving survival times for those with advanced disease. based on the need to discover better systemic therapies, we conducted a phase ii study of pemetrexed in patients with relapsed sclc. patients and methods: eligible patients had sclc or poorly differentiated neuroendocrine cancers of the lung, eastern cooper- ative oncology group performance status of to , and had received less than or equal to two prior chemotherapy regimens (additional targeted agents were allowed). both chemotherapy-sensitive (re- lapse � days from completion of first line therapy) and chemo- therapy-resistant (progressive disease during or within days from completion of first line treatment) patients were eligible and ana- lyzed separately. pemetrexed was administered at mg/m intra- venously every days for up to six cycles. all patients received folic acid, vitamin b , and steroid prophylaxis. the primary objective of the trial was to estimate the clinical benefit rate (complete plus partial response plus stable disease) in each group. results: from january to september , patients were enrolled in the chemotherapy-sensitive arm and patients in the chemotherapy-resistant arm. the majority of patients were men, the median age of the two groups were . and , respectively; % had a performance status of or , and more than % had received more than one prior regimen. grade / toxicities were as expected for pemetrexed. progressive disease was the best response in patients ( %) in the chemo-sensitive group and patients ( %) in the chemo-refractory group. one patient had a partial response and three had stable disease in each group. conclusion: pemetrexed has minimal single agent activity in re- lapsed sclc. key words: small cell lung cancer, pemetrexed. (j thorac oncol. ; : – ) small cell lung cancer (sclc) accounts for approximately to % of lung cancers. the majority of patients present with extensive stage (es) with evidence of wide spread metastases. despite an initial response to chemother- apy, patients with sclc inevitably relapse and progress resulting in death in nearly all patients. over the past decade, only minimal improvements have been made in the treatment of es sclc. many agents tested over the last three decades have demonstrated some activity, but only topotecan is ap- proved by the u.s. food and drug administration to treat sclc in the second line setting. unfortunately, therapy with topotecan results in a median survival time of only months, and most patients do not respond even for short periods of time. more effective therapies are desperately needed. some antifolates have demonstrated activity against sclc. for example, treatment with methotrexate resulted in responses in to % of patients in one study and was frequently integrated into multidrug regimens against sclc. in contrast, -fluorouracil had no activity against sclc in two studies. , pemetrexed (alimta, eli lilly and company, indianap- olis, in) is a multitargeted antifolate agent, the primary mechanism of which is inhibition of thymidylate synthase. pemetrexed also inhibits other folate-dependent enzymes in- volved in purine synthesis, including dihydrofolate reductase and glycinamide ribonucleotide formyl transferase and is u.s. food and drug administration approved to treat patients with advanced non-sclc and mesothelioma. an in vitro growth inhibition study by chan et al. showed pemetrexed *indiana university melvin and bren simon cancer center, indianapolis, indianapolis; †northern indiana cancer research consortium, south bend, indianapolis; ‡washington university medical center, siteman cancer center, st. louis, missouri; §community regional cancer cen- ter, indianapolis, indianapolis; �center for cancer care at goshen health system, goshen, indianapolis; ¶medical consultants, p.c. c/o ball me- morial hospital cancer center, muncie, indianapolis; #christiana care health services, inc., newark, delaware; **hoosier oncology group, indianapolis, indianapolis; and ††division of biostatistics, indiana uni- versity school of medicine, indianapolis, indianapolis. disclosure: ramaswamy govindan has received honoraria from eli lilly, genentech, astra zeneca and research funding from eli lilly, genen- tech, merck. nasser hanna is a consultant in eli lilly and has received honoraria and research support from eli lilly. address for correspondence: shadia jalal, md, barnhill drive, india- napolis, in . e-mail: sjalal@iupui.edu copyright © by the international association for the study of lung cancer issn: - / / - journal of thoracic oncology • volume , number , january delayed growth of sclc cell lines. a randomized phase ii trial by socinski et al. evaluating cisplatin/pemetrexed or carboplatin/pemetrexed combinations in patients with es sclc suggested these were active combinations; however, the single agent activity of pemetrexed had not been defined. we, therefore, conducted a study of single agent pemetrexed in patients with relapsed sclc to characterize its activity. patients and methods patient selection/eligibility eligible patients had histologic or cytologic proof of small cell cancers of any site or poorly differentiated neu- roendocrine cancers of the lung. all patients had measurable disease per the response evaluation criteria in solid tumors criteria and had adequate bone marrow function as defined by absolute neutrophil count more than or equal to /mm , platelet count more than or equal to , /mm and hemo- globin more than or equal to g/dl, renal function (creatinine clearance � ml/min), and liver function (bilirubin � . � upper limit of normal and aspartate aminotransferase � . � upper limit of normal). patients had an eastern cooperative oncology group performance status of to and must have received treatment with at least one but not more than two prior chemotherapy regimens (including one regimen con- taining a platinum agent). patients who received radiation must have completed their radiation at least days before being registered for the protocol. exclusion criteria included symptomatic central ner- vous system metastasis, uncontrolled pleural effusions, preg- nancy or lactation, inability or unwillingness to interrupt aspirin or other nonsteroidal anti-inflammatory agents, un- willingness to take folic acid or vitamin b supplementa- tion, or medical problems of significant severity (e.g., unsta- ble respiratory or cardiac disease). patients previously treated with pemetrexed were not eligible. the study was conducted by the hoosier oncology group, a community-based coop- erative group. the protocol was approved through institu- tional ethics review boards, and all patients provided written informed consent before treatment. pretreatment studies included history and physical ex- amination, complete blood count with differential, blood chemistries, and a computed tomography scan of chest or chest radiograph. on study evaluations (every days) in- cluded a limited history and physical examination, complete blood count and blood chemistries, performance status eval- uation, and toxicity rating. disease assessment by chest radiograph or computed tomography scan was performed before each cycle of therapy to ensure patients with other treatment options were not rapidly progressing. treatment schedule all patients received pemetrexed mg/m intrave- nously over approximately minutes. pemetrexed was re- peated every weeks for up to six cycles or until patients had either progressive disease or intolerable side effects. patients received folic acid ( – �g oral daily) approximately week before the first dose of pemetrexed and continued that daily during treatment. vitamin b was also given ( �g intramuscular injection) beginning week before the first dose of pemetrexed and was repeated every weeks during treatment. dexamethasone ( mg oral twice daily) was given for days every cycle beginning the day before pemetrexed. dose modifications dose adjustments at the beginning of a subsequent cycle were based on neutrophil and platelet nadir values from the preceding cycle of therapy. an absolute neutrophil count more than or equal to . � /l and platelet count more than or equal to � /l were required before the start of a new cycle. if a patient developed neutropenic fever, or thrombocytopenia with platelet count less than , /m with any bleeding, treatment was resumed at % of previous dose of pemetrexed. a % dose reduction was advised for grade – mucositis. treatment was resumed at % of the previous dose for other grade – nonhematologic toxicities (with the exception of alopecia, nausea, vomiting, or grade transaminase elevations). statistical considerations the trial used a two-stage design and evaluated che- motherapy-sensitive and chemotherapy-resistant patients sep- arately. for chemotherapy-sensitive patients (those who re- lapsed � days from completion of first line therapy), a total of patients were to be accrued to the first stage. if more than or equal to eight patients exhibit disease control (com- plete response (cr), partial response (pr), stable disease (sd)), the study was to be continued to the second stage and an additional patients were enrolled. if more than or equal to patients of the total patients on this arm have nonprogressive disease, the regimen was to be considered worthy of further study in this population. for chemotherapy- resistant patients (those who relapsed within days from completion of first line therapy), a total of patients were to be accrued to the first stage. further recruitment was to continue to the second stage if more than or equal to three patients had a response, whereby additional patients were then to be enrolled. if more than or equal to eight patients of had disease control, the regimen was to be considered worthy of further study. the sample size was calculated using a one-sided alpha of % and a power of %. all toxicities were graded according to the common terminology criteria for adverse events version . . study endpoints all endpoints were collected and analyzed for the two patient populations (chemo-sensitive and chemo-resistant) separately. the primary objective of the study was to estimate the clinical benefit rate (cr, pr, sd) of pemetrexed in the two patient populations. the secondary objectives were to determine the toxicity, time to disease progression, and to estimate overall survival. response was assessed using the response evaluation criteria in solid tumors criteria. time to disease progression was defined as time from start of treatment until the criteria for disease progression was met. overall survival was defined as the number of days from the day of first treatment to death or the last date of patient contact (whichever is earlier). jalal et al. journal of thoracic oncology • volume , number , january copyright © by the international association for the study of lung cancer results patient characteristics from january to september , chemo- sensitive patients and chemo-resistant patients were ac- crued to the study from sites within the hoosier oncology group. two patients had extrapulmonary small cell carci- noma, with the remaining patients having sclc. patient characteristics are displayed in table . criteria to proceed to stage ii was not met for either arm. the median number of cycles of pemetrexed was (range, – ). no dose delays or modifications were required. toxicity grade / toxicities are summarized in table . grade / hematologic toxicities were minimal, and grade / non- hematologic toxicities were consistent with previous data for single agent pemetrexed. response and survival progressive disease as the best response was noted in patients in the chemo-sensitive group ( %) and patients ( %) in the chemo-refractory group. in the chemo- sensitive group, one patient ( %) achieved a partial response and three patients ( %) achieved stable disease. in the chemo-resistant group, one patient ( %) had a partial re- sponse and three patients ( %) had stable disease. at the time of final analysis (march, ), % of patients on the trial were alive (two in each group). median time to disease progression in the chemo-sensitive and chemo-resistant arms was . months ( % confidence interval �ci�: . – . ) and . months ( % ci: . – . ), respectively, and median survival time was . months ( % ci: . – . ) and . months ( % ci: . – . ), respectively. discussion our study evaluated single agent pemetrexed in re- lapsed sclc and did not meet the predetermined criteria for increasing the sample size to patients. response rates were disappointing and the majority of patients had early progres- sive disease. single agent pemetrexed has minimal activity in patients with previously treated sclc, including those with previously chemotherapy-sensitive disease. other investigators have reported their experience with pemetrexed in patients with previously treated sclc. grøn- berg et al. treated patients with pemetrexed at mg/m and also reported minimal activity in both chemotherapy- sensitive and chemotherapy-resistant patients ( % stable disease and % progressive disease in sensitive arm, % partial response and % progressive disease in resistant arm). raju et al. enrolled patients ( chemotherapy- sensitive and chemotherapy-resistant) with relapsed sclc who had received only one prior chemotherapy regimen. patients were treated with pemetrexed at doses ranging from to mg/m . preliminary results indicated a disappoint- ing clinical benefit rate (cr, pr, sd) of % in the chemo- therapy-sensitive arm and . % in the chemo-refractory arm. there were no responses with mg/m and only one partial response was seen with mg/m . in addition, a phase iii trial comparing platinum with etoposide versus pemetrexed was closed early when an initial analysis reported that the pemetrexed arm was unable to meet the predefined end point of noninferiority. interim results of this study (known as gales) were presented at the meeting of the american society of clinical oncology. in this study, chemonaive patients with ed sclc were random- ized to either the combination of pemetrexed ( mg/m table . baseline characteristics characteristics no. of patients (%) chemo-sensitive (n � ) chemo-resistant (n � ) age (yrs) median . range – – sex male ( %) ( %) female ( %) ( %) ecog ps ( %) ( %) ( %) ( %) ( %) ( %) prior treatments ( %) one chemotherapy ( %) ( %) two chemotherapy ( %) ( %) � ( rd line targeted therapy) ( %) smoking history current ( %) ( %) former ( %) ( %) never ( %) median time since prior regimen (range) d ( – ) d ( – ) ecog, eastern cooperative oncology group; ps, performance status. table . key grade / toxicities toxicity chemo-sensitive chemo-resistant grade grade grade grade hematologic neutropenia ( %) ( %) ( . %) ( . %) thrombocytopenia ( %) ( . %) ( . %) anemia ( %) ( . %) febrile neutropenia ( %) fatigue ( %) ( . %) hyperbilirubinemia ( %) ( . %) ast elevation ( %) nausea ( . %) constipation ( . %) diarrhea ( %) ( . %) mucositis ( . %) rash ( %) infection ( %) ( . %) ast, aspartate aminotransferase. journal of thoracic oncology • volume , number , january pemetrexed beyond small cell lung cancer copyright © by the international association for the study of lung cancer intravenously day ) and carboplatin (area under the curve intravenously day ) or etoposide ( mg/m days , , ) and carboplatin (area under the curve intravenously day ). the interim analysis included patients on the pemetrexed arm and on the etoposide arm. progression free survival was noted to be inferior for patients on the pemetrexed arm compared with those on standard therapy ( . versus . months with hazard ratio � . ). overall response rates were also inferior for the pemetrexed arm ( . % versus . % with p value � . ). nearly all chemotherapy agents active against non- sclc are also active in sclc. pemetrexed seems to be an exception to this observation. recent evidence suggests that tumors associated with a high expression of thymidylate synthase may be resistant to the effects of pemetrexed. it seems that squamous cell lung cancers (with high levels of thymidylate synthase activity), for example, are resistant to pemetrexed, whereas nonsquamous lung cancers are more sensitive. , ceppi et al. have recently reported that thymi- dylate synthase expression in sclc is high. this may explain why pemetrexed has minimal activity against sclc. there have been few advances in the systemic treat- ment of sclc over the last years. this is due, in part, to the lack of a comprehensive understanding of key signaling pathways in sclc. there is clearly a chemotherapy-sensitive population of cells, which can be effectively treated with many chemotherapeutics; however, there is a chemotherapy- resistant clone, which resists most treatment and ultimately results in the patient’s death. this level of complexity of sclc is underscored by data published by davies et al., who evaluated somatic mutations of the protein kinase gene family in primary lung neoplasms (seven adenocarcino- mas, seven squamous cell carcinomas, six large cell carcino- mas and six carcinoids) and seven lung cancer cell lines (which included one neuroendocrine cancer). they reported somatic mutations in genes (higher incidence than observed in other cancers). most of the mutations were thought to be passenger ones that are not believed to play a significant role in the cancer phenotype. nevertheless, more than nonsynonymous substitutions were detected, which might possibly be implicated in oncogenesis (driver muta- tions). some heterogeneity was noted in the mutational spec- trum among the different lung cancer cells screened in that study as well. for example, the largest number of somatic mutations was noted in nci-h cell line, which was derived from a lung neuroendocrine tumor. as illustrated by our study and those involving many other therapies (chemotherapy and targeted agents), the con- tinued exploration of empiric therapies is unlikely to result in major advances for patients suffering from sclc, and a better understanding of the biology of the disease is greatly needed. acknowledgments eli lilly funded the study. references . govindan r, page n, morgensztern d, et al. changing epidemiology of small cell lung cancer in the united states over the past three decades: analysis of surveillance, epidemiologic and end results database. j clin oncol ; : – . . chute j, chen t, feigal e, simon r, johnson be. twenty years of phase iii trials for patients with extensive-stage small-cell lung cancer: per- ceptible progress. j clin oncol ; : – . . o’brien m, ciuleanu t, tsekov h, et al. survival benefit of oral topotecan plus supportive care versus supportive care alone in relapsed, resistant sclc. lung cancer ; (suppl ):s . . morrison va, luikart sd. chemotherapy of lung cancer. in mc perry, ed. the chemotherapy source book. baltimore, md: williams and wilkins, . pp. – . . stewart dj, dahrouge s, soltys km, evans wk. a phase ii study of -fluorouracil plus high-dose folinic acid in the treatment of recurrent small cell lung cancer. am j clin oncol ; : – . . havsteen h, sörenson s, rørth m, dombernowsky p, hansen hh. -fluorouracil in the treatment of small cell anaplastic carcinoma of the lung: a phase ii trial. cancer treat rep ; : – . . adjei aa. pemetrexed (alimta): a novel mutitargeted antifolate agent. exp rev anticancer ther ; : – . . chan dc, chen vj, zhang z, et al. studies of pemetrexed and gemcit- abine, alone and in combinations, in human lung cancer models. j clin oncol, asco annual meeting proceedings; vol , no. s (june suppl), : . . socinski ma, weissman w, hart ll, et al. randomized phase ii trial of pemetrexed combined with either cisplatin or carboplatin in untreated extensive-stage small-cell lung cancer. j clin oncol ; : – . . grønberg bh, bremnes rm, aasebø u, et al. a prospective phase ii study: high-dose pemetrexed as second-line chemotherapy in small-cell lung cancer. lung cancer. ; doi: . /j.lungcan. . . . . raju rn, neubauer ma, smith da, et al. pemetrexed (p) in relapsed small cell lung cancer (sclc): preliminary results of a phase ii trial. j clin oncol, asco annual meeting proceedings; vol , no. s (june suppl), : . . socinski ma, smit ef, lorigan p, et al. phase iii study of pemetrexed plus carboplatin (pc) versus etoposide plus carboplatin (ec) in che- monaive patients (pts) with extensive-stage disease small cell lung cancer (ed-sclc): interim results. asco meeting. . eismann u, oberschmidt o, ehnert m, et al. thymidylate synthase gene expression in solid tumors predicts for response to pemetrexed in vitro. j clin oncol, asco annual meeting proceedings; vol , no. s (june suppl), : . . scagliotti g, park k, patil s, et al. favorable benefit to risk profile for pemetrexed plus cisplatin versus gemcitabine plus cisplatin in a large phase iii study of first-line therapy in advanced non-small cell lung cancer. eur j cancer ;(suppl ): . . peterson p, park k, fossella f, et al. is pemetrexed more effective in patients with non-squamous histology? a retrospective analysis of a phase iii trial of pemetrexed versus docetaxel in previously treated patients with advanced non-small cell lung cancer (nsclc). eur j cancer ;(suppl ): – . . ceppi p, volante m, ferrero a, et al. thymidylate synthase expression in gastroenteropancreatic and pulmonary neuroendocrine tumors. clin cancer res ; : – . . davies h, hunter c, smith r, et al. somatic mutations of the protein kinase gene family in human lung cancer. cancer res ; : – . jalal et al. journal of thoracic oncology • volume , number , january copyright © by the international association for the study of lung cancer 	pemetrexed in second line and beyond small cell lung cancer: a hoosier oncology group phase ii study 	patients and methods 	patient selection/eligibility 	treatment schedule 	dose modifications 	statistical considerations 	study endpoints 	results 	patient characteristics 	toxicity 	response and survival 	discussion 	acknowledgments 	references wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ main body- ( ) paul r. mullins—marketing in a multicultural neighborhood paul r. mullins marketing in a multicultural neighborhood: an archaeology of corner stores in the urban midwest abstract corner stores were the most common consumer outlet in the urban midwest from the mid- th century until world war ii. in the early- th century in indianapolis, indiana, the near westside was dotted with more than modest stores managed by african americans, european immigrants, and white hoosiers. archaeological excavation of a ca. – store illuminates the widespread entrepreneurial ambition of urban newcomers, the increasing ethnic insularity of corner stores between and , and the eventual decline of corner stores in the face of urban renewal and the arrival of chain stores in the s. “those who strive”: ambition, entrepreneurship, and the color line on december , the walker theatre center in indianapolis, indiana, was unveiled. general manager freeman ransom announced that the theater complex was dedicated to “those who toil; to those who think; to those who strive readily; ... to all classes; to all races” (gibson- hudson : ). ransom’s celebration of ambi- tion and accomplishment reflected the american dream pursued by many residents of the city’s near westside, including c. j. walker herself. walker arrived in indianapolis in and launched a business selling her hair care remedies and beauty supplies. walker became enormously successful by training women to become walker “agents” who would teach women walker beauty techniques and sell the firm’s goods. walker first envisioned a black theater and commercial center for indianapolis after she was denied admission to a local theater. after her death in , her daughter a’lelia completed the project (bundles : ). the walker theatre complex on indiana avenue featured a stylistically stunning theater rendered in a unique union of afrocentric and art deco styles. today, the walker theatre is best known for this architecture. despite its unique architectural aesthetics, the building’s primary significance was as a center for black commerce. businesses, shops, and the walker manufacturing company occupied the center at the most prominent intersection in black indianapolis. beyond simply housing businesses, the walker was the city’s primary black social space. its grand ballroom hosted innumerable social events; the coffee pot was a well-known courting spot and late-night meeting place; and the offices throughout the center housed the city’s most prominent black professionals. as in many other american cities, consumer venues and districts like indiana avenue were among the community’s most significant social spaces, serv- ing narrowly defined economic functions while they also provided a profoundly consequential shared sociocultural experience. the walker loomed over a mile-long stretch of african american businesses and clubs that was the heart of black indianapolis from the late- th century into the s. entrepreneurs who shared many of madam walker’s own aspirations ran this concentration of consumer venues. like many of the europeans who migrated to indianapolis and lived in the same neighborhoods at the turn of the century, many african americans were quite committed to american dream ideology and saw entrepre- neurship and consumer space as their foothold into american society. many of these arrivals to indianapolis’s near westside optimistically lived out citizen aspirations and material ambitions much like walker’s, yet racism and xenopho- bia profoundly complicated their aspirations. in indianapolis, african american arrivals and most europeans found that provincial indiana whites erected complex boundaries to the newcomers’ rather commonplace material ambitions. with a partisan municipal government dominated by anti-black party politics and most public space governed by de facto racial and ethnic segrega- tion, many near westside residents found their foothold into american society as shopkeepers and consumers. the distinctive neighborhood marketing networks that emerged throughout historical archaeology, , ( ): – . permission to reprint required. paul r. mullins—marketing in a multicultural neighborhood the near westside mirrored and profoundly influ- enced dynamic class and color lines, forming one of the most important influences on com- munity identities. it might seem crass to place entrepreneurship and shopping at the heart of community identity, but it would be impossible to understand life in indianapolis without devot- ing considerable attention to the merchants and consumers who frequented the near westside’s hundreds of stores and clubs. in summer , indiana university-purdue university, indianapolis conducted excavations at an – corner store at camp street, a block away from the walker theater. the store was typical of those in the near westside and much like those operated by budding entrepre- neurs throughout the country. a small enterprise in a roughly sq. ft. building, the store sold everything from coal to candy and was managed by a long line of marketers that included whites and then african americans. most research on turn-of-the-century marketing has focused on department stores, mail-order sales, and adver- tising. such shifts in sales clearly impacted how consumer desire was marketed and experienced. yet, most everyday consumption in the urban midwest was conducted in much less grand, local venues like corner groceries. these insti- tutions and storekeepers assumed prominent social roles in many marginalized communities. in indianapolis, for instance, african americans and german immigrants alike cherished entrepre- neurial spirit, dotted the community with both ambitious and modest enterprises, and granted marketers considerable social prominence. the camp street assemblage and the history of neighboring marketers underscore the similar ambitions and material worlds of many white hoosiers (the local term for indiana residents), european immigrants, and african americans, despite the persistent power of racist ideology. for many marginalized citizens, such as african americans and european immigrants, entrepre- neurialism was a mechanism through which new citizens could secure social and material empowerment in a society that valued the self- made entrepreneur, individual ambition, and materialism. as madam walker herself found out, even astounding entrepreneurial success and wealth could not secure some rights across color or gender lines. african americans often were compelled to contest the implicit racialization of consumer space. in the s, for instance, walker’s former lawyer, freeman ransom, led a committee that identified more than gro- ceries and drugstores in predominately black neighborhoods that had no black employees (thornbrough : ). churches and women’s clubs led pickets against the most egregious offenders and urged african americans to take their business elsewhere. nevertheless, many americans remained invested in the ideological notion that marketing and entrepreneurialism could secure material and social stability, if not genuine affluence. these merchants and their customers in the near westside of indianapolis lived in a community in which modest corner stores and consumer spaces lay at the heart of community identity and reflected the complex position of many marginalized americans seek- ing a foothold into american society. marketing and multiculturalism in indianapolis a few african americans began to settle in indianapolis in the second quarter of the th century, and they clustered along indiana avenue in the city’s near westside. indiana avenue was one of four diagonal thorough- fares that extended from the central circle of indianapolis’s “mile square” grid plan, but the avenue’s western reaches were not considered particularly desirable areas. the western stretch of indiana avenue was marshy lowland, not far from the white river and fall creek, and some areas were poorly drained and prone to flooding. after the central canal crossed indiana avenue and almost immediately became an unpleasant open sewer that never fulfilled its intended function as a transportation corridor. the canal became an informally recognized boundary, separating the city to the east and the near westside to the west. over the late- th and early- th centuries, the near westside became a complex ethnic mosaic that included pockets of virtually every european immigrant group that came to america. the first settlers in the near westside included many german and irish immigrants who quickly established businesses along indi- ana avenue. a handful of african americans lived throughout the same neighborhoods prior to the civil war. in samuel smothers’s historical archaeology ( ) paul r. mullins—marketing in a multicultural neighborhood grocery store became the first of many african american businesses along the avenue. outside indiana avenue, the near westside contained mainly residential neighborhoods, punctuated by corner stores, a few social spaces like churches and schools, and several industrial workplaces along the white river. spaces that might reasonably be called ethnic neighborhoods were slow to emerge in the th-century near westside. in a few single blocks, most residents were from the same ethnic group, such as afri- can americans who settled around black schools or irish immigrants who lived along the white river within walking distance of the kingan’s meatpacking plant. waves of migration around the time of the civil war, at the turn of the century, and in the early s delivered new neighbors from the south and various reaches of europe, but white hoosiers remained the most common group across the near westside until the s. most indianapolis neighborhoods did not assume clear racial and ethnic segregation until the s, when de facto racism, city-sanc- tioned codes, and the local rise of the ku klux klan fueled increasingly clear spatial and social distinctions among white, european, and black residents. in the late- th century, in contrast, near westside residents routinely were served by local marketers from other ethnic groups, a situation that became increasingly uncommon after the turn of the century. by indiana avenue had a string of stores and clubs that formed black indianap- olis’s central business and leisure district. the strategically placed area became a magnet for african americans migrating northward. for those african americans migrating overland from kentucky, indianapolis was the first major northern city along the north-south artery, and many stopped at indiana avenue to settle in the area. the residential neighborhoods that fanned off indiana avenue included the area now known as ransom place. first divided into lots in , ransom place’s six-block historic area is today the only surviving remnant of the near westside’s th-century neighborhoods. like most of the near westside, ransom place was predominately residential; however, the neighbor- hood did include a few corner stores, a couple of churches, and a scatter of small businesses along the busily trafficked west street, which was one of the main northbound roadways through the city. the ransom place neighbor- hood eventually became home for many mem- bers of the african american elite, including madam walker’s lawyer freeman ransom, for whom the neighborhood is named. between and , ransom place began to be segregated along black-white lines, and the remainder of indianapolis was starkly drawn into black and white neighborhoods in the subsequent decade. in , % of ransom place’s population was black, which was slightly higher than the city’s rate of about %. by , nearly % of indianapolis’s , residents were african americans, the highest percentage of african americans in any city north of the ohio river, and two-thirds of ransom place residents were black. in that figure had ballooned to % in ransom place, a testament to how thoroughly segregated the city became within a quarter cen- tury (brady ). at the heart of these emergent neighborhoods was a vast number of small groceries catering to the immediate vicinity. small corner stores were typical features of the urban landscape in most late- th- and early- th-century cities. from a functional perspective, such venues simply pro- vided essential goods in densely settled urban communities. stores also had profound social significance because of their roles in structuring neighborhood identities and providing important spaces for entrepreneurial ambition. much of the corner store explosion came at the very moment that many central features of a consumer culture were emerging. department stores, advertising, chain stores, and public leisure were simply the most visible signals of how late- th-cen- tury citizen rights became invested in material- ism. corner stores should not be considered in isolation from such transformations. many self-made marketers tried establishing grocery stores because of their modest scale and the widespread demand for their products. in , w.e.b. du bois’s ( : ) survey of african american businesses in philadelphia found few grocery stores owned by african americans, but du bois still concluded that the “ambition of the middle class of negroes lies in this direction ... . the number of negro groceries will undoubtedly grow considerably in the next decade.” in , of businesses in indi- anapolis’s near westside were groceries, includ- ing at least run by african americans. in historical archaeology ( ) paul r. mullins—marketing in a multicultural neighborhood , paul edwards ( : ) canvassed nine southern cities and identified grocery stores out of , total retail establishments. grocery stores were the most common african american establishment in each of the nine cities. several corner stores were established in the ransom place neighborhood, which was initially a predominately white hoosier area, and these white residents included the first storekeepers at camp street. mary and william smith ran the camp street store between and . like many marginalized marketers, the smiths apparently spent some of their entrepreneurial lives as street peddlers selling goods from a wagon (praetzellis and praetzellis ). an huckster’s license was recovered from deposits dating to the smith’s occupation, a period when the city apparently licensed street peddling (figure ). since william a. smith appeared in mid- s city directories as either a salesman or canvasser, this may well have been his license. many hucksters and corner store merchants appeared to share the optimism that every venture could provide the seeds of a marketing empire. corner store merchants often used street peddling and small stores as steppingstones to grander enterprises. in typical fashion, the smiths moved their store to the more profitable indiana avenue in . the smiths lived in the grocery store (roughly sq. ft.), unlike most of the subsequent merchants, so it is problematic to distinguish between store products and the smith household discards. the deposits dating to their occupation include high percentages of household goods, and at least some of these likely belonged to the smiths. most corner stores specialized in prepackaged goods in cans and bottles, and many corner stores carried at least a modest range of fresh foods. the smith assemblage includes a handful of decorative goods. flow- erpots, glass table vessels, and white-bodied ceramic settings would have given the home a modest hint of victorian style, but such bric- a-brac and decorative goods might also have been sold in the store or distributed to custom- ers. the most unusual portion of the smith assemblage was a concentration of ceramic tiles produced at the united states encaustic tile works just a few blocks away. the tile works operated from until and specialized in ornamental tile like the fragments litter- ing the camp street site (hester ). wil- liam smith appeared in the census as a tile maker, so he likely was employed at the nearby tile works and brought the tiles to the site. the most distinctive characteristic of this tile assemblage is that very few tiles match each other in decorative style. the tiles include few matching colors, molded motifs, or even sizes. in an assemblage representing at least tiles, only % matched the color or size of other tiles in the assemblage. it is unclear exactly how the tiles were being used; none are obvi- figure . apparently city-issued, this huckster’s license was to be worn by one of the many street peddlers in turn-of-the-century indianapolis. (photo by paul r. mullins.) historical archaeology ( ) paul r. mullins—marketing in a multicultural neighborhood ous kiln wasters. only of the tiles were complete, and almost none have any evidence of having been installed. managing corner stores was one of the few entrepreneurial possibilities for women, and the camp street store, like many in indianapolis, was run by women for much of its -year operation. during the two decades between and , the store generally appeared in city directories in either mary smith’s or subsequent owner mar- garet mcguffin’s name, and from to martha miller, a canadian-born black, managed the store. yet many stores probably run by women were listed in city directories under the husband’s name. in , for example, only of the near westside’s groceries appeared in the directory under a woman’s name; however, many wives and daughters who appeared as clerks and store employees likely managed significant ele- ments of the store. the assemblage dating to the store’s african american management after does not reveal distinctive patterns that might be readily interpreted as ethnic identity. there is no clear evidence that the type of goods in the camp street store and similar african american corner venues was significantly different from that of white and european neighbors. however, the quantities, prices, store locations, and building conditions were in many cases quite different. edwards ( : ) studied african american consumers in the urban south in and concluded that african american groceries were often understocked and “consisted, in most cases, largely of staple canned and other non-perishable products.” edwards ( : ) characterized over half of nashville’s african american grocery stores as being “either in a run-down and dilapidated condition, or … otherwise unattractive.” in , ira dea. reid ( : ) concluded, “negro areas are ones in which some of the most undesirable mer- chandising practices are conducted. the negro consumer finds his neighborhood stores cluttered with a large supply of inferior and cheap goods at prices that are high under any standard.” a national negro business league survey of african american businesses found that . % were located in “negro neighborhoods,” mean- ing that their trade was overwhelmingly limited to african americans (edwards : ). edwards and many critics of black entrepre- neurship sometimes rhetorically overstated the dire conditions confronting african american grocers. edwards, for instance, was a cham- pion of academic business training and sales experience, and many african american grocery owners had neither. nevertheless, many african american groceries clearly were modest opera- tions in marginal locations. the camp street assemblage was composed almost entirely of standardized mass-produced commodities, but it also included an african- motif pot-metal brooch (ca. ) that signaled cultural and ethnic difference. the afrocentric style brooch was typical of a flood of jew- elry modeled on dancer josephine baker, and harlem renaissance artists prior to the s had championed elements of such african style. wearing such motifs in public spaces was cer- tainly politicized, because nativism, xenophobia, and racism have had a persistent impression on local life. the most infamous expression of this nativism came in the mid- s, when the ku klux klan ruled the indianapolis city govern- ment and held powerful sway over state gov- ernment as well. leonard moore’s ( : ) analysis of klan membership in indiana found that klan ranks included many residents in the near westside. an astounding . % of the city’s dues-paying klansmen lived in the area, where better than one-third of adult males were black and significant numbers were europeans and/or catholics who were also excluded from klan membership (moore : – ). the klan’s $ . initiation fee alone was equal to one-quarter or more of most men’s weekly incomes, and ongoing membership fees made joining the hooded order an even more expen- sive proposition (moore : ). even after the klan’s downfall in the late- s, racists continued to have a strong hold on most city politics and public services, devising racially based real-estate codes and school regulations that still profoundly shape indianapolis settle- ment patterns. consequently, the near westside’s mix of white hoosiers, african americans, and europeans did not necessarily signal a particu- larly settled multicultural community. in this atmosphere, the camp street brooch served a quite meaningful notice of its wearer’s identity in relation to african american culture and anti- black racism. historical archaeology ( ) paul r. mullins—marketing in a multicultural neighborhood in maurice l. shaffer and louis w. butler became the first african american proprietors of the camp street store. within a year, both men had moved on to new ventures (one as a machinist, the other as a painter), and martha miller took over the store’s operation. miller lived a few doors away from the store, so most of the sparse material culture dating to her – residency is associated with the store itself. the yard deposits associated with miller’s tenure had a concentration of marbles and toys. in the early- th century, near westside neighborhoods were rapidly subdivided to accommodate the rush of african american migrants. since many near westside homes had very small backyards and were built nearly up to the street, most yards had very little open space even before subdivision began. the camp street grocery had a large yard opening onto the street, and it sat on a heavily trafficked neighborhood corner within sight of the city’s prominent african american businesses along indiana avenue. like the many black barbershops and salons in the near westside, corner stores served as both commercial and social spaces. in some corner stores the literal corner itself and the store’s outside spaces were critical to the marketers’ social and financial viability. such spaces were critical for a black community that was marginalized in or denied access to most of the city’s implicitly white consumer and public spaces. the dense deposits of marbles, toys, and coins suggest that the camp street store lot may have been a gathering place for neighborhood residents, including children. many antebellum african americans had been ambitious entrepreneurs, but the number of afri- can american marketers expanded exponentially in turn-of-the-century indianapolis. this develop- ment reflected widespread african american dis- appointment with partisan politics. reconstruc- tion promises had passed largely unfulfilled, and politicians had willingly undercut african ameri- can voting rights, civil privileges, and the most basic human rights. stung by this racist politi- cal system and an increasingly hostile white society, many african americans turned their energy toward building a relatively independent black economy and investing considerable hope in an emergent consumer culture. that hope in progressive-era consumer society was certainly not limited to african americans. between the late- th century and about , a panoply of observers agreed that material consumption had sociopolitical import (cohen : ). most of this national commentary focused on how consumption could impact community morals, labor conditions, and standards of living. out- side african america, few observers examined consumption’s relationship to the color line, but many african american communities quickly championed consumer activism, such as patron- age of african american merchants and boycotts against racist marketers. cooperative movements received significant exposure in the african american press, but few were viable enterprises. african americans developed quite successful marketing districts in many cities, with diversi- fied african american business communities in midwestern cities like indianapolis and chicago. as in chicago, race-insular businesses without white competitors were the near westside’s most successful and long-term ventures (cohen : ). black barbers, hairdressers, and undertak- ers, for instance, had no competition from white entrepreneurs, and many such african american ventures were quite prosperous and long-lived. the willis funeral home, for instance, was established in , and today sits alongside the walker theatre, where it remains in business. nevertheless, even prosperous african american entrepreneurs could not expect complete material and social separation from the dominant white consumer economy, and even in the most pros- perous black business districts, african ameri- cans were still compelled to buy some goods from merchants who were european or white. ultimately, the notion of a black economy was a somewhat conservative structural cri- tique, despite its profound critique of consumer culture’s anti-black dimensions. for the most part, a “black economy” simply transported the basic structure of a capitalist marketplace and attempted to purge it of antiblack racism. the most significant challenge to african american corner merchants came from chain stores, which first appeared in indianapolis in the early- th century and were most successful in the grocery business. chains, however, were slow to move into marginalized neighborhoods. most of the first chains in indianapolis were located in genteel, white neighborhoods and downtown, not in the near westside. indiana avenue would always remain dominated by historical archaeology ( ) paul r. mullins—marketing in a multicultural neighborhood single-store, african american marketers. nev- ertheless, by the mid- s, chains accounted for % of sales in indiana (u.s. dept. of commerce : ). in , grocery stores were located in indianapolis, including chains that included at least two stores. only one of the near westside’s groceries in was part of a chain, but nearby chain stores certainly were frequented by neighborhood resi- dents; chain prices typically ran about % less than corner groceries (longstreth : ). one of the attractions of corner stores had been a willingness to run an “open account” for neigh- bors, but % of grocery sales in indiana were cash and carry by . some of the chains had a reputation for settling in black neighborhoods but not hiring african americans. in the s an african american coalition championed a “don’t buy where you can’t work” campaign that chose local chains kroger and standard as their primary targets (thornbrough : – ). the chains did apparently hire a few african american employees when african americans threatened to shop elsewhere. postwar demographic shifts were critical fac- tors in the decline of corner stores in india- napolis. during the depression the near westside became dominated by increasingly dense rental housing and deteriorating conditions. after world war ii, many veterans led an exodus to newly built neighborhoods to the north, and this shift continued unabated into the s. clyde bolden ( : ) found that between and music clubs dominated indiana avenue, with its earlier business diversity decreasing significantly and forcing local consumers to travel to other areas in the city. bolden’s ( : ) count of indiana avenue businesses found that restaurants always were the most common venture along the avenue, followed closely by groceries and hairstylists. in , . % of the avenue’s businesses were groceries, and in that percentage was a comparable . % ( of ). when the avenue’s gro- ceries and neighborhood corner stores are con- sidered together, it is clear that groceries were the near westside’s most common business into the s, especially in the neighborhoods off indiana avenue. in the number of businesses on indiana avenue had shrunk to , of which only ( . %) were groceries, and the corner store markets in surrounding neighborhoods felt the impact of these same shifts as residents and businesses moved out of the community. by the s, the disintegration of near westside neigh- borhoods was reflected in the camp street store, which was being managed by jesse robinson. robinson was selling coal and ice as well as standard corner store goods. in about , he filled a cistern with an assemblage that attests to the store’s decline. after heaving a toilet into the six-foot deep feature, the storekeeper threw in at least bottles, reflecting that the store was primarily selling alcohol by that time— vessels had held liquor, another were wine bottles, and at least were beer bottles, which collectively accounts for nearly half the assem- blage. the next most common vessel type was soda bottles: bottles of nehi, fanta, sun crest, and similar cold war midwestern sodas dominate the assemblage. edwards ( : ) indicated in that black corner marketers sold large volumes of canned goods, and the density of tin cans in the camp street cistern indicates that the store was selling a consider- able volume of canned foods, alongside this narrow range of bottled goods and ice. there is no evidence in the cistern for many other goods. this modest range of goods was likely what many other corner stores sold as the neighbor- hood disintegrated (mullins : – ). the social network surrounding black con- sumer spaces in the near westside significantly influenced community identities, but it was not isolated from other institutional influences. for example, hoosiers have long placed church life at the center of social and individual identity, and black churches often lent entrepreneurship and materialism a dimension of faith. in , for instance, james weldon johnson ( : – ) noted that when property was inexpensive in early- th-century harlem, “buying property became a contagious fever. it became a part of the gospel preached in churches.” african american entrepreneurs often projected faith onto secular marketing interests and made little distinction between religion and economics, but the city’s churches were divided along complex class and social lines that did not champion any particularly coherent economic vision. the near westside’s churches in ranged from small storefront churches to several large congregations. these churches certainly had a historical archaeology ( ) paul r. mullins—marketing in a multicultural neighborhood profound impact on how african americans viewed the politics of marketing and consump- tion, but they were easily outnumbered by the groceries in the same neighborhood. in con- trast to sunday services, wealthy and poor afri- can americans mingled together on the avenue and in stores, providing a significant shared cultural life. nevertheless, that appearance of african american integration across class lines in consumer space may itself have been forced on african americans through segregation. one african american who returned from world war ii acknowledged that “indiana avenue was the area where it happened for us… . but it was eventually to the point where you started going because of segregation mainly” (ridley ). the most prominent single institutional influence in the near westside probably was the segre- gated school system, especially crispus attucks high school. established as the city’s sole black high school in , attucks was staffed by a host of well-educated african americans unable to secure work in white schools, and the school championed a fierce commitment to education as the mechanism to combat racism. the importance of the near westside’s modest stores and the social significance of the entre- preneurs running those venues are typically lost. analyses often focus on prominent social institutions, such as schools and churches, or overlook the significance of small stores as social spaces structuring neighborhood identity. yet in a mostly segregated urban america, mass culture in general and consumption in particular were prominent mechanisms that bound african americans together, much as they bound many european immigrants. placing entrepreneurship at the heart of th-century african american culture or elevating marketing and consumption to a prominent role in life across the color line does not mean that african american culture became synonymous with consumer culture. african american consumers still had distinc- tive social and material politics. it seems telling that many civil rights actions were eventually directed at consumer spaces such as restaurants and stores. that makes sense in indianapolis, where african americans had a long heritage of entrepreneurship and consumption that afforded them heightened awareness of citizen rights that went along with marketing and shopping. rather than simply reduce stores and marketers to transparent intermediaries in the production and consumption process, it makes more sense to accord social meaningfulness to shopping and entrepreneurship and to recognize their signifi- cant capacity to shape community identity and negotiate dominant social ideologies. acknowledgments excavations at camp street were made pos- sible by the ransom place neighborhood asso- ciation; thanks to all the association members and neighbors who contributed to the project, especially daisy borel and thomas ridley. thanks to jeanette dickerson-putman, rick ward, liz kryder-reid, bob paynter, and mark warner for discussing various elements of this project with me. references bolden, clyde nickerson indiana avenue: black entertainment boulevard. master’s thesis, school of planning, university of cincinnati, cincinnati, oh. brady, caroline m. the transformation of a neighborhood: ransom place historic district, indianapolis, – . master’s thesis, department of history, indiana university- purdue university, indianapolis, in. bundles, a’lelia on her own ground: the life and times of madam c. j. walker. scribner, new york, ny. cohen, lisabeth making a new deal: industrial workers in chicago, – . cambridge university press, cambridge, england, uk. citizens and consumers in the united states in the century of mass consumption. in the politics of consumption: material culture and citizenship in europe and america, martin daunton and matthew hilton, editors, pp. – . berg press, new york, ny. du bois, w.e.b. the philadelphia negro: a social study. reprinted in by university of pennsylvania press, philadelphia. edwards, paul k. the southern urban negro as a consumer. negro universities press, new york. reprinted in by historical archaeology ( ) prentice-hall, new york, ny. gibson-hudson, gloria j. to all classes; to all races; this house is dedicated: the walker theatre revisited. in indiana’s african- american heritage, wilma l. gibbs, editor, pp. – . indiana historical society, indianapolis. hester, jody tile analysis camp street site. manuscript, iupui archaeology lab, indianapolis, in. johnson, james weldon black manhattan. reprinted in by arno press, new york, ny. longstreth, richard city center to regional mall: architecture, the automobile, and retailing in los angeles, – . mit press, cambridge, ma. moore, leonard j. citizen klansmen: the ku klux klan in indiana, – . university of north carolina press, chapel hill. mullins, paul r. race and affluence: an archaeology of african america and consumer culture. kluwer academic/ plenum, new york, ny. praetzellis, mary, and adrian praetzellis junk! archaeology of the pioneer junk store, – . papers in northern california archaeology, . northern california anthropological group, berkeley, ca. reid, ira dea. we do our part—but… . opportunity: journal of negro life ( ): . ridley, thomas interview by author, july. manuscript, ransom place oral history project, indiana university-purdue university, indianapolis anthropology department, indianapolis, in. thornbrough, emma lou indiana blacks in the twentieth century. indiana university press, bloomington. united states department of commerce census of business: , retail distribution, vol. : types of operation. department of commerce, washington, dc. paul r. mullins department of anthropology cavanaugh b indiana university-purdue university, indianapolis indianapolis, indiana sports science - _final editorial emotional control in the leading article, ashley stirling explore coaches’ reflection on their previous use of emotionally abusive practices in the coach-athlete relationship. from in-depth, semi- structured interviews with nine national- / international-level coaches, the author found that coaches may be well-intentioned in their use of emotionally abusive coaching practices but unaware of the harmful effects of these practices. the author recommends that prevention efforts should aim to enhance the “emotional control and coping resources of the coach” (p. ). on the matter of emotional control, the legendary ucla basketball coach john wooden comes to mind. wooden stated, “you cannot function physically or mentally unless your emotions are under control” [ , p. ]. dr. don adams, who for a time was wooden’s physician has stated: in his earlier days, when he really didn’t have the success he had, we used to have the basketball games in the old men’s gym. he used to get a little wild. he was a little less under control with his emotions. he’d yell at the players a bit, but he never used a foul word, “gosh darn it!” is his strongest statement. he used to really get after the players. i suppose this is youth and his new position. there was really a change after he got better control of his assistants. i remember a few times when i was really surprised because he was sort of the model hoosier, and he had some pretty rough, but not foul, language, chewing (players) up and down. he changed with time and experience. [ , p. ]. the transition from anchored to standard putter in the special section on golf, mark guadagnoli and justin aylsworth use the first author’s challenge point framework for motor learning to propose a specific method for transferring from the anchored to standard putter in golf. the context of this study is the decision by r&a rules limited (the r&a) and the united states golf association (usga) to prohibit – from january , - anchoring the club in making a golf stroke through adoption of rule - b of the rules of golf: in adopting rule - b, the usga and the r&a have concluded that freely swinging the entire club is integral to maintaining the traditions of the game and preserving golf as an enjoyable game of skill and challenge. the essence of the traditional method of golf stroke involves the player swinging the club with both the club and the gripping hands being held away from the body. the player’s challenge is to direct and control the movement of the entire club in making the stroke. … the concept of intentionally immobilizing one end of the golf club against the body, in a manner equivalent to creating a physical attachment point to use as a fixed fulcrum or pivot point around which the club can be swung, is a substantial departure from the traditional understanding of the golf swing. [ , p. - ] international journal of sports science & coaching volume · number · iii although no more than approximately % of all golfers in all golfers in both the usa and europe are currently using anchored putting [ , p. ], the technique has been used successfully by the winners of recent major championships (including this year’s us masters champion, adam scott). references . wooden, j. and jamison, s., wooden: a lifetime of observations and reflections on and off the court, contemporary books, chicago, il, . . johnson, n., the john wooden pyramid of success: the biography, oral history, philosophy and ultimate guide to life, leadership, friendship and love of the greatest coach in the history of sports, rev. nd edn., cool titles, los angeles, ca, . . usga and the r&a, explanation of decision to adopt rule - b of the rules of golf, may, , http://www.usga.org/uploadedfiles/usgahome/rules/explanation% draft% -% may% % revision% -% final.pdf simon jenkins (leeds metropolitan university, uk) iv editorial wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ deb performance assessment in community mental health care and at-risk populations ann m. holmes, ph.d. and partha deb, ph.d. we examine whether community mental health care centers (cmhcs) dif fer in their ability to serve at-risk populations, includ­ ing clients with dual diagnoses for sub­ stance abuse, comorbid disabilities, and par ticularly severe functional impairment. our analysis uses data from indiana’s pub­ lic mental health system. although at-risk clients experience, on average, worse out­ comes than other clients, we find that some cmhcs achieve statistically significantly better outcomes than others. although this information is useful to consumers and providers who wish to identify the most ef fective providers and treatment models for at-risk clients, it is not generated in standard per formance assessments. introduction while the health care system overall has retreated from managed care, such compe­ tition-based strategies continue to be increasingly prevalent in public mental health care (mcintyre, rogers, and heier, ). the organizational and financial arrangements associated with managed care are designed to increase efficiency and reduce waste in health care deliver y. concerns have arisen that these same incentives may lead providers to under- ser ve clients, particularly individuals with severe or complicated conditions (ellis, ; ware et al., ). to guard against ann m. holmes is with indiana university-purdue university indianapolis. partha deb is with hunter college, city university of new york. the statements in this article are those of the authors and do not necessarily reflect the views or policies of indiana university-purdue university; hunter college, city university of new york; the indiana family and social ser vices; or the centers for medicare & medicaid ser vices. such potential negative outcomes, man­ aged care is typically supplemented with monitoring of provider performance. yet, “[d]espite recent research on methods of risk adjustment…, the application of this research to medicaid populations has lagged” (ireys, thornton, and mckay, ). for instance, standard methods of performance assessment focus on average outcomes, and may not detect suboptimal quality of care provided to select groups of at-risk clients. our analysis is based on data from the indiana division of mental health and addictions (idmha). idmha is the public agency that ser ves as payer of last resort for persons with persistent and severe mental illness in indiana. care is delivered through of not-for-profit cmhcs, which act as gatekeepers to the state hos­ pitals. in , the idmha adopted the hoosier assurance plan that reformed the deliver y system along managed care prin­ ciples (family and social ser vices admin­ istration, ). subsequently, idmha produced provider repor t cards that describe various aspects of the centers that reflect the quality of care provided, includ­ ing differences in assessed mental health outcomes experienced by clients at these centers (family and social ser vices administration, ). while the idmha analysis controls for baseline functioning, it ignores variance that may be due to non- clinical client factors. in addition, the idmha analysis produces only limited subgroup analysis, in part because it uses a stratified approach that severely limits the health care financing review/fall /volume , number extent to which different subgroups can be compared. as a result, the report cards cannot identify the vulnerability of some at- risk client groups, at-risk clients cannot use the information to identify optimal choices for people most like themselves, and treatments that work best on average may be applied to some clients for whom other treatment approaches may be more appropriate. in previous analysis of these data (deb, holmes, and deliberty, ), we showed the importance of adjusting performance measures for non-clinical client character­ istics (e.g., sociodemographic variables and income), and different rates of client attrition across cmhcs. in this article, we extend this analysis to examine whether per formance dif ferentials obser ved in aggregate apply to specific, vulnerable sub­ populations of clients, including clients with dual diagnoses for substance abuse, comorbid disabilities, and mental illnesses that cause particularly severe functional impairment. methods typically, estimates of provider perfor­ mance have been generated in a fixed effects framework. we use instead a mixed random effects model to evaluate provider performance. the model includes both fixed coefficients (which permit control of client risk factors on outcomes) and ran­ dom coefficients associated with provider- specific variation. we estimate the mixed random effects model in sas® with the proc mixed procedure (sas institute inc., ). in addition, we adjust provider performance for different rates of client attrition using a non-linear selection equa­ tion. the formulation of the non-linear selection equation with fixed and random coefficients is described in deb, holmes, and deliberty ( ) and estimated using the nlmixed procedure in sas® (sas institute inc., ). as a multilevel modeling technique, the mixed random effects model offers a num­ ber of advantages over standard fixed effects specifications and is particularly attractive for the objectives of this analysis. first, because outcome analysis is typically based on data with a natural hierarchical structure (clients are grouped according to cmhc), multilevel models appropriately correct standard errors for clustering ef fects. second, unlike analysis that is stratified by patient subgroups, mixed ran­ dom effects models can accommodate the cell sizes that can arise when centers ser ve relatively small numbers of particular at- risk client types. finally, multilevel models can easily accommodate interaction terms to evaluate if relative provider performance is conditional on type of client ser ved (goldstein and spiegelhalter, ). data the data for this analysis are taken from the indiana managed care provider client based data reports for fiscal years (fys) and . these data are collected on an ongoing basis for the universe of all clients who qualify for enrollment in the hoosier assurance plan. an individual is eligible for care if ( ) he or she has a severe mental, behavioral, or emotional disorder (as defined by the fourth revision of the diagnostic and statistical manual of mental disorders) that is expected to last for more than months, and that impairs functioning, ( ) is at least years of age, and ( ) is eligible for medicaid or food stamps, or has income that is below percent of the federal poverty level. the indiana data include information on , individuals who were enrolled in the health care financing review/fall /volume , number hoosier assurance plan in fys and . performance is assessed using the sample of , individuals who were enrolled continuously through this period and, thus, for whom we can obser ve changes in health status over . dependent variable “high quality care for chronic conditions entails a focus on optimizing functional sta­ tus” state clauser and bierman ( ). in this spirit, idmha requires cmhcs to rou­ tinely and comprehensively conduct func­ tional assessment for every client for whom the division covers care, and bases its report cards on this information. functioning is assessed using the hoosier assurance plan instrument - adults (hapi-a) (family and social services admini-stration, ). the hapi-a captures severity of illness on four behavioral health dimensions (symptoms of distress and mood, community functioning, social support, and risk behavior and sub­ stance abuse) and one dimension of physical health. the hapi-a has been shown to yield reliable and valid measures of health out­ comes for this population (newman et al., ; deliberty, newman, and ward, ). centers are contractually obliged to report functioning scores biannually, and these data must be supplied before reimburse­ ments are paid. each center must have at least one designated person who receives training from idmha staff on an annual basis, where training is focused on achieving reliable scoring of clients. reliability is fur­ ther enhanced by an annual audit of a sam­ ple of hapi-a scores undertaken by an accounting firm that uses trained medical personnel to evaluate the consistency with which functioning is assessed across clients and centers. we base our analysis on one subscale from the hapi-a, mental health symptoms and mood, which is constructed from rat­ ings of the client’s depression, anxiety, and symptom distress. the scale takes on val­ ues between (most ill) and (least ill). we chose to base our analysis on this one subscale because it is the one most highly correlated with the global assessment of functioning scale (a commonly employed measure of functioning used in the mental health care field), and because it has been shown to be more sensitive to changes in mental health, with effect sizes measured at -month inter vals twice that of the global assessment of functioning scale (newman et al., ). our measure of outcome is the change in the mental health symptoms and mood score between the beginning of fy and the beginning of fy . because the focus of our analysis is on persons with per­ sistent severe mental illness, measured improvements on any outcome scale tend to be modest. indeed, the average absolute change in our functioning measure is only . in this sample, and one-third of such changes were less than one in absolute value. given that the instrument used to measure outcome in this analysis has been shown to have superior sensitivity to other instruments used in the field, we believe these minimal changes simply reflect the difficulty of achieving recover y in persons with such severe illnesses. case-mix variables given our choice of dependent variable, it is necessar y to include baseline mental health functioning (as measured at the beginning of fy ) to control for possi­ ble effects of regression to the mean. we also consider a number of client socioeco­ nomic characteristics to control for possi­ ble differences in illness perception, treat­ ment efficacy, and compliance across dif­ ferent client groups. these include age and age squared (to account for possible health care financing review/fall /volume , number non-linearities found in previous studies [cuffel et al., ]), education level, sex, family income, marital status, and race and ethnicity. we also include two clinical vari­ ables, measured at baseline, including whether or not the client had a disability other than a mental disorder (including being blind, deaf, mute, non-ambulator y, neurologically impaired, developmentally disabled, or illiterate), and whether the center considered the client to be at risk for substance use. to correct for possible sample selection bias, we also include a second order approximation of the true selection index as recommended by vella ( ). the selection index is derived from an analysis in which client retention across the cmhcs is expressed as a function of obser vable client characteristics. refer to deb, holmes, and deliberty ( ) for more detailed information. center-level variables center-level effects are inferred from indicator variables in the mixed random effects model. to determine if cmhcs dif­ fer in their ability to ser ve particular at-risk populations, we interact these center indi­ cator variables with risk indicators for client groups of concern. these groups include clients with dual diagnoses for sub­ stance abuse (icd- codes , , and/or ), clients with comorbid dis­ abling conditions (including being blind, deaf, mute, non-ambulator y, illiterate, and/or having a developmental disability or neurological impairment), and clients with mental illness that causes severe func­ tional impairment (as determined by an idmha algorithm that considers mental and physical health, social and community functioning, and risk behaviors). these interaction terms allow the slope coeffi­ cients on the client risk factors to var y by cmhc, and can be used to determine if a table prevalence of at-risk populations sub-group prevalence percent mental illness with severe functional impairment . co-occurring substance abuse . comorbid disability . note: results are based on all , clients served in fys and . source: holmes, a.m., indiana university-purdue university indianapolis and deb, p., hunter college, city university of new york: data based on estimates from the indiana managed care provider client based data reports, fys - . particular cmhc performs significantly better or worse when ser ving individuals from specific vulnerable client groups. summar y statistics for these variables are presented in table . analysis plan we estimate two models, one with center indicators interacted with at-risk variables, and one without for comparison purposes. overall model fit is evaluated using both akaike and bayesian information criteria (aic and bic). both are related to the adjusted r statistic, but with slightly dif­ ferent adjustments for the number of inde­ pendent variables, and with smaller values of the test statistic associated with better model fit. because the bic also adjusts for sample size, it is less likely to favor over-fit­ ted models. to evaluate the robustness of center per­ formance differentials across the at-risk client subgroups, we consider the follow­ ing: first, the overall importance of the variation in per formance across all cmhcs for different client subgroups is assessed by examining the covariance parameter estimates associated with each group of interaction terms. standard t-sta­ tistics associated with the individual inter­ action coefficients can be used to deter­ mine the extent to which any one cmhc may produce statistically significantly bet­ ter or worse outcomes for a particular at-risk health care financing review/fall /volume , number table model fit and covariance parameters category overall model fit covariance parameter model aic bic base model . . — intercept — — . ( . ) interaction model * . * . intercept — — . ( . ) mental illness with severe functional impairment — — . ( . ) co-occurring substance abuse — — . ( . ) comorbid disability — — . ( . ) *indicates best model based on fit criteria. notes: n= , . aic is akaike information criteria. bic is bayesian information criteria. numbers in parentheses are p-values. source: holmes, a.m., indiana university-purdue university indianapolis and deb, p., hunter college, city university of new york: data based on estimates from the indiana managed care provider client based data reports, fys - . client subgroup. second, we calculate the largest change, both positive and negative, in ranks inferred from the center indica­ tors in the overall model and the coeffi­ cients on the interaction terms associated with each at-risk client group. third, we calculate the proportion of changes in rela­ tive ranks between each at-risk group and the overall center ranks. the number of changes in relative ranks is given by -w/ , where w is kendall’s measure of concor­ dance between the ranks implied by the two groups being compared. fourth, we calculate the correlation in relative perfor­ mance and implied ranks between each at- risk group and the overall client popula­ tion. results model fit statistics are presented in table . based on both selection criteria, the model that includes interaction terms dominates the model that assumes relative performance differentials are the same for all client subgroups. the covariance para­ meter estimates, also presented in table , indicate that there are significant differ­ ences in relative center performance for each at-risk group, with the most signifi­ cant differences being obser ved for the client group with severe functional impair­ ment due to mental illness (p= . ), fol­ lowed by the group with co-occurring sub­ stance abuse (p= . ). estimates of the fixed coefficients that capture the effect of client case-mix vari­ ables (available from the authors on request) are robust to the inclusion of interaction terms. the random ef fects solutions in table provide information on the relative performance of the commu­ nity mental health centers. these random effects coefficients represent the estimat­ ed deviations for each center from the mean performance score, with positive (negative) estimates indicating the center per formed above (below) the average level. the coefficients on the interaction terms, by comparison, represent the esti­ mated deviations in performance score between the at-risk group considered and the not at-risk group for that particular cmhc. in the overall model, four cmhcs are found to perform significantly (p< . ) bet­ ter than average, and six perform signifi­ cantly worse than average. although the magnitude of these provider-level coeffi­ cients may appear to be small, it is impor­ tant to note that they measure the devia­ tions from the average change in function­ ing score. given the mean absolute change in functioning score is only . , even a health care financing review/fall /volume , number table random effects solutions for provider performance differentials interaction model mental illness with base model severe co-occurring implied intercept implied functional implied substance implied comorbid implied cmhc intercept rank (not at-risk) rank impairment rank abuse rank disability rank a * . * . - . **- . - . b * . * . . . - . c * . * . . . na — d * . * . . *- . . e ** . . - . . ** . f ** . . . . . g . . . . . h . . . - . . i . . . na — . j . . - . . - . k . . - . - . na — l . . . - . - . m . . *- . *- . ** . n . . - . . *- . o . - . . - . . p . - . - . * . - . q - . - . - . - . . r - . . - . - . - . s - . - . . . . t - . - . - . na — . u - . - . **- . na — - . v - . - . . **- . na — w - . - . - . *- . - . x **- . *- . . . . y *- . **- . - . - . - . z *- . **- . . na — - . aa *- . *- . - . - . - . bb *- . *- . . - . - . cc *- . **- . . na — - . dd *- . *- . *- . na — . *p< . . **p< . . notes: cmhc is community mental health care center. na indicates the cmhc served no clients in the at-risk category. at-risk coefficients under the interaction model represent the marginal differences in performance, while at-risk ranks are based on levels of performance. source: holmes, a.m., indiana university-purdue university indianapolis and deb, p., hunter college, city university of new york: data based on estimates from the indiana managed care provider client based data reports, fys - . -point difference would be considered substantial. also, the coefficients measure the average deviation for all clients treated at the center. thus, a coefficient of + would correspond with improving the func­ tioning of ever y client at the center by one additional point (on average) compared with the mean center. improvements above + are in the top one-third of all improve­ ments for this population, so a center with a coefficient greater than one would have essentially moved their clients from out­ comes in the middle one-third of the distri­ bution to outcomes in the top one-third of the distribution. across all centers, outcomes are much worse for clients whose mental illnesses caused severe functional impairment (aver­ age interaction coefficient of - . ), and somewhat worse for clients with co-occur­ ring substance abuse (- . ) and comor­ bid disabling conditions (- . ). although there are only a small number of signifi­ cant differences for the at-risk groups, at health care financing review/fall /volume , number table correlations in performance differentials across patient subgroups mental illness with severe functional co-occurring comorbid category not at-risk impairment substance abuse disability overall * . / . * . / . * . / . * . / . not at-risk — * . / . *- . / . *- . / . mental illness with functional impairment — — * . / . *- . / . co-occurring substance abuse — — — **- . / . *p< . . **p< . . notes: pearson’s correlation of performance differentials/spearman’s correlation of ranks. source: holmes, a.m., indiana university-purdue university indianapolis and deb, p., hunter college, city university of new york: data based on estimates from the indiana managed care provider client based data reports, fys - . least one cmhc performs statistically sig­ nificantly worse for ever y at-risk group considered. in addition, the variability in performance is much greater for clients whose mental illnesses cause severe func­ tional impairment than for other clients. a comparison of the implied ranks across the subgroups considered (also pro­ vided in table ) reveals the largest change is between the ranks for the overall client population and the ranks for the group with mental illness causing severe functional impairment. the maximum changes were a -percentile increase and a -percentile decrease in rankings), fol­ lowed by the group with co-occurring sub­ stance abuse (with a -percentile increase and -percentile decrease, respectively), followed by the group with a disabling comorbidity (with a -percentile increase and a -percentile decrease, respectively), and lastly followed by the not at-risk group (with a -percentile increase and a -per­ centile decrease, respectively). this order­ ing is preser ved when comparing the pro­ por tion of relative ranks that change between the overall and at-risk rankings: rank reversals are nearly nine times more likely between the overall ranks and the ranks for the group with mental illness causing severe functional impairment than between the overall ranks and the ranks for the not at-risk group. rank reversals for the groups with co-occurring substance abuse and other disabling conditions are, respectively, seven and three times more likely than for the not at-risk group. correlation coefficients are presented in table . the correlations across estimated performance differentials are statistically insignificant and only weakly positive in size for all at-risk groups considered and the overall client population. in contrast, the correlation between the overall differ­ entials and the not at-risk differentials is . and highly statistically significant (p= . ). thus, it appears that relative center performance overall is determined largely by its ability to ser ve less vulnera­ ble clients. although correlations between implied ranks are, by contrast, statistically significant, the strength of association is only moderately strong, par ticularly between the overall ranks and the ranks for the group with mental illness causing severe functional impairment. discussion the president’s new freedom commis­ sion on mental health ( ) identified out­ come assessment and accountability as unique challenges to the successful func­ tioning of the mental health care system. problems of asymmetr y of information, in which providers know more about patients’ conditions than either insurers or patients themselves, are particularly acute in mental health care financing review/fall /volume , number health care and, combined with incentives for risk selection, can place the neediest patients in peril (frank and mcguire, ). outcome assessment is needed to ensure these quality problems are not exac­ erbated by managed care deliver y systems that increasingly characterize publicly fund­ ed community mental health care. access to community-based care for per­ sons with even the most debilitating men­ tal illnesses was advocated by the new freedom commission on mental health ( ) which recognized that mental health care should be consumer and fami­ ly driven. with this authority comes the responsibility for selecting optimal care from community providers and the need for policymakers to provide the informa­ tion consumers need to make these choic­ es, including those consumers with partic­ ularly severe or complicated conditions. the commission also reported that dis­ parities exist in access to appropriate men­ tal health care and the burden of mental ill­ ness borne by certain segments of the population. in particular, the care for per­ sons with co-occurring disorders was found to be inadequate. administrators of public mental health care systems need to consider the extent to which they meet the needs of such at-risk subgroups. similarly, researchers who undertake effectiveness research to identify best treatment prac­ tices need to consider not only what works best for the typical client, but also whether these same practices are optimal for more vulnerable clients. standard provider-profiling exercises fail to identify whether some providers are par­ ticularly effective in the treatment of the most vulnerable at-risk clients, and these clients cannot use the resulting informa­ tion to identify optimal choices for people most like themselves (elliott et al., ). while stratified analysis has been suggest­ ed as a possible solution to these problems, strata-specific risk rates typically have unsatisfactor y statistical properties, partic­ ularly for under-represented client groups (gatsonis et al., ). this feature is par­ ticularly undesirable if the most vulnerable at-risk clients are infrequently encountered in cmhcs. in this article, we used a mixed random effects model to evaluate provider perfor­ mance. compared with standard provider- profiling exercises, such models yield more precise estimates of relative performance, especially when sample sizes are small. in addition, the model easily accommodates interaction terms to evaluate whether per­ formance differentials are robust across various client subgroups. our results sug­ gest that, for some cmhcs, relative perfor­ mance is significantly dependent on the type of client ser ved; while, on average, centers attained poorer outcomes for at-risk clients than less vulnerable clients, the dis­ crepancy was larger for some centers than others. furthermore, the estimated perfor­ mance differences for at-risk populations were only moderately related to overall per­ formance differences, with the result that standard provider profiles sometimes failed to identify the most effective providers of care for at-risk clients. we also found that per formance dif ferentials varied much more for clients with mental illnesses that resulted in severe functional impairment than for clients with less severe illnesses. policymakers need to be aware that in such situations these at-risk clients may have to travel relatively greater distances to obtain quality care, further aggravating disparities in health status and access to health care (dranove et al., ). while the number of centers with statis­ tically significantly better (or worse) out­ comes for various at-risk client groups may be small, the results still have practical relevance. by identifying a small number of exemplar y centers, we have identified health care financing review/fall /volume , number centers whose practices, etc., may be wor th emulating by other providers. similarly, by identifying a small number of centers with subpar performance, we have identified centers where quality improve­ ment initiatives by state agencies could be most effectively applied. our results can also be used to assess the distribution of quality care across different regions of the state, both overall and with respect to vul­ nerable subgroups. although we believe our empirical model offers a number of advantages over standard specifications, a number of caveats deser ve mention. first, the analy­ sis is based on only one clinical measure— change in mental health symptoms and mood over a -year period. our relative rankings may discriminate against centers that place more priority on other dimen­ sions of mental health (e.g., community functioning, reduction of substance abuse risk), or that focus on longer or shorter time horizons. second, our results are based on data for a single state over a sin­ gle year. the external validity of our find­ ings may be limited to the extent that sys­ tem, practice, or client differences may exist across geographical regions or time, although the methods we have presented for detecting differences in provider per­ formance for vulnerable at-risk populations remain valid regardless of setting. third, our analysis can only consider differences in performance across cmhcs, and not differences within a given cmhc. as a consequence, our results cannot inform consumers and insurers about the relative ef fectiveness of individual providers or treatments. however, given that idmha clients must select annually a cmhc to ser ve as a mental health care gatekeeper (rather than a specific provider or treat­ ment protocol), center-level comparisons remain useful. finally, our results only indi­ cate that differences exist, not why they exist. one of the advantages of the meth­ ods used in this article is that it is possible, in theor y, to incorporate center-level vari­ ables in the mixed random effects specifi­ cation to identify center characteristics associated with better per formance. empirically, however, our ability to assess the impact of multiple center-level charac­ teristics is limited given the small number of centers on which our analysis is based. the results of this article do provide a crit­ ical first step in quality improvement—hav­ ing identified exceptional centers, policy- makers can use this information in future studies to help determine the staffing, practice patterns or organizational struc­ tures that are associated with superior or inferior outcomes. acknowledgments we would like to thank richard n. deliberty for his assistance. references clauser, s.b. and bierman, a.s.: significance of functional status data for payment and quality. health care financing review ( ): - , spring . cuf fel, b.j., jeste, d.v., halpain, m., et al.: treatment costs and use of community mental health ser vices for schizophrenia by age cohorts. american journal of psychiatry ( ): - , . deb, p., holmes, a.m., and deliber ty, r.n.: adjusting for patient characteristics and selection ef fects in assessment of community mental health centers. medical care ( ): - , . deliberty, r. n., newman, f.l., and ward, e.o.: risk adjustment in the hoosier assurance plan: impact on providers. journal of behavioral health services & research ( ): - , . dranove, d., kessler, d., mcclellan, m., et al.: is more information better? the effects of report cards on health care providers. journal of political economy ( ): - , . elliott, m. n., swartz, r., adams, j., et al.: case-mix adjustment of the national cahps benchmarking data . : a violation of model assumptions? health services research ( ): - , . health care financing review/fall /volume , number ellis, r. p.: creaming, skimping and dumping: provider competition on the intensive and extensive margins. journal of health economics ( ): - , . family and social ser vices administration biennial report. division of mental health, office of public policy: indianapolis, in. . family and social ser vices administration the hoosier assurance plan provider profile report card. treatment outcomes for adults with a serious mental illness. division of mental health, office of public policy: indianapolis, in. . frank, r. g. and mcguire, t.g.: economics and mental health. in culyer, a.j., and newhouse, j.p. (eds.): handbook of health economics, volume . elsevier press. new york. . gatsonis, c. a., epstein, a.m., newhouse, j. p., et al.: variations in the utilization of coronar y angiography for elderly patients with an acute myocardial infarction. an analysis using hierarchical logistic regression. medical care ( ): - , . goldstein, h. and spiegelhalter, d.j.: league tables and their limitations: statistical issues in comparisons of institutional performance. journal of the royal statistical society (a) ( ): - , . ireys, h.t., thornton, c., and mckay, h.: medicaid managed care and working-age beneficiaries with disabilities and chronic illnesses. health care financing review ( ): - , fall . mcintyre, d., rogers, l., and heier, e.j.: over view, histor y, and objectives of per formance measurement. health care financing review ( ): - , spring . new freedom commission on mental health: achieving the promise: transforming mental health care in america. final report. dhhs pub. no. sma- - . rockville, md. . newman, f., deliberty, r., hodges, k., et al.: the hoosier assurance plan: repor t of research/ implementation strategy and assessment instruments to support level of care determination. indiana division of mental health working paper. indianapolis, in. . sas institute inc.: sas®/stat user’s guide: version . sas publishing. car y, nc. . vella, f.: estimating models with sample selection bias: a sur vey. the journal of human resources ( ): - , . ware, j.e., bayliss, m.s., rogers, w.h., et al.: differences in -year health outcomes for elderly and poor, chronically ill patients treated in hmo and fee-for-ser vice systems. results from the medical outcomes study. journal of the american medical association ( ): - , . reprint requests: partha deb, ph.d., economics department, hunter college, park avenue, w, new york, ny . e-mail: partha.deb@hunter.cuny.edu health care financing review/fall /volume , number - .. hoosier oncology group randomized phase ii study of docetaxel, vinorelbine, and estramustine in combination in hormone-refractory prostate cancer with pharmacogenetic survival analysis noahm.hahn, sharonmarsh, william fisher, robertlangdon, robin zon, markbrowning, cynthias.johnson, tiffany j.scott-horton, lang li, howard l.mcleod, andchristopher j.sweeney abstract purpose:to determine the safety and efficacy of two docetaxel doublets inhormone-refractory prostate cancer (hrpc) patients and to examine the prognostic role of polymorphisms in host genes important to docetaxelmetabolismand transport. experimental design: sixty-four chemotherapy-naive patients with hrpc were randomized to docetaxel and vinorelbine (d, mg/m i.v. days and ; v, mg/m i.v. days and ) or docetaxelandestramustinephosphate(d, - mg/m i.v.day ;e, mgoralthricedailydays - ) administered q d. primary end point was clinically significant toxicity. a pharmacogenetic analysis ofhostgenes was doneinpatients who receivedat leastone cycle ofdocetaxel therapy. results: grade / toxicity occurred in . % of dv patients andin . % de patients. neither armexceededthethresholdofclinicallysignificant toxicity. inthedvarm,objective responserate was %, prostate-specific antigen response rate was %, and median survival was . months. inthe de arm, objective response ratewas %,prostate-specific antigenresponserate was %,andmediansurvivalwas . months.pharmacogeneticanalyses showedasignificant associationbetweensurvivalbeyond months andtheabcg c>a(q k)polymorphism compared with the wild-type (c/c) genotype ( % versus %; p = . ). conclusions: dv and de doublets are active with a tolerable toxicity profile in patients with hrpc;however,efficacydoesnotseemsuperior tostandardsingle-agentdocetaxel.theabcg c>a(q k)polymorphismmaybeanimportantpredictorof responseandsurvival inhrpc patients treated with docetaxel-based chemotherapy. prostate cancer afflicts , men yearly and results in > , deaths ( ). recent trials have shown a survival benefit for men with hormone-refractory prostate cancer (hrpc) treated with a docetaxel-based regimen ( , ). vinorelbine is a vinca alkaloid that inhibits the microtubular apparatus in malignant cells and has documented activity in hrpc( – ).vinorelbineanddocetaxelinhibitdifferentportions ofthemicrotubuleapparatusinasynergisticmannerin preclinical in vitro prostate cancer models ( – ). with this in mind, the hoosier oncology group conducted a randomized phase ii trial of weekly docetaxel and vinorelbine (dv) with docetaxel and estramustine phosphate (de) as a parallel reference arm. this study also included a pharmacogenetic analysis of host genes with critical roles in docetaxel drug transport and metabolism. clinical studies suggest that variant alleles in the thiopurine methyltransferase gene can identify patients at risk of severe hematologic toxicity after azathioprine, -mercaptopurine, or related agents ( ). similar toxicity asso- ciations are seen with dihydropyrimidine dehydrogenase gene mutations and -fluorouracil therapy ( ) and with ugt a mutations and irinotecan ( ). single-nucleotide polymor- phisms (and other genetic variants) have also been observed, which affect docetaxel metabolism and transport. these include alterations in the degradation pathway (cytochrome p a and a ; refs. – ), microtubule associated pro- teins (map and mapt; ref. ), and drug transport proteins [multidrug resistance (abcb ) and breast cancer resistance protein (abcg ); refs. , ]. polymorphisms in these genes cancer therapy: clinical authors’affiliations: department of medicine and division of biostatistics, indiana university school of medicine, indianapolis, indiana; department of medicine,washington university schoolof medicine, st. louis, missouri; medical consultants, p.c., muncie, indiana; nebraska methodist hospital and cancer center, omaha, nebraska; northern indiana cancer research consortium, south bend, indiana; and oncology hematology associates of southwest indiana, evansville, indiana received / / ; accepted / / . grant support: glaxosmithkline (c.j. sweeney), sanofi aventis (c.j. sweeney), walther cancer institute (c.j. sweeney), and nih pharmacogenetics research networkgrant gm (h.l. mcleod). the costs ofpublicationof this articlewere defrayedinpartby thepaymentofpage charges.this article must therefore be hereby marked advertisement in accordance with u.s.c. section solely to indicate this fact. note: presented in abstract form at the proceedings of theamerican society of clinicaloncologyannualmeeting,may - , ,orlando,florida,abstract . requests for reprints: christopher j. sweeney, division of hematology and oncology, indiana university cancer center, barnhill drive, indiana university cancer pavilion, room , indianapolis, in . phone: - - ; fax: - - ; e-mail: chsweene@iupui.edu. f american associationforcancerresearch. doi: . / - .ccr- - www.aacrjournals.orgclin cancer res ; ( ) october , research. on april , . © american association for cancerclincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ were assessed to identify genetic variants that predict for toxicity and efficacy of docetaxel-based chemotherapy. materials and methods eligibility criteria. the protocol was approved by the institutional review boards of all participating centers. key eligibility criteria are listed in fig. . in addition, all patients had progressive disease after at least one hormonal therapy (orchiectomy, estrogens, luteinizing hormone–releasing hormone therapy, etc.) with castrate levels of testosterone (< ng/dl). progressive disease was defined as one of the following: an increase in prostate-specific antigen (psa) > % over nadir on hormonal therapy measured on two successive occasions at least weeks apart ( ) and/or objective evidence of progressive disease on computed tomography scan and/or new symptomatic bone metastases. patients treated with an antiandrogen must have also progressed after the antiandrogen had been discontinued for at least weeks (or for weeks for antiandrogens with a longer half-life, such as bicalutamide) and have continued evidence of disease progression (at least a % increase in psa after discontinuing). there was no maximum number of hormonal therapies allowed. patients were excluded if they had new or unstable central nervous system metastases, had received external beam radiation therapy to > % of their bone marrow, or had a history of grade z peripheral neuropathy. patients with active cardiac disease defined as active angina, symptomatic congestive heart failure, or myocardial infarction within the previous months were excluded. patients with a history of deep venous thrombosis, pulmonary embolism, or cerebrovascular attack within the last months were also excluded. patients who had received external beam radiation or samarium injection < weeks before enrollment and patients who had received a strontium injection < weeks before enrollment were ineligible. patients with a history of prior malignancy < years disease-free with the exception of curatively treated basal and squamous skin cancers were excluded. patients were excluded if they had received pc-spes within weeks before enrollment. in addition, the use of saw palmetto or lycopenes was not allowed. treatment. patients were randomized to one of two regimens. patients on the de arm received mg of oral estramustine phosphate thrice daily to hours after a meal from days to with dairy products restricted on those days. patients received i.v. infusion of docetaxel over minutes at a dose of mg/m on day . if no grade or toxicities occurred in the first cycle, docetaxel was increased to mg/m for all subsequent cycles. patients received mg dexamethasone orally every hours for five doses beginning the night before chemotherapy. from day until weeks after discontinuation of estramustine phosphate, patients received mg of enteric-coated aspirin as prophylaxis for arterial thrombosis and mg of oral warfarin. patients randomized to the dv arm received i.v. vinorelbine over to minutes at a dose of mg/m on days and . patients also received i.v. docetaxel over minutes at a dose of mg/m on days and . all patients received mg dexamethasone orally every hours for days beginning the day before treatment. in patients with a prior history of external beam radiation, vinorelbine and docetaxel were reduced to and mg/m , respectively. in both de and dv arms, treatments were repeated on a -day cycle up to a total of six cycles or until disease progression or dose-limiting toxicity. patient evaluation. all patients had a history, physical exam, and psa within week before treatment. other laboratories (complete blood count, aspartate aminotransferase, alanine aminotransferase, alkaline phosphatase, total bilirubin, testosterone, psa, and coagula- tion studies) were done within weeks before treatment. all radiologic studies (bone scan, abdomen and pelvis computed tomography scan, and chest x-ray) were done within weeks before treatment. before each cycle, a case report form, complete blood count, hepatic panel, psa, and coagulations laboratories were recorded. radiographic studies were repeated before each odd-numbered cycle and in follow-up for patients with documented objective responses to therapy. disease assessment. patients with measurable disease were assessed for response to therapy according to standard response evaluation criteria in solid tumors criteria. patients with an elevated psa were assessed for a serologic psa response. psa complete response was defined as an undetectable psa if prior prostatectomy and < ng/ml if prostate in place (after irradiation, no local therapy, etc.) on two consecutive measurements weeks apart. psa partial response was defined as a decline of psa value by > % for two consecutive determinations at least weeks apart. psa stabilization was defined as a < % increase or decrease in psa over a period of months if baseline psa > ng/ml. psa progression was defined as an increase in psa to > % above baseline on two consecutive measures at least weeks apart. psa relapse was defined as a > % increase in psa over the minimum value recorded during a response as evidenced on three successive determinations. pharmacogenetic evaluation. all patients who received at least day of docetaxel therapy on either trial arm were eligible for inclusion in the companion pharmacogenetic study. participation in the pharmacoge- netic study was not required for participation in the treatment portion of the trial. after a separate informed consent was obtained, a single -ml blood sample collected in an edta tube was drawn from each patient. samples were stored at jc to jc for v days and shipped to washington university (st. louis, mo) for analysis. dna was extracted from whole blood using the gentra puregene kit (gentra systems, inc., minneapolis, mn) following the manufacturer’s instruc- tions. eight polymorphisms in six genes associated with taxane metabolism (cyp a * b and cyp a * c), transport (abcb c>t and abcg c>a), and microtubule assembly (map g>a, map c>a, mapt- g>a, and mapt a>g) were assessed using pcr and pyrosequencing technology as described previously ( – ). statistical analysis. the primary end point of the study was the incidence of clinically significant toxicity defined as any significant adverse event, which required reporting or need for rbc or platelet transfusion that was thought to be related to treatment. events that required reporting were serious adverse events or reactions that were at least possibly related to therapy and resulted in death, or were life threatening, required inpatient hospitalization or prolongation of existing hospitalization, or resulted in persistent or significant disability.fig. . keyeligibility criteria. docetaxel,vinorelbine, and estramustine phase ii study www.aacrjournals.org clin cancer res ; ( ) october , research. on april , . © american association for cancerclincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ secondary end points included objective, psa, and clinical response rates, time to progression, and pharmacogenetic analyses. patients were stratified based on baseline performance status, pain requiring opiates, and presence of measurable disease. there was no intention to statistically compare the two arms. eighteen patients were treated on each arm in the first stage. a study arm was to be terminated if z patients, among the first , experienced clinically significant toxicity. this did not occur in either arm; therefore, additional patients were assigned to each arm in stage ii. if z patients, among the total of each arm, experienced clinically significant toxicity, the arm was to be considered not worthy of further testing. these numbers were determined by the simon two- stage phase ii design and provided the study with % power to detect a true toxicity rate of % with a % type i error rate. confidence intervals for clinically significant toxicity rates of each arm were obtained using jennison and turnbull ( ) method. confidence intervals for response rates were obtained using an exact method. curves for time to progression and overall survival were estimated using kaplan-meier method. toxicity grades were summarized using contin- gency tables. pharmacogenetic polymorphism frequencies were ana- lyzed in relation to clinical benefit, psa response, and overall survival using m proportional analyses. for the survival analysis, a cutoff of months was chosen to approximate the expected median survival in study subjects. progression-free survival was defined as the time from enrollment on study until either measurable progressive disease was noted, % increase in psa from nadir was observed and confirmed at least weeks later, or clinical deterioration (decline in karnofsky performance score of z points from baseline, an increase in opiate requirements of mg i.m. equivalents, or death due to any cause) was noted. overall survival was defined as the time from enrollment on study until death due to any cause. results patients. from march , to april , , patients were randomized, to each arm. the median age was years. median pretreatment psa level was . the median karnofsky performance score was with a median pain score of and median i.m. morphine equivalent opiate use of mg/d. a total of ( . %) patients had measurable disease at baseline. baseline characteristics are summarized in table for the entire group and each arm individually. there were no significant differences in any of the baseline stratification factors. toxicity. in general, treatment was well tolerated. nine of ( . %) patients [ % confidence interval ( % ci), . - . ] experienced grade or toxicity in the de arm. grade toxicities included diarrhea, febrile neutropenia, hypoglycemia/ hypokalemia, increased lacrimation, hypersensitivity reactions, typhlitis, urinary retention, and elevated transaminases (one patient each). grade toxicities included hematuria, hypergly- cemia/fatigue, stroke, and elevated transaminases (one patient each). five of ( . %) patients ( % ci, . - . ) expe- rienced grade or toxicity in the dv arm. grade toxicities included bone pain, edema, hyperglycemia, venous catheter- associated infection, and recurrent urinary tract infections (one patient each). grade toxicity was limited to a single episode of hypocalcemia. neither arm exceeded the prespeci- fied limit of six clinically significant toxic events as previously defined. chemotherapy administration. a total of cycles of de chemotherapy was given to patients. of these, . % of cycles were given without dose reduction or delay. the most common reasons for dose reduction or delay were infectious ( . %; cellulitis, fever, neutropenic fever, and herpes zoster; all one patient each), hematologic ( . %; four patients with neutropenia), unknown ( . %; three patients), and gastroin- testinal toxicity ( . %; hepatotoxicity and nausea/emesis; one patient each). in the dv arm, cycles of chemotherapy were adminis- tered, . % without dose reduction or delay. the most common causes for reduction or delay were prior radiotherapy ( . %; patients, mandated at baseline per treatment plan), hematologic ( . %; patients with neutropenia, patients with thrombocytopenia), and infectious ( . %; patients with febrile neutropenia). response and survival. among patients with measurable disease in the de arm, six of nine ( . %) patients had a partial or complete response. a partial psa response was observed in . % of patients. median progression-free survival was . months ( % ci, . - . ), with an overall survival of . months ( % ci, . ; not reached). in the dv arm, of ( . %) patients with measurable disease had a partial or complete response. a partial psa response was observed in . % of patients. median progres- sion-free survival was . months ( % ci, . - . ), with an overall survival of . months ( % ci, . - . ). kaplan- meier curves for progression-free survival and overall survival are summarized for both de and dv arms in figs. to . pharmacogenetic analysis. samples for pharmacogenetic analyses were available for patients. in this pilot analysis, it was noted that of ( %) patients with the abcg c>a (q k) polymorphism were alive past months compared with only of ( %) patients with wild-type (c/c; p = . ). polymorphisms in other candidate genes (cyp a , cyp a , abcb , mapt, and map ) were not table . baseline patient demographics all patients arm dv arm de n median (range) n median (range) n median (range) age ( - ) ( - ) ( - ) psa . ( . - , . ) . ( . - , . ) . ( . - , . ) kps ( - ) ( - ) ( - ) mean pain score . ( . - . ) . ( . - . ) . ( . - . ) median opiate use* ( - ) ( - ) ( - ) abbreviation: kps, karnofsky performance score. *i.m. morphine equivalents. cancer therapy: clinical www.aacrjournals.orgclin cancer res ; ( ) october , research. on april , . © american association for cancerclincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ associated with an increased chance of being alive beyond months (table ). pharmacogenomic analyses of toxicity or psa response revealed no additional significant associations (all p > . ). discussion metastatic hrpc remains the third leading cause of cancer- related mortality among men in the united states. in , petrylak et al. ( ) and tannock et al. ( ) reported results of two separate randomized controlled trials confirming a survival advantage for docetaxel-based chemotherapy. however, the potential toxic effects of chemotherapy remain a valid consideration as we evaluate new combinations. in previous studies combining estramustine phosphate and docetaxel in the treatment of hrpc patients, objective response rates of % to % were observed, with most responses characterized as partial responses. psa responses range from % to %. major toxicities include myelosuppression, fatigue, and nausea. grade or neutropenia is seen in up to half of patients. with therapy, overall survival ranges between and months with variations depending on patient characteristics at baseline ( – ). in the present study, of ( . %) patients ( % ci, . - . ) treated with de experienced grade or toxicity. an objective response rate of . %, a psa response rate of . %, and a median overall survival of . months were observed. these results compare favorably with the pivotal phase iii randomized trials described above in terms of efficacy of every -week docetaxel and toxicity of the docetaxel and estramustine combination. vinorelbine is a semisynthetic vinca alkaloid that has shown activity against solid organ malignancies, including lung, breast, and germ cell tumors. its preclinical synergism when combined with docetaxel in prostate cancer cell line experi- ments was a basis for its inclusion in this trial ( ). furthermore, as a single agent, it is tolerable in elderly patients with metastatic hrpc ( – ). in the present study, of ( . %) patients ( % ci, . - . ) treated with dv experienced grade or toxicity and had clinically significant toxicity per the protocol. an objective response rate of . %, a psa response rate of . %, and a median overall survival of . months were observed. koletsky et al. ( ) have reported the only other phase ii data combining vinorelbine and docetaxel to date. in the koletsky study, % of patients experienced grade or neutropenia. one case of acute respiratory distress syndrome fig. . a, overall survival for de arm. b, overall survival for dvarm. fig. . a, progression-free survival for de arm. b, progression-free survival for dvarm. docetaxel,vinorelbine, and estramustine phase ii study www.aacrjournals.org clin cancer res ; ( ) october , research. on april , . © american association for cancerclincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ was also observed. an objective response was seen in % of patients with a psa response in % of patients also. patients in the koletsky study had a significantly lower median psa level at baseline ( ng/ml) compared with our study population ( ng/ml), which may account for the somewhat decreased response rates seen in our experience. insufficient antitumor activity associated with weekly docetaxel dosing must also be considered. the weekly dosing of both docetaxel and vinor- elbine in our trial was based on prior results of a phase i dose escalation study in non–small cell lung cancer patients, which revealed a maximally tolerated dose of mg/m /wk of vinorelbine and mg/m /wk of docetaxel without scheduled treatment breaks ( ). the total number of patients experiencing clinically signifi- cant toxicity with the combination of vinorelbine and docetaxel did not exceed our prestudy threshold. however, the efficacy results do not suggest an advantage compared with single-agent docetaxel. as such, we would not recommend further study of the combination regimen at this dosing schedule. however, clinicians should be reminded of the single-agent tolerability of vinorelbine, particularly in elderly patients often encountered with metastatic hrpc. the companion pharmacogenetic study assessed germ-line polymorphisms in genes known to play important roles in chemotherapy drug transport, metabolism, and mechanism of action. analysis of candidate genes showed a significantly greater proportion of patients surviving beyond months with docetaxel-based therapy in the presence of the abcg c>a polymorphism. previous studies of the abcb -associated p-glycoprotein-mediated drug efflux pump have shown mini- mal, if any, effect on docetaxel plasma concentrations when an inhibitor of this pump is administered ( ). the effect of abcg polymorphisms on docetaxel pharmacokinetics is unknown. the increased survival seen in individuals with an abcg c>a polymorphism may suggest a less functional drug efflux pump, leading to increased intracellular (intra- tumoral) docetaxel concentrations and improved cytotoxic activity. this hypothesis should be interpreted cautiously due to the small patient sample size and potential confounding variables. in addition, this exploratory analysis was not subject to a multivariate analysis. therefore, at most, these data are hypothesis generating and will be used as preliminary data for future large-scale studies. nonetheless, the study shows the ability to successfully conduct translational pharmacogenetic studies in a community setting, such as the hoosier oncology group. in summary, this randomized phase ii study of two docetaxel-based regimens showed a tolerable toxicity profile with a weekly docetaxel and vinorelbine regimen. however, the efficacy data do not support evaluation in the phase iii setting. a pharmacogenetic analysis of germ-line dna showed that patients with an abcg c>a genotype had an increased chance of being alive beyond months if treated with docetaxel-based combination chemotherapy. this study shows that modern translational efforts can be accomplished beyond the confines of large tertiary academic medical centers and should serve as a model for future trial design. table . pharmacogenetic polymorphism analysis in patients surviving > months gene polymorphisms p abcb c/c, / ( %) c/t, / ( %) t/t, / ( %) . abcg c/c, / ( %) c/a, / ( %) a/a, / ( %) . cyp a * b a/a, / ( %) a/g, / ( %) g/g, / ( %) . cyp a * c a/a, / ( %) a/g, / ( %) g/g, / . map g/g, / ( %) g/a, / ( %) a/a, / ( %) . map c/c, / ( %) c/a, / ( %) a/a, / ( %) . mapt- g/g, / ( %) g/a, / ( %) a/a, / ( %) . mapt a/a, / ( %) g/a, / ( %) g/g, / ( %) . cancer therapy: clinical www.aacrjournals.orgclin cancer res ; ( ) october , references . jemal a, siegel r,ward e, et al. cancer statistics, . ca: a cancerjournal for clinicians ; : ^ . . petrylak dp,tangen cm, hussain mh, et al. doce- taxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. nengljmed ; : ^ . .tannock if, dewit r, berrywr, et al. docetaxelplus prednisone or mitoxantrone plus prednisone in advanced prostate cancer. n engl jmed ; : ^ . .matheg,reizensteinp.phaseipharmacologicstudy of a newvinca alkaloid: navelbine. cancer lett ; : ^ . . fields-jones s, koletskya,wilding g, et al. improve- ments in clinical benefit with vinorelbine in the treat- ment of hormone-refractory prostate cancer: a phase ii trial. ann oncol ; : ^ . . sweeney cj, monaco fj, jung sh, et al. a phase ii hoosier oncology group study of vinorelbine and estramustine phosphate in hormone-refractory pros- tate cancer. ann oncol ; : ^ . .bisserymc,vrignaudp,lavellef.preclinicalprofileof docetaxel (taxotere): efficacy as a single agent and in combination. semin oncol ; suppl : ^ . . aoe k, kiura k, ueoka h, et al. effect of docetaxel with cisplatin or vinorelbine on lung cancer cell lines. anticancer res ; : ^ . . kreisw, budman dr, calabro a. unique synergism or antagonism of combinations of chemotherapeutic and hormonal agents in human prostate cancer cell lines. brjurol ; : ^ . . black aj, mcleod hl, capell ha, et al.thiopurine methyltransferase genotype predicts therapy-limiting severe toxicity from azathioprine. ann intern med ; : ^ . . milanog,mcleodhl.candihydropyrimidinedehy- drogenase impact -fluorouracil-based treatment? eurjcancer ; : ^ . . innocenti f, undevia sd, iyer l, et al. genetic variants in the udp-glucuronosyltransferase a gene predict the risk of severe neutropenia of irino- tecan. j clin oncol ; : ^ . . tayeb mt, clark c, ameyaw mm, et al. cyp a promoter variant in saudi, ghanaian, and scottish caucasian populations. pharmacogenetics ; : ^ . . kuehl p, zhangj, liny, et al. sequence diversity in cyp a promoters and characterization of the genetic basis of polymorphic cyp a expression. nature gene ; : ^ . . ameyawmm,regateirof,lit,etal.mdr pharma- cogenetics: frequency of the c t mutation in exon is significantly influenced by ethnicity. phar- macogenetics ; : ^ . . sangrajrang s, denouletp, millot g, et al.estramus- tineresistancecorrelateswithh over-expressioninhu- man prostatic carcinoma cells. int j cancer ; : ^ . . doyle la, ross dd. multidrug resistance mediated research. on april , . © american association for cancerclincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ docetaxel,vinorelbine, and estramustine phase ii study www.aacrjournals.org clin cancer res ; ( ) october , by thebreastcancer resistanceproteinbcrp(abcg ). oncogene ; : ^ . . bubley gj, carducci m, dahut w, et al. eligibility and response guidelines for phase ii clinical trials in androgen-independent prostate cancer: recommen- dations from the prostate-specific antigenworking group. jclin oncol ; : ^ . . garsa aa, mcleod hl, marsh s. cyp a and cyp a genotyping by pyrosequencing. bmc med genet ; : . . mathijssenrh,marshs,karlssonmo,etal.irinote- can pathway genotype analysis to predictpharmaco- kinetics. clin cancer res ; : ^ . . sparreboom a, gelderblom h, marsh s, et al. diflomotecan pharmacokinetics in relation to abcg c>a genotype. clin pharmacol ther ; : ^ . . marsh s, king cr, garsa aa, mcleod hl.pyrose- quencing of clinically relevant polymorphisms. meth- ods mol biol ; : ^ . . petrylakdp, macarthur rb, o’connorj, et al. phase i trial of docetaxel with estramustine in androgen- independent prostate cancer. j clin oncol ; : ^ . . kreis w, budman dr, fetten j, gonzales al, barile b,vinciguerra v. phase i trial of the combina- tion of daily estramustine phosphate and intermit- tent docetaxel in patients with metastatic hormone refractory prostate carcinoma. ann oncol ; : ^ . . savarese dm, halabi s, harsv, et al. phase ii study of docetaxel, estramustine, and low-dose hydrocorti- soneinmenwithhormone-refractoryprostatecancer: a final report of calgb . cancer and leukemia group b. jclin oncol ; : ^ . . sinibaldi vj, carducci ma, moore-cooper s, laufer m, sahurak m, eisenberger ma. phase ii eval- uation of docetaxel plus one-day oral estramustine phosphate in the treatment of patients with androgen independent prostate carcinoma. cancer ; : ^ . . budman dr, calabro a, kreissw. synergistic and antagonistic combinations of drugs inhumanprostate cancer cell lines in vitro. anticancer drugs ; : ^ . . oudards, catya, humblety,etal. phase iistudyof vinorelbine in patients with androgen-independent prostate cancer. ann oncol ; : ^ . . morant r, hsu schmitz sf, bernhardj, et al.vinor- elbine in androgen-independent metastatic prostatic carcinomala phase ii study. eurj cancer ; : ^ . .tralongo p, bollina r, aiello r, et al.vinorelbine and prednisone in older cancer patients with hormone- refractorymetastaticprostatecancer.aphaseiistudy. tumori ; : ^ . . robles c, furst aj, sriratana p, et al. phase ii study of vinorelbine with low dose prednisone in the treat- ment of hormone-refractory metastatic prostate can- cer. oncol rep ; : ^ . . koletskyaj,guerraml,kronishl.phaseiistudyof vinorelbine andlow-dose docetaxel inchemotherapy- nai«ve patients with hormone-refractory prostate can- cer. cancerj ; : ^ . . johnstone,crawfordj,garstj,etal.phaseitrialof weeklydocetaxel(d)andvinorelbine(v) inadvanced non-small cell lung cancer (nsclc) [abstract ]. proc am soc clin oncol ; : . .vanzuylenl,verweijj,nooterk,brouwere,stoter g, sparreboom a. role of intestinal p-glycoprotein in the plasma and fecal disposition of docetaxel in humans. clin cancer res ; : ^ . research. on april , . © american association for cancerclincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/ ; : - . clin cancer res noah m. hahn, sharon marsh, william fisher, et al. survival analysis hormone-refractory prostate cancer with pharmacogenetic docetaxel, vinorelbine, and estramustine in combination in hoosier oncology group randomized phase ii study of updated version http://clincancerres.aacrjournals.org/content/ / / access the most recent version of this article at: cited articles http://clincancerres.aacrjournals.org/content/ / / .full#ref-list- this article cites articles, of which you can access for free at: citing articles http://clincancerres.aacrjournals.org/content/ / / .full#related-urls this article has been cited by highwire-hosted articles. access the articles at: e-mail alerts related to this article or journal.sign up to receive free email-alerts subscriptions reprints and .pubs@aacr.orgdepartment at to order reprints of this article or to subscribe to the journal, contact the aacr publications permissions rightslink site. (ccc) click on "request permissions" which will take you to the copyright clearance center's .http://clincancerres.aacrjournals.org/content/ / / to request permission to re-use all or part of this article, use this link research. on april , . © american association for cancerclincancerres.aacrjournals.org downloaded from http://clincancerres.aacrjournals.org/content/ / / http://clincancerres.aacrjournals.org/content/ / / .full#ref-list- http://clincancerres.aacrjournals.org/content/ / / .full#related-urls http://clincancerres.aacrjournals.org/cgi/alerts mailto:pubs@aacr.org http://clincancerres.aacrjournals.org/content/ / / http://clincancerres.aacrjournals.org/ northern research station 	 	u.s. forest service 	 	 united states department of agriculture 	 forest service 	 northern research station 	 	 forest service home 	 about the agency 	 contact the national office toggle navigation northern research station go 	research programs 	publications & data 	tools & applications 	locations 	scientists & staff 	about nrs 	partnerships 	nrs news contact information northern research station one gifford pinchot drive madison, wi ( ) - ( ) - tty/td contact us popular topics 	research publications 	climate change 	fire 	invasive species 	forest inventory & analysis 	urban field stations 	environmental literacy 	science of the seasons 	 	experimental forests 	research natural areas 	 forest products laboratory 	forest service research 	 	 stay connected 		 you are here: 	nrs home northern research station 	 	 	 	 	 	 featured scientist todd hutchinson thinks about forests past and present 						 			 featured research 						when a timber demonstration project became an impromptu class on fire ecology 					 						 featured partnership if a tree falls in a forest, and someone is around, should it be removed? previous next april feature - earth, wind, and fire last month was all about water, and in april we focus on the other three elements: earth, wind and fire. meet a scientist who is fascinated by the traces of past forests that remain on the landscape; learn about a study that delivered an unexpected lesson on fire ecology, and see why a partnership is recommending that some trees uprooted by wind should be left on the forest floor. 	 		 rooted in research 	 leaving it messy: using tip-up mounds to promote tree species diversity 	 	 when a tree falls in the woods, should it be removed? in managed forests, common practices call for cleaning up and removing timber following a blowdown. but when northern research station research forester christel kern sees the root mass of a fallen tree, she's more likely to see it as an opportunity. her work is described in a forest ecology and management article entitled mounds facilitate regeneration of light-seeded and browse-sensitive tree species after moderate-severity wind disturbance. 	 forestcast 	 our new podcast brings you stories, interviews, and special in-depth anthologies of the science that's examining and explaining how forests affect our lives, and how we affect our forests. 	 research highlights view all of our research highlights 	 	 	latest publications 	blog posts 		 loading publications... 		voices of maple nation - usda forest service tribal relations blogmarch , 	why the trees outside forests count - usda blogmarch , 	after a blight, the trees that survived need your help - usda blogfebruary , 	seeing the urban forest for the trees - taylor quality guitarsjanuary , 	a walk in the woods - years later - short stories about long term researchnovember , 	trees can do the dirty work of waste cleanup - usda blogaugust , 	podcast: london plane trees: more than just the hostas of street trees? - urban wildlife podcastjune , 	news releases 	nrs in the news 			decline in black cherry regeneration may herald wider forest changemarch , 	first midwest urban long-term ecological research site established in twin-citiesmarch , 	invasion by non-native insects expected to increase percent worldwide by october , 	in the arctic, extreme air pollution kills trees, limits growth by reducing sunlightseptember , 	tree planting has potential to increase carbon sequestration capacity on the nation's forestsseptember , 	a small bird's return to missouri stems from forest restoration collaborationseptember , 	more news releases... 		church creates local oasis from dying urban forest - washington postmarch , 	‘a bit of a haven’: southwest baltimore church reclaims a dying urban forest and creates a community oasis - baltimore sunmarch , 	baltimore church converts neglected urban forest into ‘peace park’ - chesapeake bay journalmarch , 	invasive insects and diseases are killing our forests - new york timesfebruary , 	movement turns abandoned housing into beautiful furniture, fighting climate change and growing jobs in the process - cbs this morning - saturdaydecember , 	a warming planet makes northeastern forests more susceptible to western-style wildfires - inside climate newsnovember , 	wisconsin northwoods: wildfire reflections & sustainable management - sustain danenovember , 	 science themes & programs 	 	 	 	 	 	 		forest disturbance processes 		climate change, invasive species, fire, forest health 	 	 	 	 	 		urban natural resource stewardship urban forestry, urban field stations, environmental justice, land use change 	 	 	 	 	 		natural resource inventory, monitoring & assessment forestry inventory, monitoring & assessment, techniques 	 	 	 	 	 		sustaining forests 		forest management, biodiversity, ecology, bioenergy, wildlife & fish 	 	 	 	 	 		providing clean air & water 		riparian forests, clean water,monitoring carbon, bioenergy 	 	 	 	 	 		experimental forests 		 long-term research, locations, national network 	 get connected 	 	 tweets by usfs_nrs visit our twitter feed u.s. forest service - research and development last modified: sunday, april , nrs at a glance research programs 	themes 	units 	lands 	facilities 	special programs publications & tools 	search 	research highlights 	data 	software 	online tools locations 	location directory 	nrs in my state 	experimental forests 	research natural areas scientists & staff 	contact us 	research units 	scientist directory 	all employees about nrs 	staff directory 	scientist directory 	research unit directory 	locations 	nrs in my state 	research highlights nrs news 	news releases 	research review usda.gov | policies & links | our performance | report fraud on usda contracts | visit oig | plain writing | get adobe reader foia | accessibility statement | privacy policy | non-discrimination statement | information quality | usda recovery | usa.gov | whitehouse.gov caballero, ricardo j., “on the sign of the investment-uncertainty relation- ship,” american economic review : (march ), – . “macroeconomic volatility in latin america: a view and three case studies,” nber working paper ( ). caballero, ricardo j., and robert s. pindyck, “uncertainty, investment, and industry evolution,” international economic review : (august ). carruth, alan, andy dickerson, and andrew henley, “what do we know about investment under uncertainty?” journal of economic sur- veys ( ), – . darby, julia, andrew h. hallett, jonathan ireland, and laura piscitelli, “the impact of exchange rate uncertainty on the level of investment,” economic journal ( ), c –c . dixit, avinash, and robert s. pindyck, investment under uncertainty (princeton, nj: princeton university press, ). easterly, william, roumeen islam, and joseph stiglitz, “explaining growth volatility,” in annual world bank conference on devel- opment economics (oxford university press, ). frankel, jeffrey, and andy rose, “a panel project on purchasing power parity: mean reversion within and between countries,” journal of international economics ( ), – . goldberg, linda, “exchange rates and investment in united states industry,” review of economics and statistics ( ), – . gul, faruk, “a theory of disappointment aversion,” econometrica ( ), – . hartman, richard, “the effects of price and cost uncertainty on in- vestment,” journal of economic theory (october ), – . interamerican development bank, overcoming volatility in latin america ( ). kraay, aart, norman loayza, luis servén, and jaume ventura, “country portfolios,” nber working paper ( ). lee, jaewoo, and kwanho shin, “the role of a variable input in the relationship between investment and uncertainty,” american eco- nomic review ( ), – . levine, ross, norman loayza, and thorsten beck, “financial intermedi- ation and growth: causality and causes,” world bank policy research working paper ( ). lothian, james, and mark taylor, “real exchange rate behavior: the recent float from the perspective of the last two centuries,” journal of political economy ( ), – . pindyck, robert s., and andrés solimano, “economic instability and aggregate investment,” nber macroeconomics annual ( ), – . sarkar, sudipto, “on the investment-uncertainty relationship in a real options model,” journal of economic dynamics and control ( ), – . servén, luis, “macroeconomic uncertainty and private investment in ldcs: an empirical investigation,” world bank policy research working paper ( ). “terms of trade shocks and optimal investment: another look at the laursen-metzler effect,” journal of international money and finance ( ), – . servén, luis and andrés solimano, striving for growth after adjustment: the role of capital formation (washington: the world bank, ). zeira, joseph, “cost uncertainty and the rate of investment,” journal of economic dynamics and control ( ), – . assessing individual risk attitudes using field data from lottery games connel fullenkamp, rafael tenorio, and robert battalio* abstract—we use information from the television game show with the highest guaranteed average payoff in the united states, hoosier million- aire, to analyze risktaking in a high-stakes experiment. we characterize gambling decisions under alternative assumptions about contestant behav- ior and preferences, and derive testable restrictions on individual risk attitudes based on this characterization. we then use an extensive sample of gambling decisions to estimate distributions of risk-aversion parameters consistent with the theoretical restrictions and revealed preferences. we find that although most contestants display risk-averse preferences, the extent of the risk aversion implied by our estimates varies substantially with the stakes involved in the different decisions. i. introduction a series of recent studies (rabin, a, b; rabinand thaler, ) have pointed out the inadequacy of expected-utility theory to provide proper characterization of risk aversion when monetary stakes are small. the basic contention of this criticism is that the concavity of the utility function would have to be so pronounced to explain small- stake risk aversion that this same utility function would imply absurd levels of risk aversion for large stakes. rabin ( b) concludes that “expected-utility theory seems to be a useful and adequate model of risk aversion for many purposes, such as understanding large-stakes insur- ance . . . ,” but “is manifestly not close to the right expla- nation of risk attitudes over modest stakes . . . .” a number of studies have attempted to characterize indi- vidual risk aversion under large stakes using expected- utility theory. experiments by binswanger ( ) and kachelmeier and shehata ( ) face subjects with deci- sions involving small nominal stakes that are large relative to the subjects’ wealth. their findings are mixed: whereas binswanger’s subjects exhibit more risk-averse behavior as stakes are increased, kachelmeier and shehata conclude that “the effects of monetary payoffs are real, albeit subtle, and are in need of further study.” unfortunately, the inherent limitations of rewards in experimental studies make it dif- ficult to get more robust conclusions about risktaking in large-stakes settings. a second group of papers analyzes risktaking behavior using information from natural experiments involving large received for publication june , . revision accepted for publica- tion january , . * duke university, depaul university, and university of notre dame, respectively. we thank gabriella bucci, larry samuelson, richard sheehan, various workshop participants, and two anonymous referees for numerous com- ments. part of this project was completed while tenorio visited the management and strategy department at northwestern university. he is thankful for their hospitality. trish and marcos delgado provided superb research assistance. all errors or omissions are the authors’. the review of economics and statistics the review of economics and statistics, february , ( ): – © by the president and fellows of harvard college and the massachusetts institute of technology stakes. gertner ( ) uses betting decisions from the bonus round of the television game show card sharks, and con- cludes that individuals exhibit a small degree of risk aver- sion. metrick ( ) and hersch and mcdougall ( ) analyze field data from wagering decisions on the television game shows final jeopardy! and illinois instant riches, respectively, and find that contestants display behavior that is statistically indistinguishable from risk neutrality. in con- trast, beetsma and schotman ( ) conclude that contes- tants in the dutch television show lingo exhibit substan- tial risk aversion. in this paper we use information from the television show hoosier millionaire to gain additional insight into the be- havior of individuals in high-stakes situations. although this show has undergone a number of regime changes, the current regime is representative of the typical decision a contestant faces: she may either (a) take a sure $ , , or (b) play a game in which she gets $ , , $ , , , , , or $ , each with equal probability. if she draws $ , or $ , , she may stop playing and keep the prize, or continue drawing from the remaining alternatives until she either decides to stop, or draws $ or $ , , . prizes are not cumulative. we believe, for various reasons, that the structure of the decision problems in hoosier millionaire provides us with a very appealing natural experiment to study individual decisions under risk. first, unlike previous game studies, the realization of the risky prospect is the outcome of a very simple probability process, where no subjectivity or skills are involved. second, the stakes are larger than those in previous studies, thus allowing a more proper characteriza- tion of risk aversion using expected-utility theory. finally, our data comprise a time series of games involving several regime changes, which allows us to study the sensitivity of individual behavior across various gambles. our approach to analyzing risk aversion differs signifi- cantly from previous game-based studies. exogenous wa- gers and lack of variation in the gambles prevent us from estimating risk-aversion parameters using regression tech- niques. instead, we use the sample properties of the deci- sions in the different regimes of hoosier millionaire to estimate the parameters of the distribution of risk-aversion coefficients in the contestant population. in addition, unlike previous studies, we explicitly allow for various degrees of contestant rationality in our estimation. our results show that most individuals behave in a way consistent with risk aversion, and as noted in recent studies, the implied degree of risk aversion varies largely with the stakes of the decision. in fact, our estimates imply substan- tial risk aversion for gambles comparable to those in hoo- sier millionaire, while also implying near-risk-neutrality for smaller gambles of similar structure. we also show that the effect of different behavioral traits on decisions and risk aversion is only evident under large stakes. ii. a brief history of hoosier millionaire the hoosier lottery’s weekly television show, hoosier millionaire, first aired on october , . since its inception, each show has involved six contestants chosen randomly from a pool of entries submitted by people playing a scratch-off ticket game. all contestants participate in the first of two phases played on each show. in phase i, all six contestants play a series of purely random draw games. the contestant who amasses the most cash wins and proceeds to phase ii; the remaining five contestants leave the show. all six contestants keep the cash and prizes accumulated in phase i. four regimes have governed phase ii of hoosier million- aire. in regime , the phase i winner has the option of choosing one of four numbers randomly associated with the dollar amounts $ , , $ , , and $ , , , and a consolation prize of $ , . if the contestant draws $ million, she wins that amount, paid out in equal annual payments, and the game is over. if the contestant draws $ , or $ , , she can keep that amount or give it up and make another choice among the remaining alternatives. she continues until she chooses to stop, wins the $ million, or draws the consolation prize. in october of , the two intermediate amounts offered in the gamble changed from $ , and $ , to $ , and $ , . all other features remained the same. we call this regime . in february of , more changes were made to phase ii. first, the lowest amount attainable in the initial drawing changed from $ , to $ , . second, the consolation prize went down from $ , to , scratch-off lottery tickets with a purchase price of $ each. finally, lottery officials changed the payment of the $ million prize to twenty annual payments of $ , . we call this regime . rules currently governing phase ii of hoosier millionaire were instituted in february of . according to pat traub, the lottery’s acting deputy director, these changes offer “. . . the highest guaranteed prize of any game show in the nation . . .” and were made to “. . . enhance the show’s entertainment value.” two revisions were made. first, a contestant is now automatically endowed with a guaranteed $ , prize. she may then decide to keep that amount and walk away from the show, or give it up and choose one of the four numbers just as in regime . the other change involved extending the period of time over which the $ , , prize is paid out from twenty to twenty-five years. we call this regime . both card sharks and illinois instant riches involve nontrivial prob- ability calculations, and final jeopardy and lingo involve subjective assessments of one’s (or other players’) ability to solve a puzzle or answer a question. this change, not widely publicized by the lottery commission, be- came a heated point of contention in the indiana state elections. notes iii. analytical issues in analytical terms, the problem a hoosier millionaire contestant faces is one of sequential decisions with no recall. this means that the outcome of the first decision may give the contestant the option to continue playing, but once a continuation decision is made at any stage, the status quo is no longer available. in a problem like this, the optimal decisions follow from using a dynamic programming (or backward induction) technique. this means starting at the final decision node, assessing the values of continuing and stopping at that stage, and then factoring in these values when making decisions at the previous stage(s). this pro- cedure is equivalent to finding the subgame-perfect equilib- rium in a game against nature. in analyzing the optimal strategy for each regime, we consider the following behavioral hypotheses: full rationality (fr). a contestant performs backward induction at each decision node, and discounts the $ , , prize according to the annuity system. bounded rationality (br ). same as fr, but no dis- counting is applied to the $ , , prize. bounded rationality (br ). contestants do not per- form backward induction at every decision node, that is, they do not take into account the value of the option to continue playing the game, and evaluate each lottery as a simple rather than a sequentially compounded lottery. an- nuities are discounted. bounded rationality (br ). same as br , but annu- ities are not discounted. our bounded-rationality scenarios are motivated by bi- ases observed in decision settings comparable to our prob- lem. empirical and experimental evidence suggests that low or even negative discounting is common in decisions in- volving evaluation of income streams over time (br , br ). loewenstein and prelec ( ), loewenstein and thaler ( ), and gigliotti and sopher ( ) note that most subjects prefer a more constant and spread-out stream of income over a strictly decreasing pattern of payments. , there is also extensive evidence that individuals fail to perform backward induction in sequential decision envi- ronments (br , br ). carbone and hey ( ) report a series of decision-making experiments where individuals do not apply bellman’s principle of optimality. similarly, camerer et al. ( ) show that subjects do not reason backward in simple alternating-offers bargaining experi- ments. instead, most subjects base their decision on current-round payoffs. figures and show the game trees associated with regimes and of hoosier millionaire. we subsequently analyze optimal decisions under each of the behavioral hypotheses outlined above. a. risk neutrality if decisions depend solely on expected values at each decision node, simple calculations show that: . boundedly rational contestants who do not discount annuities (br and br ) always choose the gamble over the sure prospect. here, the $ , , prize makes the gambles’ expected values very large across the board, even when contestants do not backward- induct. . if contestants discount annuities (fr and br ), gambling may become unattractive as discounting increases. table shows the discount rates that make a contestant indifferent between gambling and keeping the sure prospect at the various deci- sion nodes. as shown, for sufficiently low discount rates (r � . %), it is optimal to gamble in any regime at any decision node for both the fr and br contestant types. as discounting increases, some of the gambles’ expected values fall below the sure prospects. in regimes and , discount rates must be very high (above %) to discourage indi- viduals from gambling. this is because the $ million annuity payments are only spread across years and the consolation prize is substantial ($ , ). in regimes and , where annuities are paid over – years and the consolation prize is only $ , , lower discount rates may induce indi- viduals to take on various gambles. all earnings are taxable. given the amounts at stake, as long as marginal tax rates are constant over the annuity payments’ horizon, decisions should be tax-neutral. we also assume a constant inflation rate over the annuity horizon. thus our discount rates may be interpreted as real discount rates. a further possibility is that contestants are unaware of the annuity system, or that their decision frames are affected by the fact that $ million winners are presented with a large symbolic check for that amount. a variety of other decision problems involving time-delayed payoffs, but not streams of payoffs, actually show that individuals may overdis- count future payoffs. however, financial companies openly advertise their readiness to convert lottery annuity payments into lump-sum payoffs at discount rates in the %– % range. thus, overdiscounting appears unlikely in our problem. the authors conducted an experiment to gain insight into this issue. the hoosier millionaire regime decision problem was given as a final exam question in two different courses: first-year mba microeconomics and senior-level risk management and insurance. whereas the mba students had been exposed to the concept of backward induction, seniors were mostly unfamiliar with it. our results show that ( . %) of mba students and ( . %) of seniors used backward induction to solve this problem. the review of economics and statistics b. risk aversion to make our results comparable with those of previous studies, we assume that hoosier millionaire contestants display one of the following two types of preferences: constant absolute risk aversion (cara). these prefer- ences, which imply increasing relative risk aversion, are often assumed in studies of individual decision-making. their generic utility representation is u�w� � �e�aw, ( ) where w is the individual’s wealth, a is the coefficient of absolute risk aversion, and aw is the coefficient of relative risk aversion. this formulation is convenient because it allows one to calculate absolute risk-aversion coefficients without any wealth information. constant relative risk aversion (crra). these prefer- ences, which imply decreasing absolute risk aversion, are commonly used in macro and asset-pricing studies. their generic utility representation is u�w� � w �b � b , ( ) where w is the individual’s wealth, b is the coefficient of relative risk aversion, and b/w is the coefficient of absolute risk aversion. since this formulation implies that risk aver- sion depends on wealth, we must know something about that variable to make meaningful inferences. given lack of comprehensive wealth information, we make alternative it is well known that expected-utility theory fails to account for some empirical regularities in decision-making under uncertainty. the most impor- tant two violations relate to the decision-makers’ (a) asymmetric evaluations of gains and losses, and (b) use of decision weights instead of probabilities. the decisions we analyze are such that (a) is absent because there are no losses, and the influence of (b) should be minimal due to the simple probability structure of the gambles. in addition, the recent papers by rabin ( a, b), and rabin and thaler ( ) conclude that risk-aversion results based on expected-utility theory are most accurate when they pertain to decisions involving large stakes, such as those in our games. thus, we feel comfortable staying within the expected-utility framework. figure .—regime ( ) n stands for nature; ① and ② represent relevant subgames. ( ) all payments are lump sum except for $ , , , which is paid as a -year annuity due. figure .—regime ( ) n stands for nature; ① , ② , and ③ represent relevant subgames. ( ) all payments are lump sum except for $ , , , which is paid as a -year annuity. table .—discount rates that make a risk-neutral contestant indifferent between taking and not taking the gamble regime rate (%) fully rational contestants in subgame boundedly rational- contestants in subgame fully rational contestants in subgame fully rational contestants in subgame . 	 . 	 . . n.a. . . . 	 . n.a. . . 	 . . n.a. . . . . n.a. . . . . n.a. . . . . n.a. . . . n.a. n.a. . n.a. n.a. . . . n.a. n.a. . . each regime is indexed by the value of the contestant’s first draw. a . (. ) extension denotes an initial draw equal to the higher (lower) of the two intermediate dollar prizes. fully rational contestants perform backward induction at each decision node of the tree, whereas boundedly-rational- contestants do not perform backward induction (that is, they ignore the option value of continuing to play the game). both fully rational and boundedly-rational- contestants fully discount annuity payments. n.a. � nonappli- cable. notes assumptions about the reference wealth level that individu- als use to make decisions. we first use a restrictive defini- tion of wealth, w , encompassing only the winnings accu- mulated in hoosier millionaire. this partial asset integration in decision-making is to some extent consistent with kahneman and tversky’s ( ) prospect theory. as a test of robustness we will also use a broader wealth defini- tion (w ), which includes the median household income of the individual’s census tract in . table shows the values for the cara and crra (w ) parameters that make contestants indifferent between gam- bling and keeping the sure prospect on every initial sub- game in every regime of hoosier millionaire. that is, for any risk-aversion parameter larger than a given entry in this table, the contestant will prefer the sure prospect to the gamble. as expected, as discounting increases, the value of the $ , , annuity payments becomes small relative to that of the lump-sum payoffs. thus, at some discount rate the expected value of a gamble becomes smaller than the sure prospect, and the contestant would have to be a risk lover to gamble (see the appropriate entries in table when r � % and r � %). also, for a given discount rate, the risk-aversion coefficient that dissuades a contestant from gambling decreases with the amount of the sure prospect, regardless of the degree of contestant rationality. this fol- lows from certainty equivalence. iv. data we obtained the hoosier millionaire data from the hoo- sier lottery office in indianapolis, and from press releases in the south bend tribune. with a few exceptions, our data set contains the winnings and census tract characteristics of all the participants in both phases of hoosier millionaire since its inception. table presents a summary of the contestants’ endoge- nous initial gambling decisions. we also report mean phase i winnings and census tract median income figures for our sample. as seen in table , although % of the individuals in our sample take the initial gamble, this fraction varies across regimes. in fact, the rough compara- tive statics of decisions across regimes indicate that contes- tants respond correctly to changes in the award structure. for instance, in regime the sizable consolation prize ($ , ) substantially reduces the downside of gambling. as a result, all of the contestants in this regime choose to gamble. this contrasts with regime , where the small consolation prize ($ , ) makes the downside of gambling we concentrate on the initial gambling decisions in view of the very limited number of individuals choosing to gamble more than once. in regimes – contestants face a compulsory (exogenous) initial gambling decision. thus we consider only on the (endogenous) decisions after that compulsory stage. in contrast, all initial gambling decisions in regime are endogenous. table .—risk-aversion parameters that yield equal expected utilities from taking and not taking the gamble panel a: constant absolute risk aversion (cara) utility function regime fr br br br r � % r � % r � % r � % r � % r � % r � % r � % . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . � . . . . . � . e� . . � . e� � . . . . � . � . . . � . e� � . . . . . � . . . . e� � . e� . panel b: constant relative risk aversion (crra) utility function regime initial wealth fr br br br r � % r � % r � % r � % r � % r � % r � % r � % . $ , . . . . . . . . . $ , . . . . . . . . . $ , . . . . . . . . . $ , . . . . . . � . . . $ , . � . � . . . . � . . . $ , . � . �� . . � . �� . . $ , . . � . . . . . . each regime is indexed by the value of the contestant’s first draw. a . (. ) extension denotes an initial draw equal to the higher (lower) of the two intermediate dollar prizes. fully rational (fr) contestants perform backward induction at each node of the tree, and discount annuity payments. each contestant may use a different discount rate. we use discount rates of %, %, and %. thus, fr, r � %, refers to a fully rational contestant who discounts future cash flows using a % rate. boundedly-rational- (br ) contestants are the same as the fr contestants with the exception that they use a discount rate of (i.e., they do not discount future payments). though boundedly rational- (br ) contestants discount future cash flows, they act myopically in that they ignore the option value of continuing to play the game. boundedly-rational- (br ) contestants ignore the option value of continuing to play the game, and they use a discount rate of . the cara utility function is specified as follows: u(w) � �e�aw, where w is the individual’s initial wealth and a is the risk-aversion parameter. the crra utility function is specified as follows: u(w) � w �b/( � b), where w is the individual’s initial wealth and b is the risk-aversion parameter. initial wealth refers to the median amount of winnings in phase of the hoosier millionaire. the review of economics and statistics rather onerous, thus inducing of contestants to keep the sure prospect. the fourth column of table shows phase winnings for each contestant in our data set. as seen, there is not much variation in these winnings. we obtained the addresses of all phase hoosier millionaire contestants, and found match- ing data from their census tracts. table also shows the average median family and per capita income for our sample. although both of these income measures are below state averages, they are not significantly different from them. however, there is no systematic difference between the incomes of contestants that took the gambles and con- testants that chose the certain prospects. v. estimation a. methodology previous studies such as those of gertner ( ), metrick ( ), and hersch and mcdougall ( ) estimated the coefficient of absolute risk aversion using a nonlinear probit approach. unlike those studies, where individuals face gam- bles involving different stakes, each contestant within a regime of hoosier millionaire faces a gamble with identical stakes. thus the probit technique is inappropriate, because a key source of variation—that between the sizes of the gambles—is absent. instead, we use a probabilistic approach and estimate the distribution of the risk-aversion parameter. we assume that contestants are endowed with risk-aversion parameters that are independent draws from a normal population distribu- tion. if a contestant’s realization of the risk-aversion pa- rameter is less than the value that yields indifference (equal- ity of expected utilities from gambling and not gambling), she chooses the gamble. the probability that the contes- tant’s risk-aversion parameter lies below the indifference value is given by the value of the cumulative normal distribution evaluated at the indifference value. as we know, the normal distribution is determined by its mean and standard deviation. to estimate these parameters, we use the indifference values in table and the binomial probabilities implied by the empirical frequencies in table . we infer the mean and standard deviation as follows: each subgame in the history of hoosier millionaire is associated with a cara or crra indifference parameter, which we denote , and an observed frequency of taking the gamble, denoted p. we choose pairs of subgames, and solve for the (unique) values of the mean � and standard deviation � that satisfy the following system: � �
 � �� �� / e����x� / � dx � p , ( a) � �
 � �� �� / e����x� / � dx � p . ( b) we then used monte carlo simulations to place confidence intervals around the estimated mean and standard deviation. for each monte carlo trial, we generated two data sets consisting of indicator random variables, and used the implied p’s from these pseudo data to solve for � and �. the data-generating process for each set of indicator variables was a binomial distribution with parameter (probability of success on each binomial trial) equal to the observed p from the respective subgame. the number of observations in each set of pseudo data was set equal to the number of actual observations of the respective subgame. we repeated this experiment , times for each regime pair and behavioral hersch and mcdougall ( ) note that the median income of illinois lottery players is nearly identical to the statewide figure. although no figures are presented to back this claim, its accuracy is subject to the same problem regarding the coarseness of variables within census tracts. we also collected other census tract data for our contestant sample, like schooling, household size, and age. we do not show this information, due to its coarseness and insignificant interregime variation. although we choose a normal distribution for our analysis, the results are qualitatively similar if we use a one-parameter symmetric distribution, such as the logistic distribution. we do not know ex ante if risk-aversion parameters are symmetrically distributed among individuals, but we are unaware of hard evidence showing otherwise. table .—descriptive statistics regime take gamble? sample size phase winnings median income per capita income . no n.a. n.a. n.a. yes $ , $ , $ , total $ , $ , $ , . no n.a. n.a. n.a. yes $ , $ , $ , total $ , $ , $ , . no $ , missing obs. missing obs. yes $ , $ , $ , total $ , $ , $ , . no $ , $ , $ , yes $ , $ , $ , total $ , $ , $ , . no $ , $ , $ , yes $ , missing obs. missing obs. total $ , $ , $ , . no $ , $ , $ , yes $ , missing obs. missing obs. total $ , $ , $ , . no $ , $ , $ , yes $ , $ , $ , total $ , $ , $ , total $ , $ , $ , indiana $ , $ , each regime is indexed by the value of the contestant’s first draw. a . (. ) extension denotes an initial draw equal to the higher (lower) of the two intermediate dollar prizes. phase refers to the preliminary phase of hoosier millionaire, during which contestants randomly draw cash prizes. the contestant amassing the most cash in phase moves on to phase , where she is confronted with the gambles analyzed in this paper. income figures are from the census. median and per capita incomes are not available for all contestants, due to lack of demographic information. this affects seven contestants in regime , two in regime , and two in regime . “missing obs.” denotes a cell in which demographic information is missing for at least one contestant. n.a. � nonapplicable. notes assumption, ranked the estimated means and standard devi- ations, and identified % confidence intervals. although we performed this experiment using several pairs of subgames, we focus on the pairing of regimes and . . table reports estimated means and variances of the cara and crra coefficients for each behavioral assumption, along with the monte carlo-generated % confidence intervals. b. results a feature of our results is their robustness across utility specifications. the parameter estimates for the cara and crra (w ) functions yield similar distributions and exhibit comparable patterns. the following discussion makes use of this similarity by grouping the results from both specifica- tions whenever possible. first, our results consistently support risk aversion. esti- mated mean risk-aversion coefficients are always positive, and the confidence intervals around these means never include zero. mean risk-aversion parameter estimates under cara range from . e� to . e� , and those under crra range from . to . . moreover, the minimum means implied by the monte carlo simulations are always positive. this evidence supports the hypothesis that on average, the individuals on our sample are risk-averse. however, the estimated standard deviations of the risk- aversion parameters indicate that some individuals may display risk-neutral or risk-loving preferences. this is par- ticularly the case for crra utility, where a significant fraction of estimated risk-aversion parameters are nonposi- tive. a two-standard-deviation interval around the mean risk-aversion parameter generally includes only positive values for the cara utility function but encompasses zero and negative values for the crra one. our second main result is that different behavioral as- sumptions affect the estimates of mean risk aversion. for strictly speaking, the distribution parameters are overidentified, as there are several regime pairs that could be used to estimate them. however, the choice of regime pairs does not introduce major qualitative variations in the estimates. the similar results we obtained in the logistic case, which is not subject to the pairing problem, reinforce this point. table .—estimates of risk-aversion coefficients utility-function specification coefficient at rationality assumption fr r � % br r � % br r � % br r � % cara mean . . . . ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) std. dev. . . . . ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) crra mean . . . . w � $ , ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) std. dev. . . . . ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) crra mean . . . . w � $ , std. dev. . . . . estimates of risk-aversion coefficients are obtained by using maximum likelihood and observations from regimes . and . to estimate the system of equations described in equations ( a)–( b) of the text. each regime is indexed by the value of the contestant’s first draw. a . extension denotes an initial draw equal to the lower of the two intermediate dollar prizes. fully rational (fr) contestants perform backward induction at each node of the tree, and discount annuity payments. each contestant may use a different discount rate. thus, fr, r � %, refers to a fully rational contestant who discounts future cash flows using a % rate. boundedly-rational- (br ) contestants are the same as the fr contestants with the exception that they use a discount rate of (that is, they do not discount future payments). although boundedly-rational- (br ) contestants discount future cash flows, they ignore the option value of continuing to play the game. boundedly-rational- (br ) contestants ignore the option value of continuing to play the game, and they use a discount rate of . the cara utility function is u(w) � �e�aw, where w is the individual’s initial wealth and a is the risk-aversion parameter. the crra utility function is u(w) � w �b/( � b), where w is the individual’s initial wealth and b is the risk aversion parameter. two estimations are done for the crra utility function. the first uses an initial wealth (w ) of $ , , and the second uses an initial wealth (w ) of $ , . standard deviation is abbreviated std. dev., and confidence interval is abbreviated c.i. table .—certainty equivalents for various gambles implied by mean estimates of risk-aversion coefficients utility-function specification rationality assumption certainty equivalents for a - gamble with payoffs of and x x � $ , x � $ , x � $ , x � $ , , cara fr (r � %) $ . $ , . $ , . $ , . br (r � %) $ . $ , . $ , . $ , . br (r � %) $ . $ , . $ , . $ , . br (r � %) $ . $ , . $ , . $ , . crra fr (r � %) $ . $ , . $ , . $ , . w � $ , br (r � %) $ . $ , . $ , . $ , . br (r � %) $ . $ , . $ , . $ , . br (r � %) $ . $ , . $ , . $ , . estimates of risk-aversion coefficients are obtained by using maximum likelihood and observations from regimes . and . to estimate the system of equations described in equations ( a)–( b) of the text. each regime is indexed by the value of the contestant’s first draw. a . extension denotes an initial draw equal to the lower of the two intermediate dollar prizes. fully rational (fr) contestants perform backward induction at each node of the tree and discount annuity payments. each contestant may use a different discount rate. fr (r � %) refers to a fully rational contestant who discounts future cash flows using a % rate. boundedly-rational- (br ) contestants are the same as the fr contestants with the exception that they use a discount rate of (that is, they do not discount future payments). although boundedly-rational- (br ) contestants discount future cash flows, they ignore the option value of continuing to play the game. boundedly-rational- (br ) contestants ignore the option value of continuing to play the game, and they use a discount rate of . the cara utility function is u(w) � �e�aw, where w is the individual’s initial wealth and a is the risk-aversion parameter. the crra utility function is u(w) � w �b/( � b), where w is the individual’s initial wealth and b is the risk-aversion parameter. the review of economics and statistics fixed discounting, confidence intervals for these means under full rationality (fr) and lack of backward induction (br and br ) are mutually exclusive. in addition, when comparing no discounting and % discounting, holding behavior fixed, confidence intervals are mutually exclusive in three of four cases. we discuss the direction and signif- icance of these differences in the next section. as a test of robustness of our crra results, we estimated the distribution parameters using a more inclusive definition of wealth (w ). in this specification, in addition to phase i winnings, the initial wealth also included the average me- dian income for the contestant’s census tract in . as we see in table , the higher initial wealth uniformly increases the means and standard deviations of the crra parameter distributions. however, most of these increases are very small (between . and . ) and, as we show in the next section, do not affect the basic interpretation of the results. the similarity between the two sets of crra estimates is due to the fact that that both wealth definitions are a small fraction of the stakes of the gambles in hoosier millionaire. vi. discussion and conclusions the results from the previous section uniformly indicate that most individuals display some degree of risk aversion. to gain insight into the economic significance of these results, we first compare them with those obtained in related work. our estimated means of the cara parameter distri- butions are lower than the estimated cara parameters of other game-show studies. gertner ( ), using two alter- native methods, estimates statistically significant lower bounds on this parameter of . or . . met- rick ( ) and hersch and mcdougall ( ) report esti- mates ranging from . to . , but these esti- mates are not statistically different from zero, leading to the conclusion of risk neutrality. in contrast, our largest esti- mated mean cara parameter is . , and the largest upper bound on a confidence interval is . , both of which are nearly an order of magnitude smaller than those reported elsewhere. our crra parameter estimates are also generally lower than those reported elsewhere. friend and blume ( ), using data from individual portfolio holdings, find estimates of this parameter ranging from to , while chou, engle, and kane ( ), in an asset-pricing context, estimate it to be around . beetsma and schotman ( ), using game data from lingo, estimate it to be around . in our case the mean crra parameters range from . to . , once again below most estimates from other studies. to gain some intuition into the meaning of our results and how they compare with those of other studies, we calculate certainty equivalents implied by our mean estimates for a variety of gambles (see table ). for instance, for a gamble offering a - chance of winning nothing and $ , , the certainty equivalent ranges from $ . (gertner, ) to $ . (hersch and mcdougall, ), with most esti- mates implying a certainty equivalent above $ . in our experiments, certainty equivalents implied by point esti- mates of the mean of the parameter distribution range from $ . to $ . . thus for gambles of this magnitude, not only are our results similar to those in other studies, but they imply behavior close to risk neutrality. moreover, the tightness of the range of these certainty equivalents indi- cates that the differences among risk aversion parameters across behavioral assumptions may not have much eco- nomic significance when stakes like these are involved. in fact, as seen in table , near-risk-neutrality across all be- haviors obtains even as stakes rise to several thousand dollars. this point, also emphasized on table of rabin and thaler ( ), suggests that gambles of this size unavoid- ably lead to parameter estimates that imply near- risk-neutrality within the expected utility paradigm. thus, given that most previous studies based their estimates on gambles with maximum stakes around $ , , it is not surprising at all that the results point towards risk neutrality or mild risk aversion. however, we estimated our risk-aversion parameters from decisions involving unusually high stakes. in fact, stakes in hoosier millionaire are much higher ( times or more in most cases) than those in any of the games analyzed in other studies. thus, we must base our inferences on stakes of this magnitude. for instance, for a - gamble between winning nothing and $ million, the certainty equivalents that our mean estimates imply range from $ , to $ , , indicating substantial risk aversion. moreover, there is wide variation of certainty equivalents, depending on rationality assumptions. more pre- cisely, fully rational contestants display uniformly higher cer- tainty equivalents than all boundedly rational contestants, and contestants who do not backward-induct and fail to discount their payoffs display the lowest certainty equivalents. these facts suggest that the willingness to take on large gambles is inversely related to the extent of rationality of the contes- tant. because boundedly rational contestants fail to take account of for the option value of continuing to make decisions, their perceived expected utility of gambling is below that of a fully rational contestant. table also reports the ces associated with two intermediate gambles, which show how risk aversion starts to surface more dramatically as the stakes grow larger. the income statistics are from the census ( figures), whereas the gambling figures are from – . although post- contestant census tract figures are not available, the income variation within this period is not significant enough to meaningfully affect our crra estimates or its interpretation. for instance, median household income in indiana was $ , in , indicating only a modest nominal increase of . % relative to . the high risk-aversion parameters estimated by beetsma and schot- man ( ) are an exception. given the relatively small average stakes of their gambles (around $ ), it is possible that the subjective elements involved in their decisions are influencing their estimates. notes what can we learn from these results? first, one should be very careful in drawing general conclusions about pref- erences from context-specific estimates. in our case, when taken in the context of previous studies, our estimates point towards very mild risk aversion. however, once we account for the actual context of our estimation, the results indicate substantial risk aversion. thus, the recent claims in rabin ( a, b) and rabin and thaler ( ) about the limited applicability of expected-utility theory when stakes are small should be taken very seriously. second, not only does the size of the stakes affect risk-aversion inferences, but it is also very important in helping us distinguish across behaviors in decision-making problems. as we showed, estimates based on small stakes may yield decision rules that are economically indistinguishable for fully rational and boundedly rational decision-makers. only at high stakes are these behavioral differences clearly observable. finally, a common criticism of field-study data pertains to the selection of participating subjects. it is possible that indi- viduals participating in these games are in some sense different from the rest of the population. the evidence presented here and also in hersch and mcdougall ( ) suggests that this is not the case with regard to observable characteristics. how- ever, our contestant sample is more likely to be selected from individuals who are heavy lottery ticket buyers. thus our results can be interpreted as providing a quasi lower bound on the mean risk-aversion parameter for the population at large. references beetsma, roel, and peter schotman, “measuring risk attitudes in a natural experiment: data from the television game show lingo,” economic journal ( ), – . binswanger, hans, “attitudes towards risk: theoretical implications of an experiment in rural india,” economic journal ( ), – . camerer, colin, eric johnson, talia rymon, and sankar sen, “cogni- tion and framing in sequential bargaining for gains and losses” (pp. – ), ken binmore, alan kirman, and piero tani, (eds.), frontiers of game theory (cambridge, ma: mit press, ). carbone, enrica, and john hey, “a test of the principle of optimality,” university of york mimeograph ( ). chou, ray, robert f. engle, and alex kane, “measuring risk aversion from excess returns on a stock index,” journal of econometrics ( ), – . friend, i., and m. blume, “the demand for risky assets,” american economic review ( ), – . gertner, robert, “game shows and economic behavior: risk-taking on ‘card sharks’,” quarterly journal of economics ( ), – . gigliotti, gary, and barry sopher, “violations of present value max- imization in income choice,” theory and decision ( ), – . hersch, philip, and gerald mcdougall, “decision making under uncer- tainty when the stakes are high: evidence from a lottery game show,” southern economic journal ( ), – . kachelmeier, steven, and mohammed shehata, “examining risk- preferences using high monetary incentives: evidence from the people’s republic of china,” american economic review ( ), – . loewenstein, george, and drazen prelec, “negative time preference,” american economic review ( ), – . loewenstein, george, and richard thaler, “anomalies. intertemporal choice,” journal of economic perspectives ( ), – . metrick, andrew, “a natural experiment in jeopardy!,” american eco- nomic review ( ), – . rabin, mathew, “risk aversion and expected utility theory: a calibra- tion theorem,” econometrica ( a), – . “diminishing marginal utility of wealth cannot explain risk aversion,” in daniel kahneman and amos tversky, (eds.), choices, values, and frames, (new york: cambridge university press, b). rabin, mathew, and richard thaler, “anomalies. risk aversion,” journal of economic perspectives ( ), – . the small-sample bias of the gini coefficient: results and implications for empirical research george deltas* abstract—the gini coefficient is a downward-biased measure of inequal- ity in small populations when income is generated by one of three common distributions. the paper discusses the sources of bias and argues that this property is far more general. this has implications for (i) the comparison of inequality among subsamples, some of which may be small, and (ii) the use of the gini in measuring firm size inequality in markets with a small number of firms. the small-sample bias has often led to misperceptions about trends in industry concentration. a small-sample adjustment results in a reduced bias, which can no longer be signed. this remaining bias rises with the dispersion and falls with increasing skewness of the distribution. finally, an empirical example illustrates the importance of using the adjusted gini. in this example it is shown that, controlling for market characteristics, larger shipping cartels include a set of firms that is stochastically identical (in terms of relative size) to those of smaller shipping cartels. i. introduction this paper shows that the gini coefficient statisticexhibits a significant small-sample bias. the gini co- efficient of a large population estimated from a small sam- ple will be substantially smaller than the gini of the entire population. similarly, the gini of a small population will be smaller than the gini of a larger population generated by the received for publication january , . revision accepted for publication january , . * university of illinois, urbana-champaign. i would like to thank anil bera, dan bernhardt, hadi esfahani, roger koenker, richard sicotte, eleftherios zacharias, and an anonymous ref- eree for useful discussion and comments. partial financial support by the illinois ciber and the national science foundation (ecs- ) is gratefully acknowledged. responsibility for any errors rests with me. the review of economics and statistics the review of economics and statistics, february , ( ): – © by the president and fellows of harvard college and the massachusetts institute of technology submitted march accepted june published july corresponding author brice hanberry, bhanberry@fs.fed.us academic editor paolo giordani additional information and declarations can be found on page doi . /peerj. copyright hanberry distributed under creative commons cc-by . open access revisiting historical beech and oak forests in indiana using a gis method to recover information from bar charts brice hanberry rocky mountain research station, usda forest service, rapid city, sd, united states of america abstract historical gis involves applying gis to historical research. using a unique method, i recovered historical tree survey information stored in bar chart figures of a publication. i converted pdf files to tif files, which is a format for a gis layer. i then employed gis tools to measure lengths of each bar in the tif file and used a regression (r = %) to convert bar lengths to numerical values of tree composition. i joined this information to a spatial gis layer of indiana, usa. to validate results, i compared predictions against an independent dataset and written summaries. i determined that historically (circa to ) in indiana, oaks were % of all trees, beech was %, hickories and sugar maple were % each, and ash was . %. beech forests dominated (i.e., > % of all trees) % of . million ha (i.e., where data were available in indiana), oak forests dominated %, beech and oak forests dominated . %, and oak savannas were in % of indiana, resulting in beech and/or oak dominance in % of the state. this method may be valuable to reclaim information available in published figures, when associated raw data are not available. subjects biogeography, bioinformatics, data science, spatial and geographic information science keywords ash, data recovery, hickory, figure, sugar maple, tif introduction researchers increasingly are applying geographical information systems (gis) to a range of topics, including historical research (gregory & healey, ). a variety of approaches are available, with more methods and tools continually under development. currently, there is a great amount of information stored in publications that do not have associated, archived datasets, and in some cases, it may be possible to access that data using gis. one example is published information about historical tree surveys. primarily during the s, the general land office divided most of the united states into townships that were subdivided into sections, of . km ( mile) squares. surveyors recorded two to four tree species at section corners and halfway between section corners. these records provide information about forests before sustained euro-american settlement and disturbance. despite availability of valuable ecological data that provide a record of historical forests, transferring survey notes from the s to a more accessible format is time-intensive. in indiana, although survey notes were transcribed and analyzed by potzger, potzger & mccormick ( ), the data currently remain inaccessible, except in form of description, how to cite this article hanberry ( ), revisiting historical beech and oak forests in indiana using a gis method to recover informa- tion from bar charts. peerj :e ; doi . /peerj. https://peerj.com mailto:bhanberry@fs.fed.us https://peerj.com/academic-boards/editors/ https://peerj.com/academic-boards/editors/ http://dx.doi.org/ . /peerj. http://creativecommons.org/licenses/by/ . / http://creativecommons.org/licenses/by/ . / http://dx.doi.org/ . /peerj. maps, and graphs. potzger, potzger & mccormick ( ) presented the approximately , trees surveyed between and in bar chart format by township. to re-transfer data from the historical tree surveys to a gis layer, with current methods, would take about , working days at about one township per day, if survey notes are relatively legible. librarians at butler university (potzger collection) and purdue university (lindsey papers) were not able to locate any paper copies containing data tables of tree surveys. potzger, potzger & mccormick ( ) summarized in bar graph format only the five most common species or genera of american beech (fagus grandifolia), oaks (i.e., primarily white oak, quercus alba, but also including black oak, q. velutina, northern red oak, q. rubra, bur oak, q. macrocarpa, chestnut oak, q. prinus), sugar maple (acer saccharum), upland ash (primarily white ash, fraxinus americana), and hickories (carya spp.). unlike current forests, many historical forests in the united states were dominated by oaks, pines, or beech, so that information about beech and oaks alone is sufficient to describe forests (hanberry & nowacki, ). even though exact composition of the approximately tree species (potzger, potzger & mccormick, ) present historically in indiana remains unknown, species other than oak and beech were minor ( % to % of all trees; i.e., hickories, sugar maple, and upland ash) to trace (< %) components of historical forests at landscape scales. thus, where beech and oaks were not dominant, they typically were present with many relatively uncommon species (blewett & potzger, ; potzger & potzger, ; potzger, potzger & mccormick, ; lindsey, ). data in a bar chart format are not useable for other applications, beyond providing a general description, while tabular data are useful, particularly when associated with spatial location. my primary objective was to convert the bar charts presented by potzger, potzger & mccormick ( ) into tree compositional values joined to a gis layer of townships. i converted the graphs to a tif file and then used a regression between lengths in the bar charts and tree composition percentages, using previous work that provided tables of tree composition for townships within six counties (blewett & potzger, and potzger & potzger, ), to predict composition for the state. i also present ecological information reconstructed from bar charts to validate the method and show the value of recovered data. the study will provide the only source of tree percentages in historical forests of indiana available currently. this unique method may be helpful for retrieving other valuable datasets collapsed into figures that otherwise would be labor-intensive or impossible to duplicate. methods i converted the pdf file of the bar graphs from potzger, potzger & mccormick ( ) to a tif file, which is recognized as a raster (esri arcmap software v. . . , redlands, ca). the gis processing steps are as follows: . convert figure to tif format . extract by attribute to select dark colored bars (colors up to on color scale) . convert raster to a shapefile hanberry ( ), peerj, doi . /peerj. / https://peerj.com http://dx.doi.org/ . /peerj. . dissolve to turn bar multipart features to singlepart . cut any extraneous shapes that intersect bars . add geometry attributes (extent option) to measure bars . assign species and township (spatial location) information to each bar in attribute table . check data, correct or remove errors . regression between bar lengths and represented values . if data are spatial, join to spatial shapefile . validate results with independent or reserved data or written summaries. i extracted by attribute to select the dark colored bars (colors up to on color scale), converted the raster to a shapefile, and then dissolved to turn multipart features to singlepart (i.e., to make an outline of each bar; fig. ). because the bar graph contained information such as borders and township lines, i cut any extraneous shapes that intersected bars. to determine bar length (along x-axis; fig. ), i added geometry attributes (extent option). to provide location, i assigned townships to each row. because townships presented by potzger, potzger & mccormick ( ) were simplified, i processed each row to make sure it was assigned to the correct township and range, or removed rows that were indeterminate. for example, townships along the southern river border were difficult to determine correct assignment because rows were offset to avoid placement on the river. i removed townships with no data and small irregular polygons, for example, along rivers, resulting in a final total of , townships covering , , ha (of , , ha). i used township and range to join table information to a public land survey system shapefile, which is a spatially correct representation of townships and ranges. i assigned species information to each row based on row order, provided by potzger, potzger & mccormick ( ) in the bar graph legends (i.e., oak then hickory, or beech then maple then ash). for about townships, there were < rows for the beech, sugar maple, and upland ash. i matched the correct species/genus for each row based on location relative to other rows. because species/genus was difficult to match in isolated rows, i removed the few small beech, sugar maple, or upland ash rows that represented isolated groves from eight prairie counties of lake, newton, benton, warren, jasper, starke, pulaski, and white in the northwest (finley & potzger, ). i used townships, which were completely within county borders, with known beech, sugar maple, upland ash, oaks, and hickories percentage composition from blewett & potzger ( ) and potzger & potzger ( ) for a regression (sas, version . , cary, north carolina; proc reg). i regressed percentage against row length to calculate a relationship between percentage and row length (r = %). i then used the relationship to predict percent composition for each unknown row length. for validation, i compared results using root mean square error and mean absolute error to an independent dataset of four complete townships from a gis layer of glo tree surveys for the hoosier national forest in central southern indiana (available from the hoosier national forest). to compare to written descriptions and maps (presettlement vegetation map by lindsey, crankshaw & qadir ( ); generalized presettlement vegetation types hanberry ( ), peerj, doi . /peerj. / https://peerj.com http://dx.doi.org/ . /peerj. a b figure magnified image of original bar graph from potzger, potzger & mccormick ( ) as a gis layer (a) and example of row length measurements (b). full-size doi: . /peerj. /fig- of indiana, circa , indiana department of natural resources; http://maps.indiana. edu/previewmaps/environment/land_cover_presettlement_idnr.html), i distinguished prairie and oak savannas based on potzger, potzger & mccormick ( ) and classified forest type by identifying any species or genera that was greater than %. if no species hanberry ( ), peerj, doi . /peerj. / https://peerj.com https://doi.org/ . /peerj. /fig- http://maps.indiana.edu/previewmaps/environment/land_cover_presettlement_idnr.html http://maps.indiana.edu/previewmaps/environment/land_cover_presettlement_idnr.html http://dx.doi.org/ . /peerj. or genera was greater than %, then i distinguished the forest type as eastern broadleaf forest, although this may indicate both swamps and upland forests. i then overlaid beech and oak distributions on year mean precipitation gis layer for – at m resolution (prism; http://www.prism.oregonstate.edu/) and an elevation layer ( m ned; https://lta.cr.usgs.gov/ned) and derived elevation variables (e.g., slope degrees and terrain ruggedness). i also examined oak and beech distributions over statsgo soil values (miller & white, ; http://www.soilinfo.psu.edu/index. cgi?soil_data&conus&citation) related to water availability: percent clay, available water capacity, permeability, hydrologic soil groups, soil texture class, porosity, and plasticity. i summed values for the soil layers and calculated mean values for each forest type to determine if there was a strong underlying pattern missed by lindsey, crankshaw & qadir ( ). results to validate regression predictions (regression r = %) against another dataset, i compared predictions to four complete townships in a gis layer of glo surveys for the hoosier national forest. mean absolute error and root mean square error for predicted values compared to observed values were . and . for beech, . and . for oaks, . and . for hickories, and . and . for sugar maple (< for ash). furthermore, tree composition in hoosier national forest matched forest classifications. maximum predicted composition values by township for oaks ( % of all trees), beech ( %), hickories ( %), sugar maple ( %), and ash ( %) closely matched maximum values listed by potzger, potzger & mccormick ( ) of % composition for oak, % for beech, hickories and sugar maple never exceeded % composition in any township, and upland ash never exceeded % in any township. overall, for townships excluding prairie townships, oaks were % of all trees (area-weighted mean), beech was %, hickories and sugar maple were % each, and ash was . %. the other approximately species (potzger, potzger & mccormick, ) accounted for . % of all trees. in this study, beech and/or maple forests dominated (i.e., > % of all trees) % of the state (fig. ), while potzger, potzger & mccormick ( ) identified % beech-maple-ash forests and the presettlement vegetation map by lindsey, crankshaw & qadir ( ) contained % beech forest area. excluding oak openings, oak forests dominated % of the state, while potzger, potzger & mccormick ( ) identified % oak forests and the map by lindsey, crankshaw & qadir ( ) contained % oak forest area. oaks were present throughout the state, except in the northwestern prairie portion, and a few isolated townships, resulting in townships of , townships without oak. beech was present throughout the state, except in the northwestern prairie portion and southwestern portion, or townships total without beech. the presettlement vegetation map by lindsey, crankshaw & qadir ( ) generally matched forest types classified in this study (fig. ; http://maps.indiana.edu/previewmaps/environment/land_cover_presettlement_ idnr.html). lindsey, crankshaw & qadir ( ) classified beech-maple and oak-hickory forests based on if one had twice the importance value of the other; otherwise, forests were classified as beech-oak-maple-hickory. the beech-oak-maple-hickory forest type did not hanberry ( ), peerj, doi . /peerj. / https://peerj.com http://www.prism.oregonstate.edu/ https://lta.cr.usgs.gov/ned http://www.soilinfo.psu.edu/index.cgi?soil_data&conus&citation http://www.soilinfo.psu.edu/index.cgi?soil_data&conus&citation http://maps.indiana.edu/previewmaps/environment/land_cover_presettlement_idnr.html http://maps.indiana.edu/previewmaps/environment/land_cover_presettlement_idnr.html http://dx.doi.org/ . /peerj. prairie oak savanna oak oak-hickory oak-beech beech beech-maple sugar maple eastern broadleaf km " figure distribution of historical ecosystem types in indiana. full-size doi: . /peerj. /fig- match with forests in this study. also, prairies and wetlands were more extensive than indicated by potzger, potzger & mccormick ( ), which were used in this study. although the more eastern distribution of beech suggests potential for greater precipitation, year mean precipitation was cm for oak forests and . cm for hanberry ( ), peerj, doi . /peerj. / https://peerj.com https://doi.org/ . /peerj. /fig- http://dx.doi.org/ . /peerj. < - - - - > km a b figure distribution of historical oak (a) and beech (b) forests overlaid on available water capacity. full-size doi: . /peerj. /fig- beech forests. the lowest value was for prairie at cm, followed by oak openings at cm. oak-beech, oak-hickory, and beech-maple forests averaged – cm. oaks and beech were dominant in both the flat north and central portions of indiana, as well as the more dissected south. beech generally was located on moderately wet values for soil variables, but there were exceptions where beech ranged onto other soils, while wetter soil values spread beyond the border of beech distributions to where oaks and prairie also occurred (fig. ). mean value for available water capacity was . for oak forests and . for beech forests, but oak-beech forests had the lowest value of all forest types at . . sugar maple forests had the same value as oak forests, while prairies were about the same as beech forests and oak-hickory forests had greater values ( . ). discussion it was possible to convert previously published bar charts of historical composition of historically dominant tree species or genera in indiana (potzger, potzger & mccormick, ) to an accessible gis layer. despite the poor quality of the original figure (potzger, potzger & mccormick, ), the tif image was clear as a gis layer (fig. ). to my knowledge, although this method is relatively simple, there is no documentation of previous use of gis to recapture data condensed to bar chart format in publications that hanberry ( ), peerj, doi . /peerj. / https://peerj.com https://doi.org/ . /peerj. /fig- http://dx.doi.org/ . /peerj. do not have archived raw data. indeed, many current publications do not have associated datasets. the need to reconstruct data from figures also emphasizes the importance of sharing and archiving data, and the developed gis layer for this study will be archived. this method may be valuable to retrieve a variety of databases published in formats that are not archived or otherwise accessible. there were numerous sources of agreement between values recovered from bar charts and reported or independent data. there was a strong relationship (r = %) between length of rows in bar graphs presented by potzger, potzger & mccormick ( ) and compositional table values in blewett & potzger ( ) and potzger & potzger ( ). mean percentage differences between predictions and observed values from an independent dataset transcribed for the hoosier national forest were about % to %. maximum predicted values by township closely matched maximum values listed by potzger, potzger & mccormick; ( ; % compared to % for oaks, % compared to % for beech, % compared to % for hickories and sugar maple, and % compared to % for ash). forest types and areal extent generally agreed with potzger, potzger & mccormick ( ) and the presettlement vegetation map by lindsey, crankshaw & qadir ( ). recovered ecological information use of this method allowed reconstruction of historical composition by species, both spatially in a gis layer and by state. indiana is a relatively large state, and inclusion of . million ha of historical species composition will assist on-going efforts to piece together information about historical forests of the eastern united states (hanberry & nowacki, ). information about where both oak and beech were dominant particularly may be helpful to understand historical forests. excluding prairies, oaks were % of all trees (area-weighted mean), beech was %, hickories and sugar maple were % each, and ash was . %. beech and/or oak forests (i.e., > % of all trees) dominated % of townships, or % of . million ha due to township area irregularity. although there were no species/genera > % of all trees in townships, of these townships contained either beech or oak percentage > . sugar maple and hickories, which were the third and fourth most common tree species or genera (potzger, potzger & mccormick, ), were dominant alone in seven townships combined, or . % of all townships. hickories often are associated with oaks while sugar maple is associated with beech, and yet, historically, hickory and sugar maple abundance was secondary to oak and beech predominance. indeed, in beech-dominated forests (mean of % of all trees), oaks ( %) were as common as sugar maple ( %), with hickories and ash at around %. likewise, in oak-dominated forests (mean of % of all trees), beech ( %) was just slightly less abundant than hickories ( %). there was overlap of beech and oak dominance in townships. in these finer scale locations, oaks and beech may have remained separated spatially by local moisture conditions, at least in topographically variable sites. for example, beech was present on wetter north-facing slopes while oaks were present on drier south-facing slopes (potzger, potzger & mccormick, ); however, whitney ( ) was not able to determine this relationship in northeastern ohio. similarly, beech were present in protected valleys or on hanberry ( ), peerj, doi . /peerj. / https://peerj.com http://dx.doi.org/ . /peerj. lower slopes while oaks were present on drier and more exposed ridges and upper slopes (whitney, ; cowell & hayes, ). as for soils, whitney ( ) found that beech were associated with imperfectly drained soils from fine-textured glacial till soils, although he was not able to find congruence between beech forests and wisconsin drift deposits from glaciation. at a coarser scale of townships and rather than individual trees, the separation of oak and beech forests may be difficult to explain using precipitation or soil moisture based on topography and associated soils. precipitation values were approximately cm in historical beech and oak forests, albeit using modern precipitation values for – , which are available at m appropriate for resolution within states. both beech and oak forests were dominant in the flat topography of northern indiana and the more dissected topography of southern indiana. beech forests overall were restricted to moderately wet soils, with exceptions (fig. ). nonetheless, moderately wet soils were abundant in indiana and the eastern us and extended to where oaks, prairie, and other forest types were present, rather than restricted to beech distributions. lindsey, crankshaw & qadir ( ), after extensive comparison between soils and forest types, were not able to isolate forest types based on soils. fire was near-annual in prairie states of the central us and in the southeastern us and became less frequent to the east and north (day, ; fowler & konopik, ). guyette, dey & stambaugh ( ) quantified a year mean fire interval between and near the ohio river in southern indiana. frequent surface fires favored prairies, open forests of savannas and woodlands, and fire-tolerant upland oak species (hanberry, jones-farrand & kabrick, ). whitney ( ) noted that fire was recorded in < % of surveyor records for three counties in ohio along the border between beech- and oak-dominated forests. however, % of fire records occurred in % of townships, which were centers of native american activity. in indiana, the wabash confederacy, for example, probably cleared land through girdling and fire for use along the wabash river (butler, ). butler ( ) wrote about prairies and open forests filled with american bison (bison bison), particularly along river valleys and near west lafayette (i.e., ouiatanon) in indiana, and into kentucky prairies. although oaks in general were the dominant genera ( % of all trees) for the ecological region of the central eastern united states (hanberry & nowacki, ), there also were conditions that favored beech forests interspersed with oak forests. in addition to areas at fine scales that were protected from fire by water, wetlands, moist soils, rugged topography, and rocky outcrops, perhaps fire-protected conditions at landscape scales interacted with reduced or localized native american fire regimes to produce larger landscapes where fire was less frequent. butler ( ) stated that southeastern indiana was not settled by american indian villages, unlike western indiana along the wabash river and northeastern indiana. central and eastern indiana thus appeared to be one such stronghold where fire protection allowed beech forests, with some continuity into ohio, but surrounded to the north, west, and south by oak forests, woodlands, savannas, and prairies. hanberry ( ), peerj, doi . /peerj. / https://peerj.com http://dx.doi.org/ . /peerj. conclusions data published during the s may no longer be effectively available except in figures. i applied gis to recapture historical data contained in the length of hundreds of column bars in two figures from a publication. use of gis to recover information compressed into figures, with lengths that represent values, is a method that can be applied to other publications where data have been lost. acknowledgements i thank p hanberry for gis assistance and anonymous reviewers. this paper may not reflect views of the usda forest service. additional information and declarations funding the author received no funding for this work. competing interests the author declares there are no competing interests. author contributions • brice hanberry conceived and designed the experiments, analyzed the data, contributed reagents/materials/analysis tools, prepared figures and/or tables, authored or reviewed drafts of the paper, approved the final draft. data availability the following information was supplied regarding data availability: the raw data are provided in a supplemental file. supplemental information supplemental information for this article can be found online at http://dx.doi.org/ . / peerj. #supplemental-information. references blewett mb, potzger je. . the forest primeval of marion and johnson counties, indiana, in . butler university botanical studies : – . butler aw. . indiana: a century of changes in the aspects of nature. proceedings of the indiana academy of science : – . cowell cm, hayes jj. . structure, history and dynamics of a mature oak-beech forest in western indiana. the journal of the torrey botanical society : – doi . / - ( ) [ :shadoa] . .co; . day gm. . the indian as an ecological factor in the northeastern forest. ecology : – doi . / . hanberry ( ), peerj, doi . /peerj. / https://peerj.com http://dx.doi.org/ . /peerj. #supplemental-information http://dx.doi.org/ . /peerj. #supplemental-information http://dx.doi.org/ . /peerj. #supplemental-information http://dx.doi.org/ . / - ( ) [ :shadoa] . .co; http://dx.doi.org/ . / http://dx.doi.org/ . /peerj. finley d, potzger je. . characteristics of the original vegetation in some prairie counties of indiana. butler university botanical studies : – . fowler c, konopik e. . the history of fire in the southern united states. human ecology review : – . gregory in, healey rg. . historical gis: structuring, mapping and analysing geographies of the past. progress in human geography : – doi . / . guyette rp, dey dc, stambaugh mc. . fire and human history of a barren-forest mosaic in southern indiana. the american midland naturalist : – doi . / - ( ) [ :fahhoa] . .co; . hanberry bb, jones-farrand dt, kabrick jm. . historical open forest ecosystems in the missouri ozarks: reconstruction and restoration targets. ecological restoration : – doi . /er. . . . hanberry bb, nowacki gj. . oaks were the historical foundation genus of the east-central united states. quatenary science reviews : – doi . /j.quascirev. . . . lindsey aa. . vegetation of the drainage-aeration classes of northern indiana soils in . ecology : – doi . / . lindsey aa, crankshaw wb, qadir sa. . soil relations and distribution map of the vegetation of presettlement indiana. botanical gazette : – doi . / . miller da, white ra. . a conterminous united states multilayer soil characteristics dataset for regional climate and hydrology modeling. earth interactions : – doi . / - ( ) < :acusms> . .co; . potzger je, potzger me. . composition of the forest primeval from hendricks county southward to lawrence county, indiana. proceedings of the indiana academy of science : – . potzger je, potzger me, mccormick j. . the forest primeval of indiana as recorded in the original us land surveys and an evaluation of previous interpretations of indiana vegetation. butler university botanical studies : – . whitney gg. . vegetation-site relationships in the presettlement forests of north- eastern ohio. botanical gazette : – doi . / . hanberry ( ), peerj, doi . /peerj. / https://peerj.com http://dx.doi.org/ . / http://dx.doi.org/ . / - ( ) [ :fahhoa] . .co; http://dx.doi.org/ . /er. . . http://dx.doi.org/ . /j.quascirev. . . http://dx.doi.org/ . / http://dx.doi.org/ . / http://dx.doi.org/ . / - ( ) < :acusms> . .co; http://dx.doi.org/ . / http://dx.doi.org/ . /peerj. untitled erratum for ‘devonian (c. – ma) horn river group of mackenzie platform (nw canada) is an open-shelf succession recording oceanic anoxic events’, journal of the geological society, london, , - p. kabanov geological survey of canada calgary, calgary, ab t l a , canada pk, - - - correspondence: kabanovp@gmail.com the original caption of figure did not list the section names correctly. in the below figure, the numbering in a has been updated and the caption contains the names of all of the sections. pc in b has also been defined. figure . (a) geographical spread of the horn river group (on nwt side) and canol formation (on yukon side) between and °n; crosses are outcrops: ( , trail river (tr in (b)); , rumbly creek; , turnabout creek; , powell creek, and , prohibition creek); palaeogeographic areas: wob, western off-bank area; bat, bank-and-trough area; sob, southern off-bank area. (b) cross-section (black line (a)) levelled at the top of hume formation; wells from west to east: cranswick yt a- , cranswick a- , s. ramparts i- , n. ramparts a- , ramparts river f- , hume river i- , hume river d- , carcajou l- , maida creek f- , hoosier f- , nwb is norman wells oilfield, little bear n- , bluefish a- , and bracket lake c- . stratigraphic members in sob: ( fc) francis creek, ( ps) prohibition creek, (vc) vermillion creek; (dc) dodo canyon, (ml) mirror lake, and (lc) loon creek. nrtf is norman range thrust fault (other tectonic elements are not shown). the canol formation in wob area is provisionally subdivided into ( ) lower, ( ) middle, and ( ) lower traceable units or information members (cross-section b-b’ in supplementary material). (pc) on (b) is projection of the powell creek outcrop (stratotype of canol formation). © the author(s). published by the geological society of london. all rights reserved. for permissions: http://www.geolsoc.org.uk/permissions. publishing disclaimer: www.geolsoc.org.uk/pub_ethics erratum journal of the geological society published online december , https://doi.org/ . /jgs - | vol. | | p. by guest on april , http://jgs.lyellcollection.org/downloaded from https://doi.org/ . /jgs - https://doi.org/ . /jgs - http://orcid.org/ - - - mailto:kabanovp@gmail.com http://www.geolsoc.org.uk/permissions http://www.geolsoc.org.uk/pub_ethics http://crossmark.crossref.org/dialog/?doi= . /jgs - &domain=pdf https://doi.org/ . /jgs - http://jgs.lyellcollection.org/ ccr- - .. cancer therapy: clinical a phase ii trial of dovitinib in bcg-unresponsive urothelial carcinoma with fgfr mutations or overexpression: hoosier cancer research network trial hcrn - noah m. hahn , , , trinity j. bivalacqua , , , ashley e. ross , , , george j. netto , , , alex baras , , , jong chul park , carolyn chapman , timothy a. masterson , michael o. koch , richard bihrle , richard s. foster ,thomas a. gardner , liang cheng , david r. jones , kyle mcelyea , george e. sandusky , timothy breen , ziyue liu , costantine albany , marietta l. moore , rhoda l. loman , angela reed , scott a.turner , francine b. de abreu , torrey gallagher , gregory j. tsongalis , elizabeth r. plimack , richard e. greenberg , and daniel m. geynisman abstract purpose: to assess the clinical and pharmacodynamic activity of dovitinib in a treatment-resistant, molecularly enriched non– muscle-invasive urothelial carcinoma of the bladder (nmiuc) population. experimental design: a multi-site pilot phase ii trial was conducted. key eligibility criteria included the following: bacillus calmette-guerin (bcg)-unresponsive nmiuc (> prior intrave- sical regimens) with increased phosphorylated fgfr (pfgfr ) expression by centrally analyzed immunohistochemistry (ihcþ) or fgfr mutations (mutþ) assessed in a clia-licensed labora- tory. patients received oral dovitinib mg daily ( days on/ days off). the primary endpoint was -month turbt-confirmed complete response (cr) rate. results: between / and / , patients enrolled ( ihcþ mut�, ihcþ mutþ). accrual ended prematurely due to cessation of dovitinib clinical development. demographics included the following: median age years; % male; carci- noma in situ (cis; patients), ta/t ( patients), and ta/t þ cis ( patients); median prior regimens . toxicity was frequent with all patients experiencing at least one grade – event. six-month cr rate was % ( % in ihcþ mut�; % in ihcþ mutþ). the primary endpoint was not met. pharmacodynamically active ( – , nmol/l) dovitinib concentrations in urothelial tissue were observed in all evaluable patients. reductions in pfgfr ihc staining were observed post-dovitinib treatment. conclusions: dovitinib consistently achieved biologically active concentrations within the urothelium and demonstrated pharmacodynamic pfgfr inhibition. these results support systemic administration as a viable approach to clinical trials in patients with nmiuc. long-term dovitinib administration was not feasible due to frequent toxicity. absent clinical activity suggests that patient selection by pfgfr ihc alone does not enrich for response to fgfr kinase inhibitors in urothelial carcinoma. clin cancer res; ( ); – . � aacr. introduction urothelial carcinoma of the bladder is the fifth most common human cancer diagnosis. in , more than , individuals are expected to be diagnosed with urothelial carcinoma and more than , patients to die from their disease ( ). most new urothelial carcinoma cases (� , patients) are non–muscle- invasive at diagnosis with disease limited to the mucosal epithe- lium (ta/tis) and immediate connective tissue layer beneath the johns hopkins university sidney kimmel comprehensive cancer center, baltimore, maryland. the johns hopkins greenberg bladder cancer insti- tute, baltimore, maryland. the james buchanan brady urological institute, baltimore, maryland. department of urology, indiana university simon cancer center, indianapolis, indiana. department of pathology and labo- ratory medicine, indiana university simon cancer center, indianapolis, indi- ana. department of clinical pharmacology, indiana university simon cancer center, indianapolis, indiana. hoosier cancer research network, indiana- polis, indiana. indiana university department of biostatistics, schools of public health and medicine, indianapolis, indiana. division of hematology and oncology, indiana university simon cancer center, indianapolis, indiana. department of pathology and laboratory medicine, geisel school of medicine at dartmouth, dartmouth hitchcock medical center and norris cotton cancer center, lebanon, new hampshire. department of hematol- ogy and oncology, fox chase cancer center, philadelphia, pennsylvania. department of urology, fox chase cancer center, philadelphia, pennsylvania. note: supplementary data for this article are available at clinical cancer research online (http://clincancerres.aacrjournals.org/). prior presentation: presented in part as a poster at the american society of clinical oncology annual meeting, chicago, il. study clinicaltrials.gov identifier: nct . corresponding author: noah m. hahn, department of oncology and urology, the bunting-blaustein cancer research building , room m , orleans street, baltimore, md . phone: - - ; fax: - - ; e-mail: nhahn @jhmi.edu doi: . / - .ccr- - � american association for cancer research. clinical cancer research www.aacrjournals.org on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ mucosa (t ; ref. ). the clinical course of non–muscle-invasive urothelial carcinoma of the bladder (nmiuc) is dominated by frequent recurrences requiring surveillance (with cystoscopy, bladder biopsy, urine cytology, etc.). the need for long-term invasive monitoring and treatment has significant cost and mor- bidity for patients with urothelial carcinoma. compared with other malignancies, urothelial carcinoma ranks highest in lifetime per patient costs with an average cost from diagnosis to death of $ , per patient ( ). standard therapy for high-risk patients with nmiuc includes transurethral resection of bladder tumor (turbt) augmented by intravesical administration of bacillus calmette-guerin (bcg), an attenuated bovine mycoplasma–derived agent. two meta- analyses of randomized trials of turbt plus bcg versus turbt alone demonstrated a reduction in -month tumor recurrence rate from % to % (p < . ) and a reduction in progression to muscle-invasive stages from . % to . % (p ¼ . ) in association with bcg therapy ( , ). while bcg therapy is successful at preventing early tumor recurrences, most patients do not maintain sustained remissions. with -year follow-up, recurrent bladder tumors requiring repetitive turbt and further cystoscopic surveillance are observed in % to % of patients ( , ). for post-bcg tumor recurrences, bcg-unresponsive dis- ease is defined by any of the following features: recurrent nmiuc after prior adequate bcg regimens, recurrent t disease at the initial -month posttreatment turbt, recurrent nmiuc within months of last bcg administration, and nmiuc involving the prostatic urethra ( ). transient remissions are often observed with additional intravesical therapy approaches; however, only % to % of patients remain recurrence-free at year ( , ). thus, cystectomy is considered a standard treatment in bcg-unrespon- sive patients ( ). a need clearly exists to explore the clinical efficacy of novel agents in this high-risk nmiuc population. across multiple cancer types, the critical role of angiogenesis in tumor migration, proliferation, and metastasis is well established with vegf and vegfr serving as key mediators ( , ). in urothelial carcinoma, associations between increased tumor vegf expression and high-grade disease, advanced stage, and poor prognosis have been observed ( – ). initial phase ii trials in metastatic patients with urothelial carcinoma combining che- motherapy with the anti-vegfr monoclonal antibody bevaci- zumab have demonstrated promising overall survival outcomes compared with historical controls with a definitive phase iii trial of chemotherapy with or without bevacizumab completed and data maturing ( , ). in addition to vegfr, fgfr has been implicated as a critical facilitator of urothelial carcinoma carcinogenesis, particularly in nmiuc ( , ). fgfr mutations or overexpression promote fgfr dimerization and constitutive activation of downstream signaling pathways in the absence of ligand in up to % of low- grade nmiuc tumors ( ). these mutations result in a hyper- plastic phenotype dominated by frequent tumor recurrences with infrequent progression to muscle-invasive stages. while fgfr mutations are highly associated with low-grade nmiuc, over- expression of fgfr has been observed in up to % of high- grade muscle-invasive urothelial carcinoma tumors ( ). further- more, either an fgfr mutation or overexpression of the fgfr protein in the absence of mutation has been observed in % of muscle-invasive urothelial carcinoma tumors ( ). thus, while fgfr mutations likely are an early event in the tumorigenesis of low-grade noninvasive urothelial carcinoma tumors, alterations of fgfr may still play a role in the continued proliferation of high-grade urothelial carcinoma. dovitinib is an oral tyrosine kinase inhibitor of fgfr - , vegfr - , pdgfrb, c-kit, ret, trka, csf- r, and flt which has demonstrated a tolerable safety profile in single agent and combination regimens ( ). increasing evidence demonstrates that fgfr is a crucial mediator of tumor angiogenesis ( ). in preclinical tumor models, blockade of the fgf pathway has proven to be an effective method of overcoming resistance to vegfr inhibitors ( ). given the previously described impor- tance of vegf in urothelial carcinoma progression and the frequent fgfr aberrations in nmiuc, we conducted a multi- site pilot trial in patients with bcg-unresponsive nmiuc har- boring fgfr gene alterations to evaluate the clinical and biologic outcomes of oral dovitinib therapy. materials and methods study design a single-arm, nonrandomized, multicenter phase ii study (nct ) was conducted between sites: indiana univer- sity simon cancer center (indianapolis, in), fox chase cancer center (philadelphia, pa), and johns hopkins sidney kimmel comprehensive cancer center (baltimore, md). standard of care and correlative biospecimens were collected pre- and posttreat- ment from all patients. the study was approved by the institu- tional review boards of each site. patients key eligibility criteria included: histologically confirmed ta, t , or tis stage nmiuc assessed by turbt performed within days of registration; somatic tumor mutations in fgfr exons , , or (s c, g c, y c, g r, k e, k q, k t, k m, a e, s c, and r c) or tumor overexpression of phosphorylated (pfgfr ) by immunohistochemistry (ihc) translational relevance this trial reports the toxicity, pharmacodynamics, and clin- ical efficacy profiles of the oral fgfr - and vegfr - multi- targeted tyrosine kinase inhibitor, dovitinib, in a pilot phase ii investigation in patients with bacillus calmette-guerin (bcg)-unresponsive non–muscle-invasive urothelial carcino- ma of the bladder (nmiuc) with tumors harboring fgfr alterations. in addition to demonstrating reductions in post- treatment pfgfr , confirmed biologically active dovitinib concentrations were observed in the bladder urothelium. this trial is the first nmiuc study to require genomic testing as an eligibility requirement and to demonstrate successful achieve- ment of therapeutic urothelial tissue concentrations of sys- temically administered targeted therapies. thus, it greatly expands the potential therapeutic approaches to treat this high-risk population. lack of clinical efficacy was hampered by frequent drug toxicity and a paucity of patients harboring fgfr mutations. additional fgfr targeting approaches in molecularly enriched urothelial carcinoma populations are ongoing and clearly worthy of further study, including in patients with nmiuc. hahn et al. clin cancer res; ( ) june , clinical cancer research on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ defined as þ or greater tumor pfgfr staining; recurrent nmiuc despite at least prior intravesical treatment regimens (no limit), one of which must have been bcg; patients medically unfit for or refusing cystectomy; age > years; ecog perfor- mance status – ; adequate hematologic and liver function; creatinine clearance > ml/min by modified cockcroft–gault equation; and documented, written informed consent. major exclusion criteria included: evidence of muscle-invasive or met- astatic disease on pre-study screening tests; concurrent upper tract urothelial carcinoma; prior vegfr or fgfr-targeted therapy. treatment patients were treated with dovitinib mg by oral admin- istration once per day for consecutive days followed by days off each week. a cycle was defined as weeks of therapy. no maximum number of treatment cycles was stipulated. dose reductions to and mg were permitted in the event of treatment-related toxicity. because of drug–drug interactions, full dose anti-coagulation with warfarin was not allowed, however, use of low-molecular weight heparin at full-dose was permitted. usage of anti-emetic and colony-stimulating growth factor medications was at the discretion of the treating physician. disease evaluations at baseline, the absence of metastatic disease was confirmed by abdomen and pelvis ct scan and chest x-ray or ct scan. adequate cardiac function was confirmed by echocardiogram and electro- cardiogram assessments. history and physical examination find- ings, vital signs, baseline symptoms, and laboratory assessments were performed within days of registration. exams, vital signs, toxicity evaluations (per ctcae v . ), and laboratory assess- ments were performed biweekly for the first cycles of treatment and every weeks thereafter. all patients were evaluated with urine cytology and cystoscopy every months during the first year and per the treating physi- cian's discretion thereafter. turbt's were required at - and - month posttreatment with only for cause turbt's thereafter. at each turbt, biopsy tissue was obtained from all previous and new tumor sites, the bladder dome, anterior bladder wall, left lateral bladder wall, right lateral bladder wall, and the bladder trigone. patients with any nmiuc at the -month evaluation or beyond were considered relapses as were patients with carcinoma in situ (cis) at the -month evaluation. patients with papillary- only disease at the -month cystoscopy/turbt who declined further dovitinib therapy were classified as relapsed. testing of urine for evidence of relapse by fish was allowed but not required. the same was true for the use of blue light cystoscopy. an isolated fish-positive urine finding was not classified as a relapse event. complete response (cr) was defined as no evi- dence of any remaining urothelial carcinoma tumors of any t stage (including tis) as assessed by cystoscopic examination and urine cytology. in addition to these criteria, the -month cr rate required no evidence of tumor within the -month posttreatment turbt biopsies. the -year relapse-free survival rate was defined as the proportion of patients treated with dovitinib with no evidence of any urothelial carcinoma at months of follow up. patients with any evidence of muscle-invasive tumors (t or above) or metastatic disease in follow-up were considered as progressive disease. at the time of all turbts or cystectomy, tumor samples were sent for standard-of-care diagnostic evaluation and complete pathologic staging information was recorded. samples from the same blocks were cut and archived for correlative studies. resolution of any treatment-related toxicities was confirmed days after administration of a patient's last dovitinib dose. patients were not followed for long-term overall survival outcomes. fgfr mutation analysis at baseline, individual -mm-thick slides were cut from the patient's representative turbt block with the highest grade tumor and greatest volume of tumor present. in slides with less than % tumor cells present, macrodissection was performed to ensure maximum tumor cell dna content. also, at baseline, a -ml urine sample was obtained from all patients and centrifuged at , rpm for minutes. the resulting super- natant and cell pellet were transferred into separate cryovials and stored at � �c until analyzed. slides and urine cell pellets were shipped to the laboratory for clinical genomics and advanced technology (cgat) at the dartmouth hitchcock medical center. tumor and urine cell pellet dna extraction was performed per manufacturer's specification [qiagen pure- gene (tissue) and qiagen dneasy (cell pellet)]. fgfr muta- tional status was determined using a custom designed snap- shot assay (thermofisher scientific) for all common mutations in fgfr coding exons including exons , , and . the presence or absence of specific fgfr mutations (s c, g c, y c, g r, k e, k q, k t, k m, a e, s c, and r c) was communicated to hcrn within days of specimen receipt. phosphorylated fgfr ihc analysis simultaneously at baseline, individual -mm-thick slides from the patient's tumor and a single hematoxylin and eosin (h&e) slide from the same block were shipped to the iuscc immunohistochemistry core laboratory for fgfr ihc anal- ysis. slides were heated to �c for minutes. slides were deparaffinized and rehydrated sequentially with xylene ( minutes � ), % ethyl alcohol solution ( minutes � ), and % ethyl alcohol solution ( minutes � ) on a sakura linear stainer. antigen retrieval utilized pt link (pt , dako) in conjunction with envision flex high ph target retrieval solution (k , dako). cycles began at �c and were heated to �c for minutes followed by cooling back to �c and placement in wash buffer (k , dako). baseline phosphorylated fgfr staining for trial eligibility evaluation was performed on a dako autostainer platform utilizing the sc- anti-fgfr (phospho y ) antibody (santa cruz biotechnology). the sc- pfgfr antibody was optimized to a : dilution for minutes prior to the conduct of this trial utilizing breast cancer cases as positive controls. fol- lowing pfgfr staining, slides were dehydrated sequentially with % ethyl alcohol solution ( minutes � ), % ethyl alcohol solutions ( minutes � ), and xylene ( minutes � ) followed by coverslipping. the immunostained slides were evaluated by different pathologists. areas within the tumor were scored as follows: ¼ negative, þ ¼ mild staining, þ ¼ moderate staining, þ ¼ strong staining. both positive and negative controls were run in addition to the samples. because a dovitinib in bcg-unresponsive nmiuc with fgfr alterations www.aacrjournals.org clin cancer res; ( ) june , on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ clinically relevant cutoff for pfgfr ihc intensity had not previously been established, tumors with any staining intensity ( þ or greater) were considered pfgfr -overexpressing. dur- ing the conduct of the trial, improved commercial pfgfr antibodies became available. the correlative pre- and posttreat- ment pfgfr analyses were, therefore, performed utilizing the ab anti-fgfr (phospho y ) antibody (abcam). the ab evaluation scheme was validated across different individual cases of bladder cancer for antibody specificity. triplicate runs of this validation scheme showed that low strainers ( þ), moderate strainers ( þ), and strong strainers ( þ) were replicated across all runs. high, medium, and low staining positive controls were identified and used across all runs. all other antibody optimization procedures mirrored those of the sc- antibody with the exception that the ab antibody was optimized with to a : dilution for minutes. the sc- antibody continued to be utilized for eligibility determination throughout the entire conduct of the study. aperio's scanscope cs whole slide digital imaging sys- tem (leica biosystems) was used for baseline and posttreat- ment pfgfr pathology imaging. the system imaged all slides at �. the scan time ranged from / minutes to a maximum time of / minutes. the whole images were housed and stored in their spectrum software system and images were shot from the whole slides. quantification of pfgfr staining was performed on the halo image analysis platform (indica labs). an algorithm was designed on the basis of pattern recognition that quantified tumor cells within pfgfr -positive areas (tumor) and pfgfr -negative areas (invasive margin). halo's classifier package performed image analysis based on rgb (red, green, blue) spectra which was used to detect cells positively expressing pfgfr against negative expressing counterstained hematoxylin cells. the algorithm calculated the classified area (mm ) and percentage of tumor expression (% positive cells/% of all nucleated cells) using the halo classifier package. the total percentage of positive expression in each group was averaged and sd was calculated. further analysis was per- formed on hotspots on each tissue via halo's area quanti- fication package. an algorithm was designed to quantify pos- itive pfgfr -expressing tumor cells in weak, moderate, and high positivity values. an average of hotspots for the tissues collected at day was calculated along with sd. these data were compared with an average of hotspots of the tissues collected between cycle days and for their total positivity and according to weak, moderate, high, and total expression. dovitinib pharmacokinetic tissue analysis at the -month posttreatment disease assessment, a bladder biopsy of tumor or normal appearing urothelium was obtained for pharmacokinetic analysis to confirm achievement of bio- logically active dovitinib tissue concentrations via oral drug administration. the pharmacokinetic biopsy sample was flash- frozen, stored in liquid nitrogen, and shipped to the iuscc clinical pharmacology analytical core (cpac) for analyses. tissue samples were homogenized in pbs, internal standard (sorafenib) was added to each sample, the samples were extracted with ethyl acetate, and injected into an hplc-ms/ ms (api ; ab sciex). plasma was used for the matrix of the standard samples to estimate tissue concentrations. the lower limit of quantification was ng/sample. for ease of comparison, tissue concentrations (ng/g) were converted to the nanomolar concentrations (assuming -g tissue is equiva- lent to -ml water). statistical considerations the primary endpoint of the trial was -month cr rate. with a -month cr rate of clinical interest of � %, a sample size of patients provided an % power to exclude a lower bound of � % utilizing a one-sided % confidence interval (ci) of agresti–couli type. with an estimated fgfr mutation or overexpression present in % of bcg-unresponsive tumors, screening of patients' tumors was estimated to enroll the required patients on dovitinib therapy. an evaluation of early stopping was planned at the first patients completing -month assessment for progression to t or greater stages, whose objective was to stop the study if the likelihood of progression rate was more than %. a rule was chosen that the study should be terminated if or more progressions were observed of patients, which is the minimal number that leads to a % agresti–coull ci with a lower bound above %. rates of cr, progressive disease, and treatment-related toxicity were summarized by % cis. associations between pre- and posttreatment pfgfr ihc staining intensity were compared by paired t testing with significance set at p < . . results patients between november and october , patients were screened and patients were enrolled. fifteen patients ( %) had sufficient tumor tissue for fgfr mutation testing. two patients with tumors demonstrating no fgfr mutations were considered screen failures after the study amendment capping the enrollment of fgfr mutation–negative patients was in place. further accrual was stopped because of cessation of clinical development of dovitinib. patient demographics are summarized in table and included: median age years (range, – years), % male, and % caucasian. baseline turbt tumor stages were: cis, patients; ta or t , patients; and ta or t with concurrent cis, patients. patients had received a median of prior intravesical regimens (range, – ) with all patients having received at least prior bcg induction courses. the median time from last intravesical therapy was months (range, – ). tumor fgfr mutations were detected in patients ( % of screened patients) with a concordant urine fgfr mutation detected in of the patients. dovitinib treatment patients received a median of cycles of dovitinib treatment (range, – ). ten patients ( %) required dovitinib dose reduc- tions. two patients ( %) discontinued dovitinib treatment prematurely and did not undergo planned -month posttreat- ment disease evaluations. reasons for discontinuation included: physician discretion discontinuation of treatment due to a trau- matic intracranial hemorrhage sustained in a ground-level fall unrelated to study treatment ( patient) and patient choice to withdraw from study ( patient). in addition, dovitinib therapy was discontinued per treating physician's discretion in a single patient after cycles after the patient revealed a prior history of retinal detachment unknown to the treating team at study enrollment. hahn et al. clin cancer res; ( ) june , clinical cancer research on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ toxicity dovitinib therapy was associated with frequent toxicity. all patients ( %) experienced at least grade or event. treatment-related grade hypertriglyceridemia was observed in patient ( %). treatment-related grade events included fatigue, elevated g-glutamyl transferase (ggt), and elevated lipase in patients ( %) each as well as headache, hypertriglyceridemia, stomatitis, and rash in patient ( %) each. one patient ( %) suffered a subdural intracranial hemorrhage that did not require operative intervention in association with a ground-level fall on an ice-covered winter sidewalk that was not deemed treatment- related. all grade – events and other toxicities occurring in more than % of patients are summarized in table . complete all- grade toxicity is included in supplementary table s . tumor response antitumor responses to dovitinib treatment were infrequent. of the patients enrolled, a pathologic cr was observed in patient ( %). nonresponse was observed in patients ( %) and progression to muscle-invasive stage occurred in patient ( %). the single patient with cr did harbor an fgfr s c mutation. thus, the pathologic cr rate amongst fgfr mutþ patients was % ( of ) as summarized in table . the patient remains in a cr at þ months of follow-up. eight patients ( %) underwent cystectomy per the discretion of their physician at any time point following completion of study therapy with a wide variety of pathologic stages ranging from pt n to pnþ disease (supplementary table s ). dovitinib pharmacokinetic tissue analysis fresh tumor or adjacent normal urothelium biopsy tissue was available for dovitinib pharmacokinetic analysis from of the patients who underwent posttreatment disease evaluations. as shown in fig. , dovitinib was detectable at pharmacologically active levels in all patients examined with tissue concentrations ranging from to , nmol/l. ihc analysis of dovitinib treatment on pfgfr all baseline slides for eligibility determination demonstrated positive pfgfr staining as assessed by the sc- pfgfr antibody. staining intensities according to the use of the sc- and ab pfgfr antibodies showed significant heterogeneity (supplementary table s ). pre- and post-dovi- tinib treatment slides were available from patients including tumor pairs. utilizing the quantitative halo classifier imaging platform, reductions in averaged pfgfr staining area from . to . mm were observed following dovitinib treatment. as depicted in fig. , this posttreatment reduction in mean pfgfr staining area showed a strong trend but did not reach statistical significance (p ¼ . ). marked reductions in pfgfr staining were observed in of patients, of which one reduction is demonstrated in fig. . discussion until the recent fda approval of the immunotherapy agent atezolizumab in metastatic urothelial carcinoma patients, near- ly a quarter century had passed without any significant advances in systemic therapy for urothelial carcinoma ( ). while the approval of atezolizumab is encouraging, it is impor- tant to note that only a small subset of patients derive benefit. thus, additional novel approaches to treat urothelial carcinoma are clearly needed. in particular, innovative strategies for the two thirds of patients with urothelial carcinoma initially pre- senting with nmiuc are paramount. given the established relevance of vegfr in urothelial carcinoma cancer invasion and metastases and the striking frequency of fgfr aberrations in low-grade nmiuc, we postulated that an fgfr /vegfr - directed approach with dovitinib would prove both feasible and beneficial in patients with bcg-unresponsive nmiuc with tumors harboring fgfr alterations. our study failed to demonstrate significant clinical activity with dovitinib therapy in the enrolled study population. lim- itations in the enrollment criteria for the study population likely played a major factor in the absent antitumor activity observed. at the time the study was designed, the relative importance of fgfr mutations versus gene fusions versus overexpression was unknown. furthermore, clinically relevant cutoffs for pfgfr ihc staining had not been established and available commercial pfgfr antibodies were limited. there- fore, even though robust methodology was developed prior to study initiation to optimize pfgfr antibody procedures, our trial allowed patients with any degree of pfgfr ihc staining at baseline to enroll. this allowed for an early influx of ihcþ mut� patients. as demonstrated by the frequent heterogeneity that was observed in baseline ihc intensities according to the pfgfr antibody utilized, this likely resulted in a less table . baseline patient and tumor characteristics patient gender age, y race t-stage prior regimens time from last therapy, mo tumor fgfr mutation urine fgfr mutation pfgfr ihc intensity f c t þ cis bcg � , gem . g r none þ m c cis bcg � , mmc . none none þ m c ta bcg � . none none þ m c ta bcg � , mmc . none none þ m c ta bcg � , mmc, val . s c s c þ m aa t bcg � , mmc . none none þ m u cis bcg � , val . none none þ m c ta bcg � . none none þ m c ta bcg � , mmc . none none þ m c t bcg � . none none þ f c cis bcg � , mmc, val none none þ m c t þ cis bcg � , val . none none þ m c ta bcg � . s c ne þ abbreviations: aa, african-american; c, caucasian, f, female; gem, gemcitabine; m, male; mmc, mitomycin c; ne, not evaluable; u, unknown, val, valrubicin. dovitinib in bcg-unresponsive nmiuc with fgfr alterations www.aacrjournals.org clin cancer res; ( ) june , on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ biologically enriched population than intended. an amend- ment to cap the number of ihcþ mut� patient enrollment at patients was instituted; however, the trial was closed after enrolling only mutþ patients. in recent trial reports of other fgfr inhibitors (jnj- , bgj , azd ) in met- astatic patients with urothelial carcinoma, it now appears clear that activating fgfr mutations or fusions are required for tumor responses ( – ). in trials of these agents mandating either fgfr mutation or fusions, reduction of tumor size was observed in % to % of metastatic patients with urothelial carcinoma ( , ). interestingly, in a prior report of dovitinib in metastatic patients with urothelial carcinoma, no responses were observed among patients with fgfr mutations ( ). it is not clear whether differences in fgfr mutation testing methodology or individual drug fgfr -binding site properties explain discordant clinical activity. an observed cr of the mutþ patients treated with dovitinib in our trial is consistent with the more recent fgfr inhibitor results. with only mutþ patients enrolled, it is impossible for our study to provide any meaningful cis around the true cr rate. however, it is encouraging that the single cr patient has demonstrated a sustained remission out to þ months. furthermore, a strong trend in decreased posttreatment pfgfr staining was observed regardless of fgfr mutation status. in addition to patient selection limitations, the high rate of treatment-related toxicity led to frequent dose reductions including of the fgfr mutþ patients discontinuing dovi- tinib early. these dose modifications led to reduced dovitinib dose intensity in most patients and may have compromised antitumor effects. for future trials, particularly in the nmiuc population, our study provides a good example of the need to have a drug that is not only effective but also tolerable at therapeutic doses to impart true benefit. specifically, as in the case of dovitinib, the acceptance of relatively high rates of chronic toxicity in heavily pretreated metastatic solid tumor phase i trials may be greater than in nmiuc, given that nmiuc can be cured with cystectomy ( ). in future design of nmiuc trials, particular attention to high rates of acute or chronic grade – toxicities is warranted particularly if a drug will require chronic or lifelong administration to prevent tumor recurrence. in addition, perioperative complication rates from patients who proceed to posttreatment cystectomies are of critical importance in nmiuc trials, particularly when agents with known effects on bleeding and wound-healing such as fgfr or vegfr inhibitors table . all grade – adverse events and other adverse events occurring in more than % of patients adverse event grade grade grade grade constitutional fatigue ( %) ( %) ( %) ( %) pain ( %) ( %) ( %) ( % fall ( %) ( % ( %) ( %) other constitutional ( %) ( %) ( %) ( %) vascular hypertension ( %) ( %) ( %) ( %) headache ( %) ( %) ( %) ( %) intracranial hemorrhage ( %) ( % ( %) ( %) gastrointestinal gerd ( %) ( %) ( %) ( %) constipation ( %) ( %) ( %) ( %) diarrhea ( %) ( %) ( %) ( %) anorexia ( %) ( %) ( %) ( %) weight loss ( %) ( %) ( %) ( %) dysgeusia ( %) ( %) ( %) ( %) nausea/emesis ( %) ( %) ( %) ( %) emesis ( %) ( %) ( %) ( %) other gastrointestinal ( %) ( %) ( %) ( %) skin stomatitis ( %) ( %) ( %) ( %) rash ( %) ( %) ( %) ( %) hand–foot syndrome ( %) ( %) ( %) ( %) dry mouth ( %) ( %) ( %) ( %) other skin ( %) ( %) ( %) ( %) genitourinary bladder spasms ( %) ( %) ( %) ( %) other urinary ( %) ( %) ( %) ( %) infection fever ( %) ( %) ( %) ( %) infection ( %) ( %) ( %) ( %) pulmonary hoarseness ( %) ( %) ( %) ( %) other pulmonary ( %) ( %) ( %) ( %) musculoskeletal arthralgia/myalgia ( %) ( %) ( %) ( %) metabolic hypertriglyceridemia ( %) ( %) ( %) ( %) elevated alkaline phosphatase ( %) ( %) ( %) ( %) elevated ggt ( %) ( %) ( %) ( %) hypoalbuminemia ( %) ( %) ( %) ( %) elevated lipase ( %) ( %) ( %) ( %) other metabolic ( %) ( %) ( %) ( %) hematologic anemia ( %) ( %) ( %) ( %) table . tumor response to dovitinib treatment patient baseline t-stage tumor fgfr mutations duration of treatment, mo posttreatment t-stage response category t þ cis g r . ne nr cis none . cis nr ta none . ta nr ta none . ta nr ta s c . ne nr t none . t nr cis none . cis nr ta none . t nr ta none ta nr t none . t pd cis none . cis nr t þ cis none . t nr ta s c . t cr abbreviations: ne, not evaluable; nr, nonresponder; pd, progressive disease. hahn et al. clin cancer res; ( ) june , clinical cancer research on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ are studied. while no life-threatening perioperative complica- tions were observed in our trial, our sample size is insufficient to discount the possibility of such risks. despite the absent clinical activity, our study establishes several innovative principles in the design of nmiuc trials that should facilitate improved future clinical trial designs in this population. first, our study demonstrated the feasibility of tumor genomic testing as an eligibility requirement in the nmiuc population in a multisite setting. in fact, of the patients screened, ( %) had sufficient tumor available for fgfr mutation testing. while investigation of oral kinase inhibitors in patients with nmiuc has been pursued by other investigators, to our knowledge, our trial is the first to be undertaken in a molecularly enriched nmiuc population ( ). with the establishment of intrinsic basal and luminal tumor subtypes from analysis of the cancer genome atlas (tcga) urothelial carcinoma samples, we expect an increased need for future urothelial carcinoma clinical trials to target specific genomically defined patient subsets ( ). our study demonstrates that, despite the small tumor samples obtained from standard-of-care turbt specimens, enrichment of nmiuc patient subsets based on molecular testing is pos- sible and should be pursued if scientific hypotheses warrant it. in addition, our results establish the frequency of fgfr mutations in the bcg-unresponsive nmiuc population at % ( of patients), a previously unknown benchmark. our a priori design assumption that the fgfr mutation rate in patients with bcg-unresponsive nmiuc would fall somewhere between the reported rates in low-grade nmiuc ( %) and muscle-invasive urothelial carcinoma ( %) . . . . . . . . total weak moderate strong baseline pos�reatment p er ce nt p fg fr po si tiv ity b y ih c pfgfr ihc intensity group figure . pre- and post-dovitinib pfgfr ihc results. , , l og d ov iti ni b tis su e co nc en tr at io n (n m ol /l ) patient study number dovitinib ic = nmol/l figure . posttreatment dovitinib tissue concentration. dovitinib in bcg-unresponsive nmiuc with fgfr alterations www.aacrjournals.org clin cancer res; ( ) june , on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ patients proved incorrect. our results suggest that bcg-unre- sponsive nmiuc more closely resembles muscle-invasive and metastatic urothelial carcinoma than a low-grade nmiuc pre- decessor tumor. the lower rate of fgfr mutations observed in the bcg-unresponsive nmiuc population has implications on future sample size considerations of fgfr -targeting trials in this population. importantly, our trial showed that oral administration of dovitinib unanimously achieved pharmacologically active urothelial tissue concentrations. this finding suggests that lack of clinical activity was related to drug toxicity and study population design issues rather than drug delivery failure. these results support further investigation of systemically administered agents in the nmiuc population. a caveat, how- ever, is the fact that the urothelial tissue bioavailability is not usually investigated or provided in preclinical testing data provided in investigator brochures of most novel cancer drugs. a high intact urinary excretion of drug can be reassuring of adequate urothelial tumor drug concentration exposure. how- ever, if urothelial tissue concentrations are critical in the deci- sion process to assess the effectiveness of systemic versus intravesical routes of drug administration, development of clinical pharmacology assays to measure urothelial tissue drug concentrations are strongly recommended. finally, our study demonstrates the importance of multispeci- alty investigator engagement in the conduct of early-stage urothe- lial carcinoma trials. at each participating center, a urologist, medical oncologist, and pathologist were identified to serve as local champions for the trial. while multidisciplinary teams in varying forms are often utilized in the administration of neoad- juvant cisplatin-based chemotherapy for muscle-invasive urothe- lial carcinoma, our study highlights the importance of also developing highly functional cross-discipline research collabora- tions in nmiuc patients. the need for urothelial carcinoma multispecialty research infrastructure is increasing in parallel with the rapid expansion of clinical trials being conducted in the muscle-invasive adjuvant, neoadjuvant, and bcg-unresponsive nmiuc populations. in summary, our study firmly establishes that pfgfr ihc alone should not be used as a solitary qualifying criteria for enrollment in future urothelial carcinoma trials of fgfr kinase inhibitors. in addition, the unfavorable toxicity profile of dovi- tinib precludes further development in the nmiuc population. however, antitumor activity consistent with other reports in fgfr mutþ patients was observed further implying fgfr as a viable therapeutic target in urothelial carcinoma across all stages including nmiuc. the demonstration that genomic testing as an eligibility requirement in nmiuc patients is feasible and the detection of pharmacologically active dovitinib urothelial tissue concentrations by oral drug administration are novel findings with implications for future nmiuc trial designs. disclosure of potential conflicts of interest no potential conflicts of interest were disclosed. authors' contributions conception and design: n.m. hahn, m.o. koch, r.s. foster, z. liu, g.j. tsongalis development of methodology: n.m. hahn, g.j. netto, d.r. jones, g.e. san- dusky, t. breen, z. liu, s.a. turner, f.b. de abreu, g.j. tsongalis, r.e. greenberg acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): n.m. hahn, t.j. bivalacqua, a.e. ross, g.j. netto, a.s. baras, j.c. park, c. chapman, t.a. masterson, m.o. koch, r. bihrle, r.s. foster, t.a. gardner, l. cheng, t. breen, c. albany, m.l. moore, a. reed, f.b. de abreu, t.l. gallagher, g.j. tsongalis, e.r. plimack, r.e. greenberg analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): n.m. hahn, t.j. bivalacqua, g.j. netto, j.c. park, r.s. foster, l. cheng, d.r. jones, k. mcelyea, g.e. sandusky, t. breen, z. liu, c. albany, s.a. turner, f.b. de abreu, e.r. plimack writing, review, and/or revision of the manuscript: n.m. hahn, t.j. bivalac- qua, a.e. ross, g.j. netto, r.s. foster, a.s. baras, t.a. gardner, l. cheng, d.r. jones, g.e. sandusky, z. liu, c. albany, f.b. de abreu, g.j. tsongalis, e.r. plimack, r.e. greenberg, d.m. geynisman administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): n.m. hahn, a.s. baras, k. mcelyea, s.a. turner, g.j. tsongalis study supervision: n.m. hahn, t.a. masterson, r.a. loman, r.e. greenberg grant support this study was supported by grant funding from investigator-initiated funds from novartis and nci cancer center support grant funding (p ca ). the costs of publication of this article were defrayed in part by the payment of page charges. this article must therefore be hereby marked advertisement in accordance with u.s.c. section solely to indicate this fact. received september , ; revised november , ; accepted novem- ber , ; published onlinefirst december , . a b figure . dovitinib pfgfr ihc pathology samples. a, patient – baseline. b, patient – posttreatment cycle day . hahn et al. clin cancer res; ( ) june , clinical cancer research on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ references . siegel rl, miller kd, jemal a. cancer statistics, . ca cancer j clin ; : – . . ries l, harkins d, krapcho m, mariotto a, miller b, feuer e, et al. seer cancer statistics review, – . bethesda, md: national cancer insti- tute; . . botteman mf, pashos cl, redaelli a, laskin b, hauser r. the health economics of bladder cancer: a comprehensive review of the published literature. pharmacoeconomics ; : – . . shelley md, court jb, kynaston h, wilt tj, fish rg, mason m. intravesical bacillus calmette-guerin in ta and t bladder cancer. cochrane database syst rev :cd . . sylvester rj, van der ma, lamm dl. intravesical bacillus calmette-guerin reduces the risk of progression in patients with superficial bladder cancer: a meta-analysis of the published results of randomized clinical trials. j urol ; : – . . lamm dl, blumenstein ba, crissman jd, montie je, gottesman je, lowe ba, et al. maintenance bacillus calmette-guerin immunotherapy for recurrent ta, t and carcinoma in situ transitional cell carcinoma of the bladder: a randomized southwest oncology group study. j urol ; : – . . herr hw. how to manage patients who fail intravesical bcg therapy. in: proceedings of the genitourinary cancers symposium; orlando, fl; feb ; . . lerner sp,dinney c, kamat a,bivalacqua tj, nielsen m,o'donnellm,etal. clarification of bladder cancer disease states following treatment of patients with intravesical bcg. bladder cancer ; : – . . dalbagni g, russo p, bochner b, ben-porat l, sheinfeld j, sogani p, et al. phase ii trial of intravesical gemcitabine in bacille calmette-gu�erin-refrac- tory transitional cell carcinoma of the bladder. j clin oncol ; : – . . steinberg g, bahnson r, brosman s, middleton r, wajsman z, wehle m. efficacyand safetyofvalrubicinfor the treatment ofbacilluscalmette-guerin refractory carcinoma in situ of the bladder. the valrubicin study group. [erratum appears in j urol jan; ( ): ]. j urol ; : – . . babjuk m, burger m, zigeuner r, shariat sf, van rhijn bw, comperat e, et al. eau guidelines on non-muscle-invasive urothelial carcinoma of the bladder: update . eur urol ; : – . . folkman j. tumor angiogenesis: therapeutic implications. n engl j med ; : – . . kerbel r, folkman j. clinical translation of angiogenesis inhibitors. nat rev cancer ; : – . . fauconnet s, bernardini s, lascombe i, boiteux g, clairotte a, monnien f, et al. expression analysis of vegf-a and vegf-b: relationship with clin- icopathological parameters in bladder cancer. oncol rep ; : – . . swellam m, el-aal aa. correlation between tissue and released vegf levels in urine of bladder cancer patients. am j biochem biotechnol ; : – . . yang cc, chu kc, yeh wm. the expression of vascular endothelial growth factor in transitional cell carcinoma of urinary bladder is correlated with cancer progression. urol oncol ; : – . . balar av, apolo ab, ostrovnaya i, mironov s, iasonos a, trout a, et al. phase ii study of gemcitabine, carboplatin, and bevacizumab in patients with advanced unresectable or metastatic urothelial cancer. j clin oncol ; : – . . hahn nm, stadler wm, zon rt, et al. phase ii trial of cisplatin, gemcita- bine, and bevacizumab as first-line therapy for metastatic urothelial carcinoma: hoosier oncology group gu - . j clin oncol ; : – . . mitra ap, datar rh, cote rj. molecular pathways in invasive bladder cancer: new insights into mechanisms, progression, and target identifica- tion. j clin oncol ; : – . . wu xr. urothelial tumorigenesis: a tale of divergent pathways. nat rev cancer ; : – . . knowles ma. novel therapeutic targets in bladder cancer: mutation and expression of fgf receptors (report). future oncol ; : . . tomlinson dc, baldo o, harnden p, knowles ma. fgfr protein expres- sion and its relationship to mutation status and prognostic variables in bladder cancer. j pathol ; : – . . porta c, giglione p, liguigli w, paglino c. dovitinib (chir , tki ): structure, development and preclinical and clinical activity. future oncol (london, england) ; : – . . powers c, mcleskey s, wellstein a. fibroblast growth factors, their recep- tors and signaling. endocr relat cancer ; : – . . casanovas o, hicklin dj, bergers g, hanahan d. drug resistance by evasion of antiangiogenic targeting of vegf signaling in late-stage pan- creatic islet tumors. cancer cell ; : – . . rosenberg je, hoffman-censits j, powles t, van der heijden ms, balar av, necchi a, et al. atezolizumab in patients with locally advanced and metastatic urothelial carcinoma who have progressed following treatment with platinum-based chemotherapy: a single-arm, multicentre, phase trial. lancet ; : – . . pal sk, rosenberg je, keam b, wolf j, berger r, dittrich c, et al. efficacy of bgj , a fibroblast growth factor receptor (fgfr) - inhibitor, in patients (pts) with previously treated advanced/metastatic urothelial carcinoma (muc) with fgfr alterations. j clin oncol , (suppl; abstr ). . rodriguez-vida a, saggese m, hughes s, rudman s, chowdhury s, smith nr, et al. complexity of fgfr signalling in metastatic urothelial cancer. j hematol oncol ; : . . tabernero j, bahleda r, dienstmann r, infante jr, mita a, italiano a, et al. phase i dose-escalation study of jnj- , an oral pan–fibroblast growth factor receptor inhibitor, in patients with advanced solid tumors. j clin oncol ; : – . . milowsky mi, dittrich c, dur�an i, jagdev s, millard fe, sweeney cj, et al. phase trial of dovitinib in patients with progressive fgfr -mutated or fgfr wild-type advanced urothelial carcinoma. eur j cancer ; : – . . sch€afer n, gielen gh, kebir s, wieland a, till a, mack f, et al. phase i trial of dovitinib (tki ) in recurrent glioblastoma. j cancer res clin oncol ; : – . . helfand am, lee ct, hafez k, hussain m, liebert m, daignault s, et al. phase ii clinical trial of intravesical bacillus calmette-guerin (bcg) followed by sunitinib for the treatment of high-risk nonmuscle-invasive bladder cancer (nmibc). j clin oncol , (suppl ; abstr ) . the cancer genome atlas research n. comprehensive molecular charac- terization of urothelial bladder carcinoma. nature ; : – . www.aacrjournals.org clin cancer res; ( ) june , dovitinib in bcg-unresponsive nmiuc with fgfr alterations on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/ ; : - . published onlinefirst december , .clin cancer res noah m. hahn, trinity j. bivalacqua, ashley e. ross, et al. cancer research network trial hcrn - mutations or overexpression: hoosierfgfr carcinoma with a phase ii trial of dovitinib in bcg-unresponsive urothelial updated version . / - .ccr- - doi: access the most recent version of this article at: material supplementary http://clincancerres.aacrjournals.org/content/suppl/ / / / - .ccr- - .dc access the most recent supplemental material at: cited articles http://clincancerres.aacrjournals.org/content/ / / .full#ref-list- this article cites articles, of which you can access for free at: e-mail alerts related to this article or journal.sign up to receive free email-alerts subscriptions reprints and .pubs@aacr.org to order reprints of this article or to subscribe to the journal, contact the aacr publications department at permissions rightslink site. click on "request permissions" which will take you to the copyright clearance center's (ccc) .http://clincancerres.aacrjournals.org/content/ / / to request permission to re-use all or part of this article, use this link on april , . © american association for cancer research. clincancerres.aacrjournals.org downloaded from published onlinefirst december , ; doi: . / - .ccr- - http://clincancerres.aacrjournals.org/lookup/doi/ . / - .ccr- - http://clincancerres.aacrjournals.org/content/suppl/ / / / - .ccr- - .dc http://clincancerres.aacrjournals.org/content/ / / .full#ref-list- http://clincancerres.aacrjournals.org/cgi/alerts mailto:pubs@aacr.org http://clincancerres.aacrjournals.org/content/ / / http://clincancerres.aacrjournals.org/ << /ascii encodepages false /allowtransparency false /autopositionepsfiles true /autorotatepages /none /binding /left /calgrayprofile (gray gamma . ) /calrgbprofile (srgb iec - . ) /calcmykprofile (u.s. web coated \ swop\ v ) /srgbprofile (srgb iec - . ) /cannotembedfontpolicy /error /compatibilitylevel . /compressobjects /off /compresspages true /convertimagestoindexed true /passthroughjpegimages true /createjobticket false /defaultrenderingintent /default /detectblends true /detectcurves . /colorconversionstrategy /leavecolorunchanged /dothumbnails false /embedallfonts true /embedopentype false /parseiccprofilesincomments true /embedjoboptions true /dscreportinglevel /emitdscwarnings true /endpage - /imagememory /lockdistillerparams true /maxsubsetpct /optimize false /opm /parsedsccomments true /parsedsccommentsfordocinfo true /preservecopypage false /preservedicmykvalues true /preserveepsinfo true /preserveflatness false /preservehalftoneinfo false /preserveopicomments false /preserveoverprintsettings true /startpage /subsetfonts true /transferfunctioninfo /remove /ucrandbginfo /remove /useprologue false /colorsettingsfile () /alwaysembed [ true ] /neverembed [ true ] /antialiascolorimages false /cropcolorimages false /colorimageminresolution /colorimageminresolutionpolicy /warning /downsamplecolorimages true /colorimagedownsampletype /bicubic /colorimageresolution /colorimagedepth - /colorimagemindownsampledepth /colorimagedownsamplethreshold . /encodecolorimages true /colorimagefilter /dctencode /autofiltercolorimages true /colorimageautofilterstrategy /jpeg /coloracsimagedict << /qfactor . /hsamples [ ] /vsamples [ ] >> /colorimagedict << /qfactor . /hsamples [ ] /vsamples [ ] >> /jpeg coloracsimagedict << /tilewidth /tileheight /quality >> /jpeg colorimagedict << /tilewidth /tileheight /quality >> /antialiasgrayimages false /cropgrayimages false /grayimageminresolution /grayimageminresolutionpolicy /warning /downsamplegrayimages true /grayimagedownsampletype /bicubic /grayimageresolution /grayimagedepth - /grayimagemindownsampledepth /grayimagedownsamplethreshold . /encodegrayimages true /grayimagefilter /dctencode /autofiltergrayimages true /grayimageautofilterstrategy /jpeg /grayacsimagedict << /qfactor . /hsamples [ ] /vsamples [ ] >> /grayimagedict << /qfactor . /hsamples [ ] /vsamples [ ] >> /jpeg grayacsimagedict << /tilewidth /tileheight /quality >> /jpeg grayimagedict << /tilewidth /tileheight /quality >> /antialiasmonoimages false /cropmonoimages false /monoimageminresolution /monoimageminresolutionpolicy /warning /downsamplemonoimages true /monoimagedownsampletype /bicubic /monoimageresolution /monoimagedepth - /monoimagedownsamplethreshold . /encodemonoimages true /monoimagefilter /ccittfaxencode /monoimagedict << /k - >> /allowpsxobjects false /checkcompliance [ /none ] /pdfx acheck false /pdfx check false /pdfxcompliantpdfonly false /pdfxnotrimboxerror true /pdfxtrimboxtomediaboxoffset [ . . . . ] /pdfxsetbleedboxtomediabox true /pdfxbleedboxtotrimboxoffset [ . . . . ] /pdfxoutputintentprofile (none) /pdfxoutputconditionidentifier () /pdfxoutputcondition () /pdfxregistryname () /pdfxtrapped /false /createjdffile false /description << /enu ([based on '[high quality print]'] use these settings to create adobe pdf documents for quality printing on desktop printers and proofers. created pdf documents can be opened with acrobat and adobe reader . and later.) >> /namespace [ (adobe) (common) ( . ) ] /othernamespaces [ << /asreaderspreads false /cropimagestoframes false /errorcontrol /warnandcontinue /flattenerignorespreadoverrides true /includeguidesgrids false /includenonprinting false /includeslug false /namespace [ (adobe) (indesign) ( . ) ] /omitplacedbitmaps false /omitplacedeps false /omitplacedpdf false /simulateoverprint /legacy >> << /addbleedmarks false /addcolorbars false /addcropmarks true /addpageinfo false /addregmarks false /bleedoffset [ ] /convertcolors /noconversion /destinationprofilename () /destinationprofileselector /na /downsample bitimages true /flattenerpreset << /clipcomplexregions true /convertstrokestooutlines false /converttexttooutlines false /gradientresolution /linearttextresolution /presetname ([high resolution]) /presetselector /highresolution /rastervectorbalance >> /formelements false /generatestructure false /includebookmarks false /includehyperlinks false /includeinteractive false /includelayers false /includeprofiles false /marksoffset /marksweight . /multimediahandling /useobjectsettings /namespace [ (adobe) (creativesuite) ( . ) ] /pdfxoutputintentprofileselector /na /pagemarksfile /romandefault /preserveediting true /untaggedcmykhandling /leaveuntagged /untaggedrgbhandling /leaveuntagged /usedocumentbleed false >> << /allowimagebreaks true /allowtablebreaks true /expandpage false /honorbaseurl true /honorrollovereffect false /ignorehtmlpagebreaks false /includeheaderfooter false /marginoffset [ ] /metadataauthor () /metadatakeywords () /metadatasubject () /metadatatitle () /metricpagesize [ ] /metricunit /inch /mobilecompatible /namespace [ (adobe) (golive) ( . ) ] /openzoomtohtmlfontsize false /pageorientation /portrait /removebackground false /shrinkcontent true /treatcolorsas /mainmonitorcolors /useembeddedprofiles false /usehtmltitleasmetadata true >> ] >> setdistillerparams << /hwresolution [ ] /pagesize [ . . ] >> setpagedevice [pdf] dynamics of chromosome number and genome size variation in a cytogenetically variable sedge (carex scoparia var. scoparia, cyperaceae). | semantic scholar skip to search formskip to main content> semantic scholar's logo search sign increate free account you are currently offline. some features of the site may not work correctly. 	doi: . /ajb. corpus id: dynamics of chromosome number and genome size variation in a cytogenetically variable sedge (carex scoparia var. scoparia, cyperaceae). @article{chung dynamicsoc, title={dynamics of chromosome number and genome size variation in a cytogenetically variable sedge (carex scoparia var. scoparia, cyperaceae).}, author={kyong-sook chung and j. weber and a. hipp}, journal={american journal of botany}, year={ }, volume={ }, pages={ - } } 	kyong-sook chung, j. weber, a. hipp 	published 	biology, medicine 	american journal of botany premise of the study high intraspecific cytogenetic variation in the sedge genus carex (cyperaceae) is hypothesized to be due to the "diffuse" or non-localized centromeres, which facilitate chromosome fission and fusion. if chromosome number changes are dominated by fission and fusion, then chromosome evolution will result primarily in changes in the potential for recombination among populations. chromosome duplications, on the other hand, entail consequent opportunities for divergent evolution… expand view on pubmed systematics.mortonarb.org save to library create alert cite launch research feed share this paper citationshighly influential citations background citations methods citations results citations view all figures, tables, and topics from this paper 	 figure 	 table 	 figure 	 figure 	 figure view all figures & tables 	scoparia <angiosperm> 	carex genus 	cyperaceae 	genome size 	chromosome pairing 	gene duplication abnormality 	angiosperms 	largest 	flow cytometry 	ninety nine 	centromere 	dna sequence rearrangement 	kochia scoparia ab.iga:acnc:pt:ser:qn citations citation type citation type 	all types 	cites results 	cites methods 	cites background has pdf publication type author more filters more filters filters sort by relevance sort by most influenced papers sort by citation count sort by recency species coherence in the face of karyotype diversification in holocentric organisms: the case of a cytogenetically variable sedge (carex scoparia, cyperaceae). 	m. escudero, j. weber, a. hipp 	biology, medicine 	annals of botany 	 	 	pdf 	view excerpts, cites methods and background save alert research feed selection and inertia in the evolution of holocentric chromosomes in sedges (carex, cyperaceae). 	m. escudero, a. hipp, t. hansen, k. l. voje, m. luceño 	biology, medicine 	the new phytologist 	 	 	 save alert research feed chromosomal rearrangements in holocentric organisms lead to reproductive isolation by hybrid dysfunction: the correlation between karyotype rearrangements and germination rates in sedges. 	m. escudero, marlene hahn, b. brown, kate lueders, a. hipp 	biology, medicine 	american journal of botany 	 	 	 save alert research feed genome size stability despite high chromosome number variation in carex gr. laevigata. 	m. escudero, e. maguilla, j. loureiro, m. castro, s. castro, m. luceño 	biology, medicine 	american journal of botany 	 	 	 save alert research feed chromosome number evolves independently of genome size in a clade with nonlocalized centromeres (carex: cyperaceae) 	kyong-sook chung, a. hipp, e. h. roalson 	biology, medicine 	evolution; international journal of organic evolution 	 	 	 	view excerpts, cites background and methods save alert research feed chromosomes tell half of the story: the correlation between karyotype rearrangements and genetic diversity in sedges, a group with holocentric chromosomes 	a. hipp, p. e. rothrock, r. whitkus, j. weber 	biology, medicine 	molecular ecology 	 	 	pdf save alert research feed analysis of retrotransposon abundance, diversity and distribution in holocentric eleocharis (cyperaceae) genomes 	t. b. de souza, s. chaluvadi, + authors a. l. l. vanzela 	biology, medicine 	annals of botany 	 	 	pdf 	view excerpt, cites background save alert research feed karyotype diversity and chromosomal organization of repetitive dna in tityus obscurus (scorpiones, buthidae) 	bruno rafael ribeiro de almeida, susana suely rodrigues milhomem-paixão, + authors j. pieczarka 	biology, medicine 	bmc genetics 	 	 	 save alert research feed karyotype diversity and evolutionary trends in angiosperms 	h. weiss-schneeweiss, g. schneeweiss 	biology 	 	 save alert research feed meiotic chromosome numbers of five carex taxa in korea (cyperaceae) 	kyong-sook chung, hyoung-tak im 	biology 	 	 	pdf 	view excerpts, cites background save alert research feed ... ... references showing - of references sort byrelevance most influenced papers recency the evolution of chromosome arrangements in carex (cyperaceae) 	a. hipp, p. e. rothrock, e. h. roalson 	biology 	the botanical review 	 	 	pdf save alert research feed nonuniform processes of chromosome evolution in sedges (carex: cyperaceae) 	a. hipp 	biology, medicine 	evolution; international journal of organic evolution 	 	 	pdf save alert research feed chromosomal evolution in genus carex as viewed from nuclear dna content, with special reference to its aneuploidy 	k. nishikawa, y. furuta, k. ishitobi 	biology 	 	 	 save alert research feed chromosomes tell half of the story: the correlation between karyotype rearrangements and genetic diversity in sedges, a group with holocentric chromosomes 	a. hipp, p. e. rothrock, r. whitkus, j. weber 	biology, medicine 	molecular ecology 	 	 	pdf save alert research feed cytotaxonomic studies in iberian, balearic, north african, and macaronesian species of carex (cyperaceae). ii 	m. luceño 	biology 	 	 save alert research feed genome size and karyotype evolution in the slipper orchids (cypripedioideae: orchidaceae). 	a. cox, g. abdelnour, m. bennett, i. leitch 	biology, medicine 	american journal of botany 	 	 save alert research feed genome size in hieracium subgenus hieracium (asteraceae) is strongly correlated with major phylogenetic groups. 	j. chrtek, j. zahradníček, k. krak, j. fehrer 	medicine, biology 	annals of botany 	 	 	pdf save alert research feed chromosome studies in the genus carex, with special reference to aneuploidy and polyploidy 	n. tanaka 	biology 	 	 	 save alert research feed chromosome evolution in cyperales 	e. h. roalson, a. mccubbin, r. whitkus 	biology 	 	 	 save alert research feed speciation by monobrachial centric fusions ( chromosomes / reproductive isolation / cryptic species / population genetics 	r. j. baker, john w. bickhamt 	 	 	pdf save alert research feed ... ... related papers 	abstract 	figures, tables, and topics 	 citations 	 references 	related papers stay connected with semantic scholar sign up about semantic scholar semantic scholar is a free, ai-powered research tool for scientific literature, based at the allen institute for ai. learn more → resources datasetssupp.aiapiopen corpus organization about usresearchpublishing partnersdata partners   faqcontact proudly built by ai with the help of our collaborators terms of service•privacy policy the allen institute for ai by clicking accept or continuing to use the site, you agree to the terms outlined in our privacy policy, terms of service, and dataset license accept & continue amid- - - .. fire and human history of a barren-forest mosaic in southern indiana richard p. guyette, daniel c. dey and michael c. stambaugh u.s. forest service, north central research station, abnr bldg., university of missouri-columbia, missouri . telephone ( ) - ext. ; fax ( ) - ; e-mail: ddey@fs.fed.us department of forestry, abnr bldg., university of missouri-columbia, missouri . telephone ( ) - ; fax ( ) - ; e-mail: stambaughm@missouri.edu abstract.—the purpose of this paper is to provide quantitative fire history information from a historically unique region, the oak barrens of the interior low plateau ecoregion. we sampled post oak (quercus stellata wangenh.) trees from the boone creek watershed in southern indiana. the period of tree-ring record ranged in calendar years from to and fire scar dates (n ¼ ) ranged from to . the mean fire interval for the period to was . y and individual fire intervals ranged from to y. the average percentage of trees scarred at the site was % or about in trees sampled. no significant relationship was identified between fire years and drought conditions however, variability in the fire record coincided with native american migrations and euro-american settlement periods. temporal variability in the fire record illustrates not only the dynamic nature of anthropogenic fire regimes but also the importance of humans in culturing presettlement barrens communities. introduction during the period of euro-american settlement of the united states the term ‘‘barrens’’ was used by land surveyors and others to describe areas of land that had no or sparse tree cover (bourne, ; englemann, ; sauer, ; braun, ). the term was used loosely for a variety of forest types (e.g., pine barrens of new jersey and northern lake states, oak barrens of the midwest) that occurred on a wide range of soils, geology and with variable climate (tyndall, ). early land descriptions distinguished barrens vegetation from prairies and forests. botanical manuals and land classification descriptions list barrens as a habitat type that is dominated by drought tolerant plants (lounsberry, ; gray, ; homoya, ), and many of the species that inhabit barrens are representative of the substrate type (i.e., chert, limestone, sandstone, siltstone, gravel, clay and sand) (bacone et al., ; homoya, ). oak barrens were common throughout the eastern united states (bartgis, ; heikens and robertson, ; heikens et al., ; homoya, ), particularly in the interior low plateau ecoregion (bailey, ). the ‘‘big barrens’’ region, estimated to cover , to , km (mcinteer, ) and located primarily on the mississippian limestone karst plain of kentucky and tennessee (baskin et al., ; deselm, ), was one of the largest barrens at the time of european settlement. in indiana, presettlement barren acreage was estimated to be approximately , ha (homoya et al., ). however, the aerial extent of barrens has decreased dramatically since european settlement because of: ( ) conversion to agriculture, ( ) urban development (hutchison, ) and ( ) reduction in the widespread use of fire (robertson and heikens, ). in addition, the replacement of the present address: department of forestry, abnr bldg., university of missouri-columbia, missouri . telephone ( ) - ; fax ( ) - ; e-mail: guyetter@missouri.edu am. midl. nat. : – term barrens with terms such as glade, savanna and woodland (baskin et al., ; heikens and robertson, ) has caused problems in assessing the aerial extent of past and present barrens. olson ( ) emphasized the importance of fire in maintaining the openness of forests in the central hardwood region of the united states based on a synthesis of general land office (glo) surveyor notes, early travelers’ descriptions of the vegetation composition and characteristics and accounts of native american burning practices. many authors have also concluded that burning was an important factor in creating and maintaining barrens in the eastern united states (sauer, ; gleason, ; mcinteer, ; cottam, ; bielmann and brenner, ; schroeder, ; anderson and brown, ; bacone and post, ; anderson and schwegman, ; tyndall, , ; mcclain et al., ). it is likely that anthropogenic ignitions were the primary source of fire in the barrens region because of the extremely low frequency of lightning caused fires (schroeder and buck, ) due largely to heavy rains that accompany storms in this region. the purpose of this study is to quantify the fire history of an oak barren site in the interior low plateau ecoregion (fig. ) and relate the variability in fire frequency to historic periods of drought and changes in anthropogenic ignitions. knowledge of the fire history of barrens is important for understanding the importance of fire in creating and maintaining barrens communities. furthermore, understanding the historic fire regime provides an ecological basis for using prescribed fire to mimic historic fire regimes and restore barrens. this study contributes to our understanding of the historical role of fire in the eastern united states— a region where knowledge of the spatial and temporal dynamics of fire is limited. materials and methods the study site is located in the boone creek watershed on the tell city ranger district of the hoosier national forest (fig. ). the boone creek watershed is approximately km north of the ohio river in perry county, indiana ( n, w). the site is underlain by limestone and characterized by rolling topography with slopes and elevation ranging from to and to m, respectively. barrens occupy about % of the area with barren canopy closure ranging from – %. currently, the boone creek area is largely forested with canopy closure of about – % (usda, ). oaks and hickories are the most important species in the overstory tree layer, are widely spaced and rarely exceed m in height. grasses, forbs and small trees dominate the surface and understory vegetation. comparisons of old ( s– s) with new aerial photographs of the boone creek area reveal increases in canopy closure and woody vegetation in previous barren openings; both attributed to fire suppression. additionally, fire intolerant tree species such as sugar maple (acer saccaharum marsh.) and american beech (fagus grandifolia ehrh.) have recently increased in importance in the understory vegetation layer. several prescribed fires were conducted in the vicinity of the study site during the s including two landscape level burns (approximately ha) and two smaller burns of barren openings (steven olson, pers. commu). field sampling.—during october of wedges from post oak (quercus stellata wangenh.) trees and snags were collected for the purpose of reconstructing the fire history of the site. wedges were taken exclusively from a ha area in the boone creek watershed. wedges, . to . cm thick, were cut from the tree bole within cm of ground level and from the side of the tree that showed external fire scar evidence. we sampled post oaks because the species is long-lived, highly resistant to heart rot, and distributed throughout the study area. sample trees were selected based on the presence of fire scars, their age and landscape position. the majority of trees were sampled on open southern aspects (barrens) the american midland naturalist ( ) high on the landscape (i.e., shoulder and upper slope positions), however sample aspect ranged from – with some representation from lower slope positions. we sampled both young and old trees in order to minimize possible age related scarring biases (due to the effect of tree mass and bark thickness) on the homogeneity of our fire event chronology. aspect and slope of tree locations were marked on wedges in the field and then transferred to a database. all samples collected were used in creating the fire chronology. tree-ring and fire scar dating.—annual growth increments and fire scars were dated using fig. .—study site is in perry county on the southern border of indiana and within the interior lowland plateau. the boone creek barrens area is currently managed by the u.s.d.a. forest service, hoosier national forest guyette et al.: barren-forest mosaic dendrochronology, a method that uses tree-ring variation caused by climate to date annual growth (stokes and smiley, ; guyette and cutter, ). samples were prepared by sanding their surface to a high polish (e.g., grit sandpaper). annual rings were measured to an accuracy of . mm using a moving stage with an electronic transducer and binocular microscope. tree-ring measurements were imported into cofecha (holmes et al., ), a program that checks the accuracy of dating and aids in quality control of measurements. fire scars were then identified following criteria by smith and sutherland ( ) and assigned to the calendar year of the first year of growth response to the fire injury (e.g., callus tissue, cambial death). we determined the diameter of the tree when scarred by measuring the mean radial distance from the scar to pith and multiplying by two. using fhx software (grissino-mayer, ), we developed the fire chronology and analyzed fire scar years. we computed a mean fire interval and descriptive statistics for both the composite and individual tree mean fire intervals, constructed a composite graph of fire scarring years and tested the actual data frequency distribution using a kolmogorov-smirnov goodness-of-fit test and weibull distribution. individual tree mean fire intervals were grouped into cultural periods (e.g. ( ) native american (pre ), ( ) native american and european transition ( to ), ( ) european grazing ( to ), ( ) fire suppression (post )/prescribed fire ( to ). when intervals spanned multiple cultural periods they were assigned to the period in which the most years of the mean fire interval record represented. the fire intervals and statistical information in this study are based on all fire scars within the ha study area and, therefore, the mean fire interval is defined as the occurrence of fire somewhere in the study site. since some fire injuries are not readily distinguishable from injuries caused by other factors (e.g., logging) we used characteristics common to fire scars (e.g., size, shape, orientation, location on tree bole, presence of charcoal, synchrony of scarring among trees and related ring-width response) and fire scar characteristics from known prescribed fires that occurred at the study site during the s to ensure the accurate identification of historic fire scar years. comparisons and similarities between fire scar characteristics of the prescribed burn and fire scar characteristics before the prescribed burn helped to ensure all scars in our sample were fire-related injuries. in addition, fire scar characteristics were compared to those studied and documented from post oaks in other regions (guyette and cutter, ; cutter and guyette, ). low merchantability and value of barrens trees due to species, bole form, defects and soundness likely inhibited logging therefore, the chance of tree scarring by logging operations is probably low. the lack of stumps and the old age of many of the trees at the study site suggest that logging has been infrequent and has not occurred recently. fire scarring biases.—each tree has a unique fire scarring potential that is based on tree and site characteristics such as species, tree size, bark thickness, previous fire injury, fuel near the bole, thermal xylem characteristics and landscape position (cutter and guyette, ; guyette and cutter, ). we sampled young and old post oaks to minimize scarring biases due to tree size and age. post oak is one of the most fire resistant deciduous tree species in the midwest, thus making determinations of the historic fire frequency difficult because of the relatively low probability that surface fires have scarred an individual tree. for example, in a study of external scarring from surface fires in the ozark region of missouri, paulsell ( ) found that % of post oaks were scarred in annual burns while periodic burns scarred %. fire and drought event analysis.—palmer drought severity index (pdsi) reconstructions were used to analyze the potential relationship between fire years and historic drought the american midland naturalist ( ) conditions (cook et al., ). the four nearest pdsi grid points in each of the cardinal directions from the study site were averaged to calculate a mean regional pdsi for the study site. we used pdsi reconstructions from tree-ring records spanning the period of years to . t-tests were used to determine differences between the mean pdsi for periods in the fire history that were absent of fire and those with frequent fire. fire years before were not included in the analysis due to the absence of pdsi reconstruction data. results composite fire chronology.—the -y tree-ring record spanned the period of to and fire-scar dates ranged in calendar years from to (fig. ). eighty-four fire scars from trees were dated yielding fire intervals ( fire years). the composite fire chronology was characterized by high variability in fire frequency with a century of no fire and a decade of near annual burning ( to , fig. ). the mean fire interval for the period to was . y and fire intervals ranged from to y. the mean fire interval for the early half of the fire chronology ( to ) was . y with a range in fire intervals from to y. in contrast, fires were more frequent and occurred more regularly in the latter half of the fire chronology ( to ) where the mean fire interval was . y and fire intervals ranged from to y. the greatest number of fires scars per year of tree growth occurred from to ; however, our sample size is small in the early period. no fire scars were recorded after this period until the early s. years of relatively large fires, calculated as the percentage of trees scarred, were , , and . with the exception of the to period (low sample size, fig. ), the highest percentage of sample trees scarred in any fire year was % ( ). the average percentage of sampled trees scarred at the site during all fire years was . % or about in post oak trees. we examined the relationship between the number of fire years per decade and the number of sample trees to address the hypothesis that number of fire years is not a function of the number of sample trees and found that the number of sample trees may explain % of the variability in fire years per decade. however, this relationship was not statistically significant, likely because of auto correlation in the time series of fire years per decade. individual tree fire history.—fifty-seven individual tree fire intervals resulted from the sample trees (fig. ). the mean and mode intervals were and y respectively. individual tree fire intervals ranged from to y and were positively skewed (skewness ¼ . ). a weibull distribution adequately fit the fire interval data and yielded a median interval of . y (fig. ). weibull % and % exceedance probabilities were . and . y and intervals , . and . . y were significantly short and long intervals respectively. fire scarring.—a histogram of tree basal diameter at the time of fire scarring resulted in a trend of decreasing scarring frequency with increasing tree diameter (fig. ). the majority of trees scarred were , cm in diameter and ranged from . to . cm with a mean of . cm (sd ¼ . ). on average, post oaks � cm showed a lower fire scar frequency than those trees , cm. the average number of years from the initial scar to the second scarring event was . y and ranged from to y. drought and fire.—pdsi values were similar for all time periods and categories compared (table ). there was no significant difference in pdsi during a -y period of no fire ( to ) and the period of highest fire frequency ( to ). few fires occurred during years that were drier than years of recent history (i.e., to y before fire year), particularly in fire years before (fig. ). pdsi values were not statistically different between years of no fire and fire years post- . no correlations between lag effects (e.g., t- , t- , t- ) of pdsi and fire years (kaye and swetnam, ) were significant. guyette et al.: barren-forest mosaic discussion interpretation of the fire scar record.—woody plant encroachment and succession of barrens to closed-canopy forests was reported as early as the mid- s (chester, ). in the absence of fire, succession from barrens to closed canopy forests can occur rapidly (e.g., – y) (grimm, , usda, ). assuming a transition from a barren/open dry upland forest (e.g., historic conditions of the site) to closed canopy forest vegetation (e.g., current conditions of the site) could occur from y of fire suppression, potential for woody encroachment at the boone creek study site was high during the th century; a century comprised of a larger -y fire free interval ( to ). although occurring during a time of low sample size ( to trees), the -y fire free interval is likely an accurate representation of the fire regime. both small trees and previously scarred post oaks (both likely sensitive to scarring) occurred in this portion of the fire chronology. a second period of succession to forest occurred from to (usda, ), a period of low occurrence of fire that allowed fire intolerant woody plants to colonize the site. there is little evidence that historic fire events corresponded to droughts. this is fig. .—a plot of tree samples, their dated fire scars, and a composite fire chronology. the y-axis is calendar years and bold vertical bars represent fire scars. horizontal lines represent the period of record for each sample tree. pith dates are represented by short thin vertical lines at the left end of the samples while inside ring dates are represented by diagonal lines. a composite fire scar chronology with all fire years is given at the bottom of the plot the american midland naturalist ( ) evidenced by few fire years directly corresponding to drought conditions and significant changes in the fire regime being unrelated to pdsi variability. there is some evidence that fires before were more strongly correlated with drought conditions, however, this was not statistically significant. inhabitants of the region may have taken advantage of drought conditions when burning. regional inhabitants likely used drier than average conditions to modify the landscape with fire as burning during a drought would require less effort in spreading the fire (e.g., number of ignitions) and would likely be more effective in killing woody plants than burning during wetter conditions. as human population and the importance of anthropogenic ignitions decrease it is likely that the correlation between fire and drought would increase. the lack of evidence supporting a relationship between fire and drought and the presence of a temporal sequence of variability in the fire history suggest that ignitions were largely anthropogenic. the fire history spans five pyro-cultural periods: ( ) native american (pre- ), ( ) native american and european transition ( to ), ( ) european grazing ( to ), ( ) fire suppression ( to ) and ( ) prescribed fire ( to ) (fig. ). as has been documented elsewhere (guyette and dey, , , ; guyette et al., ), the sequence and abrupt changes in the frequency of fire at the study site suggest that a strong relationship exists between fire frequency and human population density, settlement and migration. the temporal variability in fire frequency can be explained by the movement of human populations near the boone creek watershed. the long and unusual fire interval in the s ( to ) at this site may be the result of native americans in the ohio river fig. .—a chronology of the influence of fire at the study site. the influence of fire (y axis) is the number of fire scars per decade divided by the number of years of record within a decade on all sample trees. bars may indicate that either many trees scarred in different years or many trees scarred in a single fire year. pyro-cultural periods (see text) are identified and represent possible changes in human-fire interactions guyette et al.: barren-forest mosaic valley migrating into arkansas and missouri. several authors (hudson, ; obrien and wood, ; vehik, ) point to archeological, linguistic and historic evidence suggesting that the quapaw and osage, who lived in missouri and arkansas from about to , had lived in the ohio river valley before european arrival. their migration out of the ohio river valley in the latter half of the s may have been a retreat from iroquois invaders, who had recently acquired european firearms (baird, ). waldman ( ) reported the ohio river valley having the lowest native american population density in eastern north america at the time of contact. nevertheless, considerable archaeological evidence of native american occupation in the ohio river valley and indiana exists for many periods before contact with europeans (kellar, ) thus humans cannot be discounted as a potential ignition source during early portions of the fire record. the return of frequent fire in the early s ( to ) was likely due to humans. native americans ceded parts of southern indiana to the united states in the treaties of and (kellar, ). westward migrating tribes, such as the shawnee and delaware, moved through southern indiana (anson, ; clark, ). in the early th century, europeans settled along the ohio river in southern indiana and population density and fire ignitions began to increase even in topographically rough areas such as the study site (figs. , ) (guyette and dey, ; guyette et al., ). between and indiana population increased from , to , , , a more than six-fold increase (stats indiana, ). population density reached a maximum of about people/km circa in indiana. an abrupt increase in the frequency of fire at the study site occurred between and compared to the previous y (fig. ). increases in human population in indiana between and are correlated with increases in fire years per decade (r ¼ fig. .—frequency of individual tree fire intervals for important pyro-cultural periods: native american (pre- ), native american and european transition ( to ), european grazing ( to ) and fire suppression/prescribed fire (post- ) the american midland naturalist ( ) . ) and the number of fire scars per decade (r ¼ . ), however these correlations are not statistically significant because of the short sample length and autocorrelation within the decadal human population data. landscape level burning in the late s and early s was followed by fire suppression beginning in the mid- s. during the most recent part of the fire event chronology the dates of fire scars are associated with the dates of prescribed fires. fire scar dates of and are associated with prescribed burns conducted in april of and by the hoosier national forest. prescribed fires also occurred in and in the vicinity of the study site but did not scar the study trees. we do not know if the prescribed fires of and actually burned in the study area. the fire was the first large ( ha) prescribed burn conducted in the boone creek watershed and it scarred % of the trees in our sample. despite our targeting of sample trees, the percentage of trees scarred at the study site in ( %) is similar to that documented for post oak trees ( %) scarred after prescribed fire in the ozark region of missouri (paulsell, ). the average percentage of trees scarred at the boone creek study site during all fire years was . % or about in post oak trees. this statistic may overestimate the percentage of trees scarred in a fire because we targeted trees that had external evidence of scarring. however, many trees that were scarred were not sampled because of extensive rot and many fire scarred trees may have died when young and thus were not accounted for in the sample population. topographic influences on the fire regime.—several pyro-topographic features may be relevant to the fire history of this site. although the site is located approximately km northwest of the ohio river (fig. ), the spread of anthropogenic fire from this major travel corridor to the study site would have been inhibited due to several conditions. no terraces or other potential agricultural sites are located on the side of the river adjacent to the study site, fig. .—histogram of the diameter of trees when they were fire-scarred indicating the greater likelihood of smaller diameter trees being scarred by a fire guyette et al.: barren-forest mosaic thereby decreasing the potential for human occupation and agricultural activity as a source of anthropogenic ignitions. secondly, if a fire were to initiate near the river, the steep step and bluff topography along the river would have inhibited the northward spread of fire because fire does not travel well over steep bluff terrain (anderson, ; batek et al., ). finally, if a fire were to reach the river’s bluff top, it would need to spread down slope to enter the study site, thus further minimizing its potential to spread and scar trees. the highly dissected topography of the surrounding landscape also likely inhibited the spread of fires into the study site, particularly from the north, west and south. isolation of the site from fire spreading from other areas makes the fire history all that more responsive to changes in local human populations and land uses. fire was undoubtedly an important disturbance agent in the maintenance of the boone creek barrens site. fire scars from the period of to resulted in a mean fire interval of . y and we attribute the variability in fire intervals ( to y) within this period to changes in human populations. today, many extant barrens (e.g., boone creek barrens) are small isolated remnants on xeric sites with rocky, shallow or infertile soils (bacone et al., ; mcinteer, ). these remnant barrens remnants were probably core areas in what were much larger barrens complexes. core barrens areas are important in barren restoration because they have relic populations of barren dependent species and they can be used to target areas on the landscape for barrens restoration. barrens restoration.—prescribed burning can be an effective technique for barren restoration (anderson and brown, ; bacone and post, ; wade and menges, ; stritch, ). where available, land managers should use local fire histories as an ecological basis for developing their fire management plans and prescriptions. fire managers should recognize that results of mean fire intervals from composite fire chronologies were produced by stochastic presettlement fire events and prescribed burns may not produce similar fire scarring and other fire effects. prescribed burning practices that mimic presettlement fires might include landscape-scale fires started from point ignitions, fires that only burn portions of the landscape, low to moderate intensity fires (i.e., , % trees scarred during a fire event) and variability in burning intervals. prescribed burning practices such as repeated ignitions to ensure an area is burned homogeneously and static fire intervals are not congruent with the presettlement fire regime we observed in the boone creek watershed. this empirical evidence supports fire as a key disturbance to maintain barren communities and it provides managers with information for understanding the nature of presettlement fire regimes, how dynamic fire regimes can be and how humans have modified the landscape with five centuries of burning. historic fire frequencies vary greatly between areas due to three and one half factors such as fuel, vegetation, weather and table .—pdsi values for contrasting periods of the fire record showing the similarity in drought conditions. there is no significant difference between the period – and – (tcrit ¼ . ) period mean fire interval (y) mean regional pdsi t stat (df ¼ ) – no fires . . – . . all nonfire years — . all fire years (n ¼ ) — . the american midland naturalist ( ) climate, topographic roughness and pyro-cultural influence. therefore, fire histories are typically applicable only to the area surrounding the study site. additional fire histories are needed from other historic barrens before variability in fire regimes in barrens communities throughout southern indiana and the larger interior low plateau can be fully characterized. acknowledgments.—this research was funded by a grant from the u.s. forest service, north central research station, columbia, missouri (rwu the ecology and management of central hardwoods unit). we thank steve olson and tom thake, hoosier national forest, for their assistance in this study. additional thanks to kevin hosman (university of missouri–columbia) for gis support. literature cited anderson, r. c. . presettlement forest of illinois, p. – . in: proc. of the oak woods management workshop. eastern illinois university, charleston. ——— and l. e. brown. . comparative effects of fire on trees in a mid-western savanna and an adjacent forest. bull. torr. bot. club, : – . ——— and j. e. schwegman. . twenty years of vegetational change on a southern illinois barren. nat. areas j., : – . anson, b. . the miami indians. university of oklahoma press, norman, oklahoma. p. bacone, j. a. and t. w. post. . effects of prescribed burning on woody and herbaceous vegetation in black oak sand savannas at hoosier prairie nature preserve, lake county, indiana. indiana acad. sci. proc., : – . ———, l. a. casebere and m. d. hutchison. . glades and barrens of crawford and perry counties, indiana. indiana acad. sci. proc., : – . fig. .—regional palmer drought severity index (pdsi) reconstructed from tree-rings (cook et al., ) and a smoothed pdsi utilizing a -y moving average for the purpose of displaying droughts with longer term influence guyette et al.: barren-forest mosaic bailey, r. g. . description of the ecoregions of the united states, nd ed. misc. pub. no. , usda for. serv. p. baird, w. d. . the quapaw indians: a history of the downstream people. univ. of oklahoma press, norman oklahoma. p. baker, w. l. and d. ehle. . uncertainty in surface-fire history: the case of ponderosa pine forests in the western united states. can. j. for. res., : – . bartgis, r. l. . the limestone glades and barrens of west virginia. castanea, : – . baskin, j. m., c. c. baskin and e. w. chester. . the big barrens region of kentucky and tennessee: further observations and considerations. castanea, : – . batek, m. j., a. j. rebertus, w. a. schroeder, t. l. haithcoat, e. compas and r. p. guyette. . reconstruction of early nineteenth-century vegetation and fire regimes in the missouri ozarks. j. biogeog., : – . bielmann, a. p. and l. g. brenner. . the recent intrusion of forests in the ozarks. ann. missouri bot. garden, : – . bourne, a. . on the prairies and barrens of the west. amer. j. sci. art, series , : – . braun, e. l. . deciduous forests of eastern north america. blakiston co., philadelphia. p. chester, e. w. . the kentucky barrens of northwestern middle tennessee: an historical and floristic perspective, p – . in: proc. st annual symp. on nat. hist. of lower tennessee and cumberland river valleys. the center for field biology of land between the lakes, clarksville, tennessee. clark, j. e. . the shawnee. univ. press of kentucky, lexington, kentucky. p. cook, e. r., d. m. meko, d. w. stahle and m. k. cleaveland. . drought reconstructions for the continental united states. j. climate, : – . cottam, g. . the phytosociology of an oak woods in southwest wisconsin. ecology, : – . cutter, b. e. and r. p. guyette. . fire frequency of an oak-hickory ridgetop in the missouri ozarks. am. midl. nat., : – . deselm, h. r. . tennessee barrens. castanea, : – . engelmann, h. a. . remarks upon the causes producing the different characters of vegetation known as prairies, flats, and barrens in southern illinois with specific reference to observation made in perry and jackson counties. am. j. sci. art, series , : – . gleason, h. a. . the individualistic concept of the plant association. am. midl. nat., : – . gray, a. . gray’s new manual of botany, th ed. american book company, new york. p. grimm, e. c. . chronology and dynamics of vegetation change in the prairie-woodland region of southern minnesota, u.s.a. new phyt., : – . grissino-mayer, h. d. . users manual fhx . laboratory of tree-ring research, university of arizona, tuscon. p. guyette, r. p. and b. e. cutter. . tree-ring analysis of fire history of a post oak savanna in the missouri ozarks. nat. areas j., : – . ——— and ———. . fire history, population, and calcium cycling in the current river watershed, p. – . in: proc. th central hardwood forest conference. usda for. serv. gen. tech. rep. nc- . ——— and d. c. dey. . a dendrochronological fire history of opeongo lookout in algonquin park, ontario, p. . in: ontario for. res. inst. for. res. rep. no. . sault ste marie, ontario. ——— and ———. . fire and logging history at huckleberry hollow, shannon county, missouri. forestry research paper no. . missouri dept. cons. jefferson city, missouri. p. ——— and ———. . humans, topography, and wildland fire: the ingredients for long-term patterns in ecosystems, p. – . in: proc. workshop on fire, people, and the central hardwoods landscape. usda for. serv. gen. tech. rep. ne- . ——— r. muzika and d. c. dey. . the dynamics of anthropogenic fire regimes. ecosystems, : – . heikens, a. l. and p. a. robertson. . barrens of the midwest: a review of the literature. castanea, : – . the american midland naturalist ( ) ———, k. a. west and p. a. robertson. . short-term response of chert and shale barrens vegetation to fire in southwestern illinois. castanea, : – . holmes, r. l., h. r. k. adams and h. fritts. . quality control of crossdating and measuring: a users manual for program cofecha, p. – . in: tree-ring chronologies of western north america: california, eastern oregon and northern great basin. laboratory of tree ring research, university of arizona, tuscon. homoya, m. a. . indiana barrens: classification and description. castanea, : – . ———, d. b. abrell, j. r. aldrich and t. w. post. . the natural regions of indiana. proc. indiana acad. sci., : – . hudson, c. . introduction, p. . in: cultural encounters in the early south. univ. of arkansas press, fayetteville. hutchison, m. d. . the barrens of the midwest: an historical perspective. castanea, : – . kaye, m. w. and t. w. swetnam. . an assessment of fire, climate, and apache history in the sacramento mountains, new mexico. physic. geog., : – . kellar, j. h. . an archaelological survey of perry county. indiana historical bureau, indianapolis. p. ———. . an introduction to the prehistory of indiana. indiana hist. soc., indianapolis. p. lounsberry, a. . a guide to the trees. frederick a. stokes co., new york. p. mcclain, w. e., m. a. jenkins, s. e. jenkins and j. e. ebinger. . changes in the woody vegetation of a bur oak savanna remnant in central illinois. nat. areas j., : – . mcinteer, b. b. . a change from grassland to forest vegetation in the ‘‘big barrens’’ of kentucky. am. midl. nat., : – . obrien, m. j. and w. r. wood. . the prehistory of missouri. univ. of missouri press. columbia. p. olson, s. d. . the historical occurrence of fire in the central hardwoods, with emphasis on southcentral indiana. nat. areas j., : – . paulsell, l. k. . effects of burning on ozark hardwood timberlands. univ. of missouri agric. exper. stat. bull. no. . p. ricketts, t. h., e. dinerstein, d. m. olson and c. j. loucks. . terrestrial ecoregions of north america. island press, washington d.c. p. robertson, p. a. and a. l. heikens. . fire frequency in oak-hickory forests of southern illinois. castanea, : – . sas institute inc. . sas/stat guide for personal computers, version . sas institute inc., cary, north carolina. sauer, c. o. . geography of the pennyroyal of kentucky, the barrens. kentucky geolog. survey, series , : – . schroeder, w. a. . presettlement prairie of missouri. natural history series, no. , missouri dept. cons. p. schroeder, m. j. and c. c. buck. . fire weather. usda for. serv., ag. handbook no. . p. smith, k. t. and e. k. sutherland. . fire-scar formation and compartmentalization in oak. can. j. for. res., : – . stats indiana. . indiana business research center, indiana university, bloomington. stokes, m. a. and t. l. smiley. . introduction to tree-ring dating. univ. of chicago press, illinois. p. stritch, l. r. . landscape-scale restoration of barrens-woodland within the oak-hickory forest mosaic. rest. mngmt. notes, : – . tyndall, r. w. . historical considerations of conifer expansion in maryland serpentine ‘‘barrens’’. castanea, : – . ———. . contributions of the barrens symposium: preface. castanea, : – . usda, forest service, hoosier national forest. . environmental assessment: ecological restoration in the boone creek watershed. tell city ranger district, perry county, indiana. march . vehik, s. c. . dhegia origins and plains archaeology. plains anthr., : – . guyette et al.: barren-forest mosaic wade, k. a. and e. s. menges. . effects of fire on invasion and community structure of a southern indiana cedar barrens. indiana acad. sci., : – . waldman, c. . atlas of the north american indian. facts on file. new york. p. submitted december accepted august the american midland naturalist ( ) evaluation of stress in laboratory rabbits used for teaching purposes monika urbanova , eva kramarova , jan chloupek , martina najmanova university of veterinary and pharmaceutical sciences brno, faculty of veterinary hygiene and ecology, department of veterinary public health and forensic medicine, faculty of veterinary medicine, department of pharmacology and pharmacy, brno, czech republic received november , accepted april , abstract this study was intended as a contribution to the argument about possible suffering of animals used for demonstrative purposes during teaching at universities. pharmacology lectures at the university of veterinary and pharmaceutical sciences brno, czech republic were selected for this experiment. the study focused on stress assessment using haematological and biochemical indices. blood samples were drawn from rabbits (oryctolagus cuniculus f. domesticus l.) days after arrival at the institute (before the first practical lesson), and then in the th and th weeks of the semester with practical lessons. the haematological and biochemical indices were compared between the experimental group (n = ) used for practical demonstrations and the control group (n = ) which was kept in its hutch. practical lessons included manipulation with rabbits, health examination, topical or total application and observation of the drugs’ effects. all the acts were carried out by students. significant changes were detected in some of haematological (erythrocytes, haemoglobin, haematocrit) and biochemical (glucose, total protein, enzymes aspartate aminotransferase and lactate dehydrogenase) variables compared to the control group. the values obtained from the last blood sampling indicate a possible habituation process in the experimental group between the th and th week. the results were compared between the experimental and control group and also with the results of other studies with different stressors. in conclusion, the stress the rabbits used for teaching purposes at universities are exposed to, is tolerable. animal stress, haematological indices, biochemical properties, glucose, total protein our study was intended as a contribution to the ongoing discussion about the level of stress in animals used for research and teaching demonstrations at universities. for this type of study, there is no classic stress model that could be applied which has led to a lack of studies and reference values in this field. rabbits are used as laboratory animals very often and they are suitable model organisms for teaching purposes, too. practical lessons have often been conducted using rabbits. stress is the organism’s response to a stimulus which leads to activation of the hypothalamic–pituitary–adrenocortical axis (hpa) and the sympatho- adrenomedullary system (m ö s t l and p a l m e ). both chains of hormonal reactions influence the adrenal glands and the glucocorticoids that are excreted lead to many hormonal changes and changes in the metabolic status (m ö s t l and p a l m e ). the main effect of elevated glucocorticoid concentrations (cortisol and corticosterone) is an increase in the glucose level (m ö s t l and p a l m e ; r o m e r o ; g r i s s o m and b h a t n a g a r ). glucose and cortisol are designated markers of acute stress in humans and animals (a r m a r i o et al. ; r o m e r o ; b r e i n e k o v a et al. ). however, severe chronic stress (prolonged periods of high cortisol concentrations in the blood stream and their long term effect) may negatively influence the overall health condition (m ö s t l and p a l m e ; g r i s s o m and b h a t n a g a r ) which is why concentrations of these hormones are used as health and welfare indicators. the stress reaction can alter many haematological and biochemical indices which is why the least invasive method for sample collection was acta vet. brno , : – ; https://doi.org/ . /avb address for correspondence: mvdr. monika urbanova palackého tř. / , brno, czech republic phone: + e-mail: urbanovam@vfu.cz http://actavet.vfu.cz/ chosen. in the event of a more pronounced chronic stress, not only do the haematological and biochemical indices change, but the overall health condition is also affected, e.g. by reduction of feed intake and weight loss (c a b e z a s et al. ; o k a b et al. ). the main aim of the study was to objectively determine the possible degree of stress impact on the rabbits caused by manipulation during practical lessons. the other aim was to determine whether the haematological and biochemical indices were affected in the long term. our intention was to contribute an objective argument to the discussion about the possible suffering of animals used for demonstration purposes in practical lessons. materials and methods twenty new zealand white rabbits (oryctolagus cuniculus f. domesticus l.) from a commercial breeding farm (lapins s.r.o., dolní dubňany, czech republic) were used. they were weeks old at the beginning of the experiment. the rabbits were housed individually in stainless steel cages but they could see each other. each cage was marked by a number. the rabbits were kept under controlled environmental conditions, and their biorhythm was set to h of light and h of dark. the temperature was kept between and °c and the humidity was between % and %. the rabbits were checked on every day. feed and water were provided ad libitum. the complete feed mixture for all the rabbits, biostan kbo (biokron s.r.o., blučina, czech republic), consisted of pellets, oats and other grains. cleaning of the hutches was performed every days or, when necessary, more frequently. this study was carried out during pharmacology lectures at the university of veterinary and pharmaceutical sciences brno, czech republic. the study was conducted over the weeks of the winter semester. twenty rabbits were divided into two groups, an experimental (n = ) and a control group (n = ). the control group was kept constantly in cages except during routine maintenance tasks. selected rabbits from the experimental group and other rabbits (which were not included into the study) were repeatedly used during practical lessons. there were rabbits in total in the facility. ten of them were randomly chosen as the experimental group and other rabbits were chosen as the control group. this selection was done before the study started. in each practical lesson, rabbits were used, which meant that rabbits of the experimental group were used or times a week like the other rabbits. the rabbits selected for the experimental group were transported from the housing facilities into individual cages in the classroom before the lesson began. the basic examination of the rabbits’ health condition was performed by students at the beginning of every lesson. this examination consisted of measuring the pulse, heart rate, and temperature. all procedures that were carried out on the rabbits by students during the semester are listed in table . the below mentioned substances were applied according to the instructions and then the therapeutic effect of the substance was observed or measured. during these procedures, the rabbits were usually fixated with towels. every rabbit was returned into the cage after the end of the practical act. they were housed in the cages in the classroom during lessons which lasted h. table . practical procedures on rabbits during the study. week of semester practical procedures according to the syllabus of practical lessons st and nd manipulation, fixation, basic examination rd and th drops of . % atropini sulfas intraconjunctival drops drops of . % physostigmini salicylas application drops of % physostigmini salicylas th and th mg/ml epinefhrini hydrochloridum . ml intramuscular local anaestesia mg/ml epinefhrini hydrochloridum mg/kg intramuscular intraconjunctival drops drops of . % cocaini hydrochloridum application or intramuscular drops of . % procaini hydrochloridum application th and th % solution morphini hydrochloridum, mg/kg sub cutis th and th inhalation anaesthesia - isoflurane, premedication medetomidini anaesthesia hydrochloridum mg - . ml pro toto, ketamine mg/kg intravenous anaesthesia - . % thiopental mg/kg th and th manipulation, fixation, basic examination, blood sampling . ml this experimental project was approved by the animal use committee of the university of veterinary and pharmaceutical sciences brno no. pp - and was in compliance with the national legislation (act no. / coll., on the protection of animals against cruelty, as amended and decree no. / coll., on the protection of experimental animals, as amended). both groups of rabbits had their blood collected three times. the first sampling was done ten days after they were acclimated to the new conditions and before the first practical lesson. the second sampling was done in the th week of the semester, and the third sampling was done in the th week of the semester. every rabbit was taken from its cage by hand and wrapped in a towel on the sampling table. their keepers manipulated and fixated them, and blood was always collected without sedation. daily rhythms did not affect the results because the sampling was done between : and : h. ethyl alcohol was applied to dissinfect the collection site and . millilitres of blood were obtained from the vena auricularis marginalis of each rabbit. the blood was collected into two commercial tubes (dispolab s.r.o., brno, czech republic), first into a tube with edta (ethylenediaminetetraacetic acid) for haematology and then into one containing amorphous heparin for biochemical examination. after collecting blood samples from both groups, the samples were transported to the university laboratory where they were immediately analysed. all samples were analysed for a complete blood count. haematological indices such as the amount of erythrocytes (ery), leukocytes (leu), thrombocytes (throm) and haemoglobin (hb) was done by the haematological analyser sysmex xt- iv (sysmex, kobe, japan). from these values, other indices were calculated such as haematocrit (ht) and red blood cell indices such as the mean corpuscular volume (mcv), mean corpuscular haemoglobin (mch) and mean corpuscular haemoglobin concentration (mchc). the manual evaluation differential of leucocytes from the blood stains with hemacolor was done by the microscope olympus, type bx tf (olympus, prague, czech republic). the blood samples in the tubes with amorphous heparin were centrifuged (at g for min). biochemical tests of the blood plasma were performed to obtain the following indices: total protein, albumin, glucose, aspartate aminotransferase (ast), alanine aminotransferase (alt) and lactate dehydrogenase (ldh). the measurement was done by the analysator abbott architect c (abbott, princeton, new jersey, usa) with a photometric determination. the rest of the plasma samples were placed in eppendorf tubes and frozen at - °c, and later used for hormone analysis. corticosterone concentration was measured by an enzyme immunoassay commercial kit (corticosterone stressxpress eia kit stressmarq biosciences inc., victoria, bc canada). sensitivity of this test was determined as . pg/ml and detection limit was determined as . pg/ml. photometric measurement of colour intensity was performed on the calorimetric reader at a wavelength of nm. the results were determined after reading from the standard curve which was proposed in the programme free elisa software (elysaanalysis.com). determination of cortisol in blood plasma was measured by liquid chromatography with tandem mass spectrometry (lc/ms/ms). the thermo scientific uhplc accela system was connected to the thermo scientific tsq quantum access max triple quadrupole instrument (thermo, san jose, ca, usa) equipped with a heated electrospray ionization (hesi-ii) probe. the mobile phase consisted of water containing . % formic acid (v/v) and acetonitrile containing . % formic acid. cortisol was separated by an isocratic elution method. the heated electrospray ionization was operated in positive-ion mode. the detection limit for cortisol was , pg/ml. some plasma cortisol and corticosterone reference values are listed in table . it is difficult to determine the basal values of these hormones, because any interference before blood sampling may affect them. all statistical analyses were performed using the statistical software unistat for excel version . the tests of normality were carried out in the first phase. the values of the corresponding gauss curve were evaluated by t-test, non-parametric values were evaluated by mann-whitney’s test. table . selected plasma corticol and corticosterone reference values and maximal and minimal value influenced by the daily rhytm as determined for rabbits. glucocorticoid hormone values condition study cortisol reference values . – . μg/dl washington and van hoosier minimal . μg/dl at : h szeto maximal . μg/dl at : h corticosterone reference value . μg/dl washington and van hoosier baseline values . ± . μg /dl rosenthal minimal . ng/ml at : h s z e t o maximal . ng/ml at : h differences were considered to be significant when the probability of the null hypothesis was less than . (i.e., p < . ) and highly significant if the probability of the null hypothesis was less than . (i.e., p < . ). the average values of the experimental group and the control group were compared within one sampling. results the mean values of haematological indices are shown in table . the erythrocyte count of animals of the experimental group was significantly lower (p < . ) compared to the control group in the second and third sampling. haematocrit mean values in rabbits of the experimental group were significantly lower (p < . ) compared to the control group after the second blood sampling, and highly significantly lower (p < . ) compared to control group after the third blood sampling. table . haematological indices in rabbits of the control (n = ) and the experimental group (n= ). first sampling second sampling third sampling control experimental control experimental control experimental group group group group group group leu × /l . ± . . ± . . ± . . ± . . ± . . ± . ery × /l . ± . . ± . . * ± . . * ± . . * ± . . * ± . hb g/l . ± . . ± . . ± . . ± . . * ± . . * ± . ht l/l . ± . . ± . . * ± . . * ± . . ** ± , . ** ± . thrombocytes × /l . ± . . ± . . ± . . ± . . ± . . ± . neutrophils × /l . ± . . ± . . ± . . ± . . ± . . ± . lymphocytes × /l . ± . . ± . . ± . . ± . . ± . . ± . monocytes × /l . ± . . ± . . ± . . ± . . ± . . ± . eosinophils × /l . ± . . ± . . ± . . ± . . ± . . ± . basophils × /l . ± . . ± . . ± . . ± . . ± . . ± . mcv fl . ± . . ± . . ± . . ± . . ± . . ± . mch pg . ± . . ± . . ± . . ± . . ± . . ± . mchc g/l . ± . . ± . . ± . . ± . . ± . . ± . leu - leukocytes, ery - erythrocytes, hb - haemoglobin, ht - haematocrit, mcv - mean corpuscular volume, mch - mean corpuscular haemoglobin, mchc - mean corpuscular haemoglobin concentration all values were expressed as the mean value ± standard deviation. the mean values of the experimental and the control group were compared within one sampling period.* p < . significant; ** p < . highly significant. table . biochemical indices in rabbits of the control (n = ) and experimental group (n = ). first sampling second sampling third sampling control experimental control experimental control experimental group group group group group group total protein g/l . ± . . ± . . * ± . . * ± . . * ± . . * ± . albumin g/l . ± . . ± . . ± . . ± . . ± . . ± . glucose mmol/l . ± . . ± . . ** ± . . ** ± . . ± . . ± . alt μkat/l . ± . . ± . . ± . . ± . . ± . . ± . ast μkat/l . ± . . ± . . * ± . . * ± . . ± . . ± . ldh μkat/l . ± . . ± . . *± . . * ± . . ± . . ± . alt - alanine aminotransferase, ast - aspartate aminotransferase, ldh - lactate dehydrogenase all values were expressed as the mean value ± standard deviation. the mean values of the experimental and the control group were compared within one sampling period. * p < . significant; ** p < . highly significant. significantly lower (p < . ) differences were found between the mean haemoglobin values in rabbits of the experimental group compared to the control group after the third sampling. other haematological indices were without significant differences. the differences between biochemical indices and other mean values are listed in table . total protein values in animals of the experimental group were significantly lower (p < . ) than the control group values after the second and third blood sampling. a highly significant elevation of glucose (p < . ) was determined after the second sampling. the mean glucose value in the control group was higher compared to mean value of the experimental group, which was an unexpected result for this group. aspartate aminotransferase was significantly increased (p < . ) in the experimental group compared to the control group after the second sampling. also, the elevation of ldh activity in the experimental group was significant (p < . ) compared to the control after the second sampling. albumin and alt values in rabbits were without significant differences (p > . ) during the experimental period. the mean corticosterone value of both groups is presented in table . the measured corticosterone values were low; the mean values were around pg/ml and the highest value was , pg/ml. the differences were non-significant. plasma cortisol levels were below the cortisol detection limit, less than , pg/ml. discussion the aim was to find out whether the contact with students associated with practical lessons caused any stress effects in the rabbits. the value of erythrocytes, haemoglobin and haematocrit were determined as indices that may vary depending on the duration of stress. chronic stress results in a decrease in these indices (o k a b et al. ; e v e r d s et al. ). w e i et al. ( ) noted in their study that psychological stress in rats decreased the values of erythrocytes, haemoglobin and erythropoietin. our study shows the same result, however, the values correspond to the reference range (ve l l a and d o n n e l l y ; k n o t k o v a et al. ). the loss of blood was minimal during the study. our study did not show any significant changes in the leukocyte profiles. the first sampling of our study was done when the rabbits were seven weeks old and the total protein and albumin values were normal and without significant differences. total protein values in young rabbits are lower than in adults; rabbits older than weeks achieve normal adult-like values (m e l i l l o ). values of the second and third samplings were significantly lower in the experimental group; the differences were especially evident in the second sampling. reasons for reduction of protein are chronic malnutrition or loss of proteins (m e l i l l o ). no health problems were observed during the study. at the end of the study, rabbits from both groups weighed about . kg. total protein could be designated as a long-term stress indicator in rabbits, as a study of the effect of noise on rabbits showed significantly lower total protein values after days of daily exposure (e l w a s i f e et al. ) and another one reported that a higher concentration of rabbits in a cage also resulted in a reduction in protein values (k a l a b a ). table . mean values of corticosterone which were analysed after the first, second and third blood sampling. first sampling second sampling third sampling control experimental control experimental control experimental group group group group group group corticosterone pg/ml . ± . . ± . . ± . . ± . . ± . . ± . all values are expressed as the mean value ± standard deviation. in our study, a higher concentration of glucose was determined in the control group during second sampling which was an unexpected outcome. it could have been triggered by unusual sounds in the room, by manipulation and fixation because the control group was not as used to these procedures as the experimental group which was used for demonstration purposes in practical lessons. the trend of gradual increase in glucose levels was observed in both groups during our study. the level of glycaemia is considered a stress indicator in a number of other cases because many factors lead to the change of the glucose concentration in rabbits (wa s h i n g t o n and va n h o o s i e r ), e.g. chronic noise exposure can cause a reduction (e l w a s i f e et al. ) but heat stress (s k ł a d a n o w s k a - b a r y z a et al. ), higher concentration of rabbits in the cage (k a l a b a ) or transport to slaughter (s k ł a d a n o w s k a - b a r y z a et al. ) can cause its elevation. transport and manipulation in the laboratory can increase the glucose concentration (and other variables) in laboratory rats (c a s t e l h a n o - c a r l o s and b a u m a n s ). it is assumed that the increase of enzymes ast and ldh in the second sampling in our study was due to repeated stress in the experimental group. later during the third sampling, no differences were found, presumably due to the experimental group becoming used to the manipulation. muscle enzymes such as ldh, ast and alt increase after capturing (s á n c h e z et al. ), however, rough manipulation or fixation, yet gentle handling does not lead to changes in the concentration of these enzymes (m e l i l l o ). these enzymes can be elevated after muscle damage and blood sampling coupled with restrictive activities, but they can also be increased by haemolysis of the sample (wa s h i n g t o n and va n h o o s i e r ). haemolytic samples were not reported by the laboratories, though. the results of cortisol and corticosterone were lower than known basal values. we assumed that the morning samples would be lower but still measurable. chronic elevation of glucocorticoid hormones were not proven after repeated exposure during lessons in the experimental as well in the control group. in our study the average manipulation and sampling time was min per one rabbit, from the opening of the cage to the return there. the response to the stress usually takes about – min after exposure to the stressor (r o m e r o ) and the level of elevation of glucocorticoids is individual and also depends on the type of stressor (d e k l o e t et al. ). it was expected that repetitive stress would affect these levels in the long term. if there was an increase in the concentrations of these hormones in blood due to manipulation (acute stress), we were not able to detect it. in summary, our study showed significant changes in the experimental group compared to the control group. however, these indices are within the reference range established for the rabbit (ve l l a and d o n n e l l y ; k n o t k o v a et al. ) the habituation process during the semester also had some influence, because the repeated process during lessons stimulated the hpa less, and subsequent reactions were milder (g r i s s o m and b h a t n a g a r ). good overall health was maintained. the results indicate that the effects of stress were at a tolerable level. the use of a large number of rabbits during education is not ideal but this system allows for students to gain knowledge and skills needed in follow-up disciplines without exhausting the rabbit. thanks to the rotation of the rabbits, rest and partial regeneration were also possible. acknowledgement the study was supported by the iga vfu / fvl. references armario a, marti o, molina t, de pablo j, valdes m : acute stress markers in humans: response of plasma glucose, cortisol and prolactin to two examinations differing in the anxiety they provoke. psychoneuroendocrinology : - breinekova k, svoboda m, smutna m, vorlova l : markers of acute stress in pigs. physiol res : - cabezas s, blas j, marchant ta, moreno s : physiological stress levels predict survival probabilities in wild rabbits. horm behav : - castelhano-carlos mj, baumans v : the impact of light, noise, cage cleaning and in-house transport on welfare and stress of laboratory rats. lab anim-uk : - de kloet er, sibug rm, helmerhorst fm, schmidt mv : stress, genes and the mechanism of programming the brain for later life. neurosci biobehav r : - elwasife k, aziz ia, shabat m, shahwan o, el hamidi am . effects of noise on rabbit’s blood. european journal of biophysics : - everds ne, snyder pw, bailey kl, bolon b, creasy dm, foley gl, rosol tj, sellers t : interpreting stress responses during routine toxicity studies: a review of the biology, impact, and assessment. toxicol pathol : - grissom n, bhatnagar s : habituation to repeated stress: get used to it. neurobiol learn mem : - kalaba zm : physiological response and stress indicators of california rabbits under intensive conditions in egypt. asian j poultry sci : - knotkova z, haupman k, jekl v : laboratorní parametry. in: knotek z (ed.): nemoci zvířat zájmových chovů: drobní savci. profi press, praha, pp. - melillo a : rabbit clinical pathology. j exot pet med : - möstl e and palme r : hormones as indicators of stress. domest anim endocrin : - okab ab, el-banna sg, koriem aa : influence of environmental temperatures on some physiological and biochemical parameters of new-zealand rabbit males. slovak j anim sci : - romero lm : physiological stress in ecology: lessons from biomedical research. trends ecol evol : - rosenthal ke : ferret and rabbit endocrine disease diagnosis. in: fudge am (ed.): laboratory medicine: avian and exotic pets, saunders, philadelphia, pp. - sánchez o, arnau a, pareja m, poch e, ramírez i, soley m : acute stress-induced tissue injury in mice: differences between emotional and social stress. cell stress chaperon : - składanowska-baryza j, ludwiczak a, pruszyńska-oszmałek e, kołodziejski p, bykowska m, stanisz m : the effect of transport on the quality of rabbit meat. anim sci j : - szeto a, gonzales ja, spitzer sb, levine je, zaias j, saab pg, schneiderman n, mccabe pm : circulating levels of glucocorticoid hormones in whhl and nzw rabbits: circadian cycle and response to repeated social encounter. psychoneuroendocrino : - vella d, donnelly tm : basic anatomy, physiology and husbandry. in: quesenberry ke, carpenter jw (eds.): ferrets, rabbits, and rodents: clinical medicine and surgery. elsevier, st. louis, missouri, p. washington im, van hoosier g : clinical biochemistry and hematology. in: suckow ma, stevens ka, wilson rp (eds): the laboratory rabbit, guinea pig, hamster, and other rodents. waltham, ma: academic press/elsevier. american college of laboratory animal medicine series, pp. - wei c, zhou j, huang x, li m : effects of psychological stress on serum iron and erythropoiesis. int j hematol : - taking the country barefooted: the indiana colony in southern california | california history | university of california press skip to main content close 	ucpress 	about us 	blog 	support us 	contact us search search input search input auto suggest search filter all content california history search user tools register carnegie mellon university carnegie mellon university sign in toggle menumenu 	 content 	 recent content 	 browse issues 	 all content 	 purchase 	 alerts 	 submit 	 info for 	 authors 	 librarians 	 reprints & permissions 	 about 	 journal 	 editorial team 	 richard j. orsi prize 	 contact us skip nav destination article navigation close mobile search navigation article navigation volume , issue october 	 previous article 	 next article article navigation research article| october taking the country barefooted: the indiana colony in southern california james h. madison james h. madison search for other works by this author on: this site pubmed google scholar california history ( ) ( ): – . https://doi.org/ . / 	 views icon views 	article contents 	figures & tables 	video 	audio 	supplementary data 	 share icon share 	facebook 	twitter 	linkedin 	email 	 	 tools icon tools 	 get permissions 	 cite icon cite 	 search site citation james h. madison; taking the country barefooted: the indiana colony in southern california. california history october ; ( ): – . doi: https://doi.org/ . / download citation file: 	ris (zotero) 	reference manager 	easybib 	bookends 	mendeley 	papers 	endnote 	refworks 	bibtex toolbar search search search input search input auto suggest search filter all content california history search this content is only available via pdf. copyright chs article pdf first page preview close modal you do not currently have access to this content. sign in don't already have an account? register client account you could not be signed in. please check your email address / username and password and try again. email address / username ? password sign in reset password register sign in via your institution sign in via your institution citing articles via web of science ( ) google scholar crossref 	 latest 	 most read 	 most cited mexican american parrhesia at troy: the rise of the chicana/o movement at the university of southern california, – breaking the eleventh commandment: pete mccloskey’s campaign against the vietnam war fishing on porpoise: the origins of dolphin bycatch in the american yellowfin tuna industry keepers of the culture at adeline street: locating black power theater in berkeley, california suburban cowboy: country music, punk, and the struggle over space in orange county, – email alerts article activity alert latest issue alert close modal 	 recent content 	 browse issues 	 all content 	 purchase 	 alerts 	 submit 	 info for authors 	 info for librarians 	 about 	 editorial team 	 richard j. orsi prize 	 contact us 	online issn - 	print issn - 	copyright © stay informed sign up for enews 	twitter 	facebook 	instagram 	youtube 	linkedin visit the uc press blog disciplines 	ancient world 	anthropology 	art 	communication 	criminology & criminal justice 	film & media studies 	food & wine 	history 	music 	psychology 	religion 	sociology 	browse all disciplines courses 	browse all courses products 	books 	journals resources 	book authors 	booksellers 	instructions 	journal authors 	journal editors 	librarians 	media & journalists support us 	endowments 	membership 	planned giving 	supporters about uc press 	careers 	location 	press releases 	seasonal catalog contact us 	acquisitions editors 	customer service 	exam/desk requests 	media inquiries 	print-disability 	rights & permissions 	royalties 	uc press foundation 	© copyright by the regents of the university of california. all rights reserved. privacy policy   accessibility close modal close modal this feature is available to subscribers only sign in or create an account close modal close modal this site uses cookies. by continuing to use our website, you are agreeing to our privacy policy. accept news & notes promotions ruth anita bevan, yeshiva university: associ- ate professor and chairman. sharon e. doerner, university of toledo: asso- ciate professor. abel jacob, york college, cuny: associate professor. manindra k. mori apatra, old dominion univer- sity: associate professor. david j. olson, indiana university: associate professor. edward rogowsky, york college, cuny: assis- tant professor. kim e. shienbaum, york college, cuny: assistant professor. john sullivan, indiana university: associate professor. joseph s. szyliowicz, university of denver: professor. ronald e. weber, indiana university: associate professor. retirements joseph dunner, david petegorsky professor of political science, yeshiva university, will retire in june . john brown mason, professor emeritus, califor- nia state university, fullerton, as of july , . in memoriam carlton clymer rodee carlton clymer rodee, professor emeritus of political science at the university of southern california, died in pasadena on september . he was seventy-four years of age and had been a usc faculty member for thirty-five years prior to his retirement in . born and raised in milwaukee, wisconsin, carl- ton did his baccalaureate and master's degree at madison and took his ph.d. at yale, where he was a cowles fellow. carlton rodee's entire teaching career was at usc. after a short term in the school of public administration, he was promoted to an associ- ate professorship in political science and public administration, and in , to a full professor- ship in political science. his leadership in the department of political science was unmis- takable; in fact, he served as chairman twice, - and - . his faculty leadership at usc was evidenced in other ways as well. he was active in the american association of university professors and the so-called faculty uprising ("revolt ' "). he was one of the founders of the faculty senate and elected vice chairman in - . as a political scientist, he spanned all levels of professional organization: international politi- cal science association; the california ful- bright commission from - ; the ameri- can political science association; the western association; the board of editors of the west- ern political quarterly for a number of years; and the southern california association, one of its founders and president, - . in addi- tion, over the years, he served in a number of civic capacities, such as a member of the panel for the western region of the war labor board during world war ii and as a member of the team which surveyed the university of nevada in . as a scholar, dr. rodee's major writings were closely related to his teaching. he was editor and co-author of problems of democratic soci- ety, , and twentieth century political thought, . he was principal architect and co-author of introduction to political science, . he was a contributor to the encyclo- pedia of the social sciences, collier's encyclo- pedia and to a number of academic journals. professor rodee's great strength was his teach- ing. he was unusually effective even with large undergraduate classes. many years after these classes, his former students bring up his name as "my most memorable professor." the graduate students whose doctoral work professor rodee chaired, regard themselves as his personal pro- ducts. each pridefully speaks of himself as a "rodee man." they mention his idealism and his inspirational qualities; they even refer ap- provingly to his perfectionist traits. he came by his idealism ancestrally; his namesake was george clymer, signer of both the declaration of independence and the constitution. carlton's colleagues in the department and throughout the university of california held him in high regard and received continuing inspiration and guidance from him following his retirement. his family survives him: his widow, frances fritz, whom he married as an under- graduate, and their daughter, georgia (mrs. gayle e. post). henry reining, jr. university of southern california wayne a. wilcox wayne a. wilcox, together with his wife ouida and children kailan and clark, was killed in the mid-air explosion of the turkish airliner outside paris on march , . he was forty-one years old. two other children, shelley and spencer, survive, as do wayne wilcox's mother and brother and ouida wilcox's mother. wayne wilcox became chairman of columbia university's department of political science in the autumn of , at a most difficult and painful moment, when both the department and the university at large were still deeply divided into angry camps in the aftermath of ps spring the previous spring's crisis, and when many students were pressing for sweeping curricular and structural changes. positions were trumpet- ed as non-negotiable, tempers ran short, and moods were unforgiving. combining, in equal measure, calm, cajolery, humor, guile, pru- dence, tolerance, and occasional pretenses to amnesia, wayne wilcox (an old navy hand) steered us through this minefield and on a course that brought us to major changes (hope- fully they are improvements) in the depart- ment's doctoral program, to new (and certainly desirable) procedures for involving students in areas of departmental decision-making that had previously been the exclusive preserve of the faculty — such as admissions, personnel search- es, and curricular experimentation — and, most important, to a return of civility and goodwill within a departmental community many of whose members had been barely on speaking terms at the beginning of wilcox's term-of- office. while engaged in this healing task, wilcox also recruited to the department, and made wel- come within it, a number of its finest current teachers and scholars. in addition, he was an active member of the university's southern asian institute and helped found and guide the international fellows program. despite these heavy administrative responsibilities, he en- joyed teaching and was good at it, and he also proudly and effectively pursued his scholarly work, usually during his summers at rand on the west coast. indeed, this scholarly activity continued even into the recent years of his diplomatic assignment. one of his last articles, on current american policy toward india, pakistan, and bangladesh, was sent us from london and is assigned in at least one of our courses. wilcox's scholarly work combined an expertise in the politics of a geographic region, that is, the indian subcontinent, with a subtle under- standing of the comparative problems of devel- opment, and with a semi-insider's knowledge of american policy toward such areas and prob- lems. his publications are the product of a skillful mixture of documentary analysis, field work, and intelligence. the author of three books and of many articles, he was also an ever-alert prober of methodological innovations in the discipline political science. he was the first, and for several years the only, member of this department who knew what is a stochastic model and, in a tangential area, was the co-author of the all-too-true issawi-wilcox prin- ciple which stipulates that "problems increase in geometric ratio, solutions in arithmetic ratio." it is unlikely that wayne would have explicitly termed himself an intellectual, still less a member of the intelligentsia. his hoosier home- spun horse-sense cut against such grandiloquen- ces. still, he had a deep commitment to the application of academic and cultivated educa- tion to the world's problems. congratulating one of his former students upon the latter's receipt of the doctorate, he wrote last year from london: " . . . it remains an act of faith on my part that bright people ought to cultivate their intellects and turned (sic) loose against society which is clearly inefficient, rigged, and disagreeable. i can't see how more educated people could harm it and it is just possible, hobbes notwithstanding, that they might inch it a little closer to a humane civilization." though the syntax of those lines is somewhat uncertain, their import is clearly benevolent and true. joseph rothschild columbia university wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ phase ii study of fosaprepitant + ht receptor antagonist + dexamethasone in patients with germ cell tumors undergoing -day cisplatin-based chemotherapy: a hoosier cancer research network study phase ii study of fosaprepitant + ht receptor antagonist + dexamethasone in patients with germ cell tumors undergoing -day cisplatin-based chemotherapy: a hoosier cancer research network study nabil adra , costantine albany , mary j. brames , somer case-eads , cynthia calley , ziyue liu , christopher a. fausel , timothy breen , nasser h. hanna , ralph j. hauke , joel picus , lawrence h. einhorn : division of hematology/oncology, melvin & bren simon cancer center, indiana university school of medicine, indianapolis, indiana. : department of biostatistics, indiana university, indianapolis, indiana. : hoosier cancer research network. : methodist hospital/nebraska cancer specialists, omaha, nebraska. : division of hematology/oncology, washington university school of medicine, st. louis, missouri. trial supported by merck & co. inc. -presented in part at the st annual meeting of the american society of clinical oncology (asco), chicago, may -june , . -corresponding author: nabil adra, md. email: nadra@iu.edu address: barnhill drive, rt , indianapolis, in telephone: - - ; pager: - - ; fax: - - -dr. einhorn is supported in part by the national cancer institute ( ro ca ) a . -dr. einhorn is a livestrong foundation distinguished professor of medicine. _________________________________________________________________________________ this is the author's manuscript of the article published in final edited form as: adra, n., albany, c., brames, m. j., case-eads, s., johnson, c. s., liu, z., … einhorn, l. h. ( ). phase ii study of fosaprepitant + ht receptor antagonist + dexamethasone in patients with germ cell tumors undergoing - day cisplatin-based chemotherapy: a hoosier cancer research network study. supportive care in cancer, ( ), – . https://doi.org/ . /s - - -y https://doi.org/ . /s - - -y     phase ii study of fosaprepitant + ht receptor antagonist + dexamethasone in patients with germ cell tumors undergoing -day cisplatin-based chemotherapy: a hoosier cancer research network study abstract    purpose a phase iii study adding aprepitant to a htt receptor antagonist plus dexamethasone in germ cell tumor (gct) patients treated with day cisplatin combination chemotherapy demonstrated a significant improvement in complete response (cr) (j clin onc : - , ). fosaprepitant has demonstrated non-inferiority compared to aprepitant in single day cisplatin chemotherapy and is approved as a single-dose alternative. this single arm phase ii study is the first clinical trial evaluating fosaprepitant in patients receiving multi-day cisplatin regimen. methods gct patients receiving day cisplatin combination chemotherapy were enrolled. fosaprepitant mg was given iv on days and . a ht antagonist days - (days , , , if palonosetron) plus dexamethasone mg days , , and mg po bid days , , was administered. rescue antiemetics were allowed. the primary objective was to determine the cr rate – no emetic episodes or use of rescue medications. accrual of patients was planned with expected cr> %. results patients were enrolled of whom were eligible for analysis. median age was . patients received bep (bleomycin, etoposide, cisplatin) chemotherapy. cr was observed in ( . %) patients ( % agresti-coull binomial c.i. . %, . %). conclusion the data in this phase ii study, in contrast to our prior phase iii study, appears to indicate a lower cr rate with the substitution of fosaprepitant for aprepitant. it is unknown whether the substitution of fosaprepitant for aprepitant provides the same benefit in multi-day cisplatin that was achieved with single day cisplatin. -keywords: germ cell tumor, testicular cancer, fosaprepitant, chemotherapy-induced nausea and vomiting     introduction germ-cell tumors (gcts) represent the most common carcinoma in men ages to years. there is an estimated , new cases to be diagnosed in the united states in [ ]. the introduction of cisplatin-based combination chemotherapy has established testicular cancer as a model for a curable neoplasm [ ]. acute and delayed nausea and vomiting is a universal adverse effect of cisplatin chemotherapy. despite significant progress in the prevention of chemotherapy- induced nausea and vomiting (cinv), emesis continues to be associated with a significant deterioration in quality of life in patients treated with combination chemotherapy [ ]. the addition of the brain-penetrant neurokinin- receptor antagonist (nk -ra) aprepitant, to a - hydroxytryptamine- receptor antagonist ( ht -ra) and dexamethasone improved the prevention of cinv in patients receiving highly emetogenic single-day cisplatin chemotherapy [ - ]. in the hoosier cancer research network (hcrn), a phase iii study was conducted with a ht -ra plus dexamethasone plus either aprepitant or placebo in a randomized double-blind placebo controlled study of patients receiving day cisplatin-based combination chemotherapy. this study demonstrated a significant improvement in complete response (cr) rate of % versus % (p< . ) favoring aprepitant [ ]. although the oral capsule is appropriate for many patients, the availability of an intravenous (iv) formulation would provide further convenience and flexibility. fosaprepitant is a water-soluble prodrug of aprepitant available in iv formulation. it is rapidly converted to the active form     (aprepitant) by phosphatase enzymes and is expected to provide the same aprepitant exposure in regards to auc and hence similar antiemetic effect [ ]. a randomized, double-blind phase iii study was conducted to assess the efficacy and safety of fosaprepitant in the prevention of cinv in cisplatin-naive patients who were treated with chemotherapy regimens that included single-day cisplatin ≥ mg/m [ ] enrolled over , patients and demonstrated that a single dose of intravenous fosaprepitant was non-inferior to standard -day oral aprepitant. here we report the results of the first clinical trial evaluating fosaprepitant in patients receiving multi-day cisplatin combination chemotherapy. methods patient selection eligible patients were ≥ years of age with histologically or cytologically confirmed diagnosis of germ cell tumor who were scheduled to receive a standard day cisplatin based chemotherapy regimen. prior chemotherapy was allowed and patients did not have to be chemotherapy naive. patients had to be without nausea or vomiting for hours before study entry and no antiemetic use for hours prior to starting protocol therapy. absolute neutrophil count ≥ , cells/µl, wbc count ≥ , cells/µl, platelet count ≥ , cells/µl, ast and alt ≤ x upper limit of normal, bilirubin ≤ . x upper limit of normal, and creatinine less than mg/dl were required. patients should have had an ecog performance status of - and no active central nervous system (cns) metastasis.     this single arm hcrn phase ii study was conducted from to after approval by each site’s institutional review board. informed consent was obtained from all individual participants included in the study. study design and treatment regimen all participating patients received germ cell combination chemotherapy utilizing cisplatin mg/m x days. all patients received an anti-emetic regimen consisting of dexamethasone mg orally daily on days and then mg orally bid (twice daily) on days , , and . a ht - ra was administered to all patients minutes before starting chemotherapy: ondansetron mg orally bid on days to or palonosetron . mg iv on days , , and . fosaprepitant mg was administered iv on days and over - minutes (table ). patients were permitted to take rescue therapy of the treating investigator’s choice for nausea and/or emesis/retching based on clinical circumstances. no additional doses of ht -ra, dexamethasone, or fosaprepitant were given during the acute or delayed treatment periods. patients who required rescue therapy were permitted to continue the study at the discretion of the treating investigator and in consultation with the patient. the primary endpoint was complete response (cr) of both acute (days through ) and delayed (days through ) cinv, defined by no emetic episodes or use of rescue medications. secondary endpoints included incidence of emetic episodes via patients logs (days through ), use of rescue medications (days through ), patient’s self-reported assessment of nausea (days through ) using a - mm visual analog scale (vas), safety, and toxicity. study visits and assessment procedures     in the pre-study period, all pertinent demographics (age, gender, height, and weight) and medical data (site and stage of disease, ecog performance status, laboratory values, medications, and prior oncologic therapies) were recorded. all patients were provided a diary inclusive of days through of the chemotherapy cycle. patients were asked to complete a daily log of any episodes of vomiting or retching and the time of these episodes beginning with day and daily through day . the use of rescue therapy, defined as any medication taken to treat established nausea or emesis, was also recorded. on days through , patients rated nausea by using a -mm horizontal vas ranking their nausea for the prior hours from no nausea to the worst nausea with a measurement of - mm. a vas score of to mm was considered as no nausea. statistical methods the cr rate of aprepitant in gct patients receiving day cisplatin-based combination chemotherapy was % [ ]. we expected fosaprepitant to have similar cr rate. a one-sided score test was used to compare the cr rate of fosaprepitant with the historical data of aprepitant. denoting the cr rate of fosaprepitant as p, the hypotheses are h : p ≤ % versus ha: p ≥ %. a cr rate of fosaprepitant that is no worse than % lower than aprepitant was considered worth further investigation. enrollment of patients was required to attain a power of . with type i error level as . . continuous variables were summarized by mean, median, and range. categorical variables were summarized by frequencies and percentages. overall cr was tested according to the aforementioned hypotheses. in addition, % confidence interval of agresti-coull type was constructed for the overall cr. adverse events were summarized by their grades and types.     results patient characteristics from january till may , patients with gct who were scheduled to receive day cisplatin chemotherapy were enrolled. a total of patients were enrolled on study. one patient had a reaction to the infusion and was taken off study. patients were not evaluable because they did not complete the vas on all days. hence, patients were evaluable and eligible for analysis demographic and treatment data are listed in table . median age was (range - ). all patients were caucasian and male. eastern cooperative oncology group (eccog) performance status was in ( . %) patients, in ( . %) patients, and in ( . %) patient. chemotherapy regimen consisted of bep (bleomycin, etoposide, cisplatin) in ( . %) patients, ep (etoposide, cisplatin) in ( . %) patients, veip (vinblastine, ifosfamide, cisplatin) in ( . %) patients, and cisplatin/epirubicin in ( . %) patient. all evaluable patients received fosaprepitant mg iv on days and . among evaluable patients, received ondansetron and and received palonosetron. all evaluable patient received days of dexamethasone. among evaluable patients, ( . %) were chemotherapy-naïve. table depicts patient and treatment characteristics. efficacy endpoints primary and secondary endpoints were analyzed in the evaluable patients. complete response (cr), defined by no emetic episodes and no use of rescue medications, was observed in ( . %) patients in the overall treatment period day - ( % agresti-coull binomial c.i. . %, . %). this was insufficient to reject the null hypothesis h : p ≤ % (p= . ). cr was observed in ( . %) patients during the acute phase (days through ) and ( . %)     patients in the delayed phase (days through ). figure depicts the percentage of cr in the acute, delayed, and overall phases. sixteen ( . %) patients had at least one emetic episode. total number of emetic episodes was . nine patients had emetic episode; patients had episodes; patients and episodes and patient had emetic episodes. the largest number of emetic episodes occurred on days , , and with patients having an emetic episode on each of these days. the fewest number of emetic episodes occurred on days and with patients having an emetic episode on these days. figure depicts a histogram indicating the percentage of patients who had emetic episodes on days through . thirty-seven ( . %) patients received rescue medications. rescue medications consisted of lorazepam, prochlorperazine, promethazine, or dexamethasone. lorazepam was the most common rescue medication used. there were episodes in patients of reported nausea > mm on the vas in the day reporting period. figure depicts the median vas score on days through . no patient had a change in their planned cisplatin-based combination chemotherapy schedule. safety sixty-four patients were evaluable for toxicity. administration of fosaprepitant was well tolerated. there was a total of grade iii or iv toxicity events on this study that were possibly, probably, or definitely related to study drug. two patients had grade iii toxicity consisting of leukopenia and thrombocytopenia. four patients had grade iv toxicity consisting of leukopenia and neutropenia. one patient had grade iv toxicity consisting of febrile neutropenia. there was no cases of reaction at the infusion site attributable to fosaprepitant infusion on this study.     discussion severe cinv were common adverse events associated with certain chemotherapeutic regimens and forced a substantial number of patients to delay or even refuse potentially curative therapy [ ]. despite the compelling advances made in recent years, cinv continues to be among the most distressing and feared adverse effects of chemotherapy [ ]. prior to the introduction of ondansetron, the first ht receptor antagonist, the typical patient with testicular cancer would experience a median of emetic episodes on day of a day cisplatin-based chemotherapy course, episodes on day , and decreasing emetic episodes on later days [ ]. the introduction of ondansetron was fundamental in preventing acute nausea and vomiting associated with cisplatin-based chemotherapy [ , ]. more recently, this trend has been reversed in multi-day cisplatin as patients experience more severe symptoms of cinv on later days of the chemotherapy cycle (days - ) as well as delayed cinv on days - [ ]. the addition of dexamethasone in phase iii studies resulted in further improvement in the prevention of acute cinv [ ]. delayed cinv was not adequately controlled by ht receptor antagonists and dexamethasone [ ]. aprepitant, a nk -ra, demonstrated efficacy in controlling both acute and delayed cinv and hence it was combined with previous antiemetic regimens. aprepitant was initially given for days [ ] but eventually a day course was deemed sufficient [ ]. aprepitant proved to be efficacious in preventing cinv in a randomized double-blind placebo-controlled trial which enrolled cisplatin-naïve patients who were treated with chemotherapy regimens including cisplatin ≥ mg/m administered on a single day [ ]. in the hcrn, a phase iii study was conducted to test the efficacy of aprepitant in patients undergoing a -day cisplatin-based     combination chemotherapy regimen [ ]. in this randomized double-blind placebo-controlled crossover study, % of patients achieved a cr, defined as no emetic episodes and no use of rescue medications, with aprepitant compared to % with placebo (p < . ). an intravenous formulation of a nk -ra was developed. fosaprepitant is a water-soluble phosphoryl prodrug of aprepitant which when administered intravenously, rapidly converts to aprepitant within minutes of administration [ ]. a phase iii trial compared single dose fosaprepitant mg iv in combination with granisetron µg/kg iv and dexamethasone to a control regimen of placebo, granisetron, and dexamethasone [ ]. complete response, defined by no emesis and no rescue therapy, was higher in the fosaprepitant arm than in control arm; % vs. % (p = . ). fosaprepitant was more effective than control regimen in both acute and delayed phases. a phase iii randomized, double-blind, non-inferiority study was conducted comparing fosaprepitant to aprepitant in chemotherapy-naïve patients receiving cisplatin ≥ mg/m [ ]. a single dose of intravenous fosaprepitant was shown to be non-inferior to standard -day oral aprepitant in preventing cinv in patients receiving single-day cisplatin. however, there is a paucity of data with fosaprepitant in patients receiving a -day cisplatin combination chemotherapy regimen. the hcrn conducted this single arm phase ii study as a first evaluation of fosaprepitant in patients receiving -day cisplatin-based combination chemotherapy for gct. we have used dexamethasone only on days and of the acute phase in an attempt to decrease the adverse effects of corticosteroids. patients with gct who are receiving three to four cycles of cisplatin combination chemotherapy will be subjected to high doses of dexamethasone which will put them at risk for hyperglycemia, anxiety, agitation, insomnia, gastrointestinal irritation, and peptic ulcer disease. we attempt to avoid the potential long-term adverse effects of dexamethasone in     this young patient population such as obesity, avascular necrosis of the hip, and cataracts [ - ]. in the phase ii study reported here, fosaprepitant combined with dexamethasone and ht -ra was very well tolerated with minimal grade iii or iv toxicities. however, the cr rate observed in this study appears to be significantly lower than our previous study with aprepitant. fosaprepitant had a cr rate of . % on this study compared to a cr rate of % with aprepitant [ ]. while there may be several factors that contributed to the low cr rate, a notable one is that % of the patients were not evaluable for the primary endpoint due to incomplete patient logs. to the best of our knowledge, this is the first clinical trial evaluating fosaprepitant in patients receiving -day cisplatin-based chemotherapy regimens for gct. the data in this small phase ii study, in contrast to our prior phase iii study, appears to indicate a significantly lower cr rate with the substitution of fosaprepitant for aprepitant.     acknowledgement: -funding/support: trial supported by merck & co. inc. -all authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. references . siegel rl, miller kd, jemal a ( ) cancer statistics, . ca: a cancer journal for clinicians ( ): - . doi: . /caac. . einhorn lh ( ) curing metastatic testicular cancer. proceedings of the national academy of sciences of the united states of america ( ): - . doi: . /pnas. . bloechl-daum b, deuson rr, mavros p, hansen m, herrstedt j ( ) delayed nausea and vomiting continue to reduce patients' quality of life after highly and moderately emetogenic chemotherapy despite antiemetic treatment. journal of clinical oncology : official journal of the american society of clinical oncology ( ): - . doi: . /jco. . . . campos d, pereira jr, reinhardt rr, carracedo c, poli s, vogel c, martinez-cedillo j, erazo a, wittreich j, eriksson lo, carides ad, gertz bj ( ) prevention of cisplatin-induced emesis by the oral neurokinin- antagonist, mk- , in combination with granisetron and dexamethasone or with dexamethasone alone. journal of clinical oncology : official journal of the american society of clinical oncology ( ): - . gralla rj, de wit r, herrstedt j, carides ad, ianus j, guoguang-ma j, evans jk, horgan kj ( ) antiemetic efficacy of the neurokinin- antagonist, aprepitant, plus a ht antagonist and a corticosteroid in patients receiving anthracyclines or cyclophosphamide in addition to high-dose cisplatin: analysis of combined data from two phase iii randomized clinical trials. cancer ( ): - . doi: . /cncr. . hesketh pj, grunberg sm, gralla rj, warr dg, roila f, de wit r, chawla sp, carides ad, ianus j, elmer me, evans jk, beck k, reines s, horgan kj ( ) the oral neurokinin-     antagonist aprepitant for the prevention of chemotherapy-induced nausea and vomiting: a multinational, randomized, double-blind, placebo-controlled trial in patients receiving high-dose cisplatin--the aprepitant protocol study group. journal of clinical oncology : official journal of the american society of clinical oncology ( ): - . doi: . /jco. . . . albany c, brames mj, fausel c, johnson cs, picus j, einhorn lh ( ) randomized, double-blind, placebo-controlled, phase iii cross-over study evaluating the oral neurokinin- antagonist aprepitant in combination with a ht receptor antagonist and dexamethasone in patients with germ cell tumors receiving -day cisplatin combination chemotherapy regimens: a hoosier oncology group study. journal of clinical oncology : official journal of the american society of clinical oncology ( ): - . doi: . /jco. . . . lasseter kc, gambale j, jin b, bergman a, constanzer m, dru j, han th, majumdar a, evans jk, murphy mg ( ) tolerability of fosaprepitant and bioequivalency to aprepitant in healthy subjects. journal of clinical pharmacology ( ): - . doi: . / . grunberg s, chua d, maru a, dinis j, devandry s, boice ja, hardwick js, beckford e, taylor a, carides a, roila f, herrstedt j ( ) single-dose fosaprepitant for the prevention of chemotherapy-induced nausea and vomiting associated with cisplatin therapy: randomized, double-blind study protocol--ease. journal of clinical oncology : official journal of the american society of clinical oncology ( ): - . doi: . /jco. . . . herrstedt j ( ) nausea and emesis: still an unsolved problem in cancer patients? supportive care in cancer : official journal of the multinational association of supportive care in cancer ( ): - . doi: . /s - - - . schwartzberg ls ( ) chemotherapy-induced nausea and vomiting: clinician and patient perspectives. the journal of supportive oncology ( suppl ): - . herman ts, einhorn lh, jones se, nagy c, chester ab, dean jc, furnas b, williams sd, leigh sa, dorr rt, moon te ( ) superiority of nabilone over prochlorperazine as an antiemetic in patients receiving cancer chemotherapy. the new england journal of medicine ( ): - . doi: . /nejm . cubeddu lx, hoffmann is, fuenmayor nt, finn al ( ) efficacy of ondansetron (gr f) and the role of serotonin in cisplatin-induced nausea and vomiting. new england journal of medicine ( ): - . doi:doi: . /nejm . einhorn lh, nagy c, werner k, finn al ( ) ondansetron: a new antiemetic for patients receiving cisplatin chemotherapy. journal of clinical oncology : official journal of the american society of clinical oncology ( ): - . einhorn lh, grunberg sm, rapoport b, rittenberg c, feyer p ( ) antiemetic therapy for multiple-day chemotherapy and additional topics consisting of rescue antiemetics and high-dose chemotherapy with stem cell transplant: review and consensus statement. supportive care in cancer : official journal of the multinational association of supportive care in cancer suppl :s - . doi: . /s - - -z . hesketh pj, harvey wh, harker wg, beck tm, ryan t, bricker lj, kish ja, murphy wk, hainsworth jd, haley b ( ) a randomized, double-blind comparison of intravenous ondansetron alone and in combination with intravenous dexamethasone in the prevention of high-dose cisplatin-induced emesis. journal of clinical oncology : official journal of the american society of clinical oncology ( ): -     . grunberg sm, deuson rr, mavros p, geling o, hansen m, cruciani g, daniele b, de pouvourville g, rubenstein eb, daugaard g ( ) incidence of chemotherapy-induced nausea and emesis after modern antiemetics. cancer ( ): - . doi: . /cncr. . chawla sp, grunberg sm, gralla rj, hesketh pj, rittenberg c, elmer me, schmidt c, taylor a, carides ad, evans jk, horgan kj ( ) establishing the dose of the oral nk antagonist aprepitant for the prevention of chemotherapy-induced nausea and vomiting. cancer ( ): - . doi: . /cncr. . hale jj, mills sg, maccoss m, dorn cp, finke pe, budhu rj, reamer ra, huskey se, luffer-atlas d, dean bj, mcgowan em, feeney wp, chiu sh, cascieri ma, chicchi gg, kurtz mm, sadowski s, ber e, tattersall fd, rupniak nm, williams ar, rycroft w, hargreaves r, metzger jm, macintyre de ( ) phosphorylated morpholine acetal human neurokinin- receptor antagonists as water-soluble prodrugs. journal of medicinal chemistry ( ): - . saito h, yoshizawa h, yoshimori k, katakami n, katsumata n, kawahara m, eguchi k ( ) efficacy and safety of single-dose fosaprepitant in the prevention of chemotherapy- induced nausea and vomiting in patients receiving high-dose cisplatin: a multicentre, randomised, double-blind, placebo-controlled phase trial. annals of oncology : official journal of the european society for medical oncology / esmo ( ): - . doi: . /annonc/mds . cook am, dzik-jurasz as, padhani ar, norman a, huddart ra ( ) the prevalence of avascular necrosis in patients treated with chemotherapy for testicular tumours. british journal of cancer ( ): - . doi: . /bjoc. . . jobling ai, augusteyn rc ( ) what causes steroid cataracts? a review of steroid- induced posterior subcapsular cataracts. clinical & experimental optometry ( ): - . vardy j, chiew ks, galica j, pond gr, tannock if ( ) side effects associated with the use of dexamethasone for prophylaxis of delayed emesis after moderately emetogenic chemotherapy. british journal of cancer ( ): - . doi: . /sj.bjc. -conflicts of interest: nabil adra: none costantine albany: none mary j. brames: none somer case-eads: none cynthia calley: husband works for lilly and has lilly stock in his retirement account. ziyue liu: none     christopher a. fausel: none timothy breen: none nasser h. hanna: received grant support from merck for another study. ralph j. hauke: none joel picus: none lawrence h. einhorn: none   figure . percentage of complete response (cr) in acute, delayed, and overall phase     figure . percentage of patients with emetic episodes on days through   figure . median vas (visual analog scale) score on days through acute delayed overall cr p e rc e n t percent of patients with complete response day p e rc e n t day percent of patients with an emetic episode by day     table : study drug schedule day day day day day days – dexamethasone  mg po dexamethasone  mg po fosaprepitant  mg iv  - fosaprepitant  mg iv dexamethasone  mg po bid palonosetron .  mg iv   palonosetron .  mg iv -  palonosetron .  mg iv   *alternatively, ondansetron mg po bid can be utilized on days to if palonosetron is not available. abbreviations: po, orally; iv, intravenous; bid, twice daily   table . patient and treatment characteristics v a s  s co re day median vas score by day     all patients (n= ) evaluable patients (n= ) median range median range age - - n % n % sex male . . race caucasian . . ethnicity non-hispanic . . hispanic . . ecog ps . . . . . . stage i . . ii . . iii . . is . . unknown . . prior chemotherapy yes . . no . . chemotherapy regimen bep . . ep . . veip . . cisplatin/epirubicin . . bep=bleomycin, etoposide, cisplatin; ep=etoposide, cisplatin; veip=vinblastine, ifosfamide, cisplatin revisiting the swcs international annual conference a journal of soil and water conservationsept/oct —vol. , no. jim gulliford revisiting the swcs international annual conference doi: . /jswc. . . a jim gulliford is executive director of the soil and water conservation society, ankeny, iowa. m any great cities of the world are located on the banks of similarly great rivers. louisville, kentucky, is no exception, ranking th in us popula- tion and established immediately upstream of the falls of the ohio river. it was here that early river travelers paused to contem- plate the challenge of navigating the falls. it was here that more than conserva- tionists gathered for the st international annual conference of the soil and water conservation society (swcs) to consider the challenge of “managing great river landscapes.” from the world’s largest riv- ers to the smallest upstream landscape features, understanding the hydrology of and human influence on agricultural and urban landscapes is important to address the challenges of agricultural production, resource conservation, and environmen- tal quality protection. this theme was the focus of this year’s outstanding conference. plenary sessions and panels the conference began with the fellows forum, which examined the challenge of managing great rivers and the impact of those management decisions on urban and agricultural landscapes. flood events on great rivers have profound impacts on communities and farms that exist in floodplains. presentations described his- toric flood events on both the ohio and mississippi rivers, flood management plans that existed (or didn’t exist), and the timing and effect of decisions made imple- menting those plans. watch for a new book, managing mississippi and ohio river landscapes, by forum moderators kenneth olson and lois wright morton that will be released by swcs this year. our goal for each annual conference is to use the plenary sessions to inform, challenge, and inspire attendees. chad pregracke delivered this year’s pritchard lecture to open the conference and deliv- ered the goods as well. observing the unsightly and toxic accumulation of trash in our rivers, he began a one-man cru- sade to clean america’s rivers. he shared his story and conveyed the importance of each individual’s contribution when faced with a great resource conservation challenge. if you missed chad’s presenta- tion, take a moment to visit the living lands and waters website (http://living- landsandwaters.org/). you will gain an appreciation for the accomplishments they have achieved and see the parallel to the work that each of us contributes to soil and water conservation. day two featured two plenary panels. nick goeser returned for a third year with an update on the soil health partnership. joint presentations by farmers and their crop consultants provided insight into the agronomic and conservation relationship that is important as farmers commit to improving soil health. the second plenary examined the challenges conservation professionals have assisting farmers as they address impaired waters under the clean water act. an important take-away message from this plenary was that conser- vation practice standards must be specific enough to address the water quality objec- tive and flexible enough to fit into the specific landscape being farmed. conservation science learning, networking, and recognition opportunities an important feature of the con- ference each year is the breadth of learning opportunities available to par- ticipants. workshops provide in-depth training; tours provide access to unique, local conservation activities; and poster presentations provide an opportunity for one-on-one interaction between students, scientists, and participants regarding spe- cific research or project activities. the exhibit hall offers agribusiness, agency, and conservation organizations a space to connect with participants and share their products, services, and education materials. all of these activities make the conference a complete professional development and learning experience. the society recognized jerry hatfield as this year’s recipient of the hugh hammond bennett award and con- ferred the honor of swcs fellow upon kenneth olson. the hoosier, north dakota, michigan, and hugh hammond bennett chapters were recognized for their accomplishments and sustained con- servation performance. individuals were recognized for their conservation achieve- ments, journal of soil and water conservation contributions, and swcs leadership. usda natural resources conservation service (nrcs) chief jason weller addressed the awards luncheon and spoke to the challenge and importance of conservation planning and voluntary con- servation service to farmers and ranchers. his presentation added important empha- sis to the tuesday plenary presentations on soil health and water quality challenges. the society’s first photo contest was a smashing success! initial entrees were screened to a final set of —all of which could have been winners. in the end, con- ference voting resulted in a first place tie. the winning photos can be seen on the swcs conference website, and watch swcs conference echoes jim gulliford c o p yrig h t © s o il a n d w a ter c o n serva tio n s o ciety. a ll rig h ts reserved . w w w .sw cs.org ( ): a - a jo u rn a l o f s o il a n d w a ter c o n serva tio n http://www.swcs.org asept/oct —vol. , no. journal of soil and water conservation future journal covers as you may see some of the entries there as well. the annual conference also provides an opportunity for swcs partner orga- nizations to “piggyback” their activities around or during the conference. this year, the nrcs conservation innovation grant (cig) program featured six project presentations. over the years, the cig pro- gram has matured into an excellent locally led, “science to practice” opportunity for conservation districts and organizations to test their creative approaches to conserva- tion challenges. a second organization that joins us each year for a luncheon and busi- ness meeting is the association of retired conservation service employees. these individuals may be retired, but they have not lost their passion for conservation and the society. they enjoy fellowship with each other but also engage the conference with their views. participant-driven success finally, it is important to understand that there are two essential elements to a suc- cessful conference: you must have a great program and a great participating audience. special thanks are due to many individuals and organizations for their contributions. deanna osmond chaired the program committee that built the conference around its theme. the kentucky bluegrass chapter members were remarkable hosts. they arranged tours, made us all feel wel- come, and by my observation had the most fun of all our participants. the contribu- tions of syngenta and all of our sponsors and exhibitors helped us engage agri- culture and agribusiness as conservation partners, and made our networking events successful. and again, the most important contributions came from the more than conference participants from states; washington, dc; canadian provinces; additional countries; guam; and puerto rico. these individuals contributed more than presentations in workshops, sym- posia, individual presentations, and posters. this audience diversity brought a rich- ness to the program, conversation, and networking that is only experienced by conference participants. in closing, i can report that plans are well underway for next year’s conference, which will be held july through august , , in madison, wisconsin. watch the society website this fall for a call for abstracts and conference arrangements. plan now to bring your conservation voice to madison. c o p yrig h t © s o il a n d w a ter c o n serva tio n s o ciety. a ll rig h ts reserved . w w w .sw cs.org ( ): a - a jo u rn a l o f s o il a n d w a ter c o n serva tio n http://www.swcs.org journal of environmental management ( ) , - . simulating the effects of alternative forest management strategies _l_ on landscape structure _ •l i eric j. gustafson and thomas r. crow x u.s. department of agriculture, forest service, north central forest experiment station, forestry sciences laboratory, rhinelander, wi , u.s.a. received february quantitative, spatial tools are needed to assess the long-term spatial consequences -., i:i.... " of alternative management strategies for land use planning and resource . ::_?i,__ ,_,. management. we constructed a timber harvest allocation model (harvest) ;.!_;_:i;_, that provides a visual and quantitative means to predict the spatial pattern of . ._%.,;_-_,. . , forest openings produced by alternative harvest strategies. harvest was used to investigate the effect on landscape structure of alternate management plans • _i/i.; " •formulated for the hoosier national forest (hnf) in southern indiana, u.s.a. ..... :- , the forest plan for the hnf specified primarily clear,cutting across most of the forest, and the amended forest plan specified primarily group selection (removal of small groups of trees), limited to a much smaller portion of the forest. we assessed the relative importance of variation in the extent and intensity of o timber harvest on landscape spatial pattern. we simulated years for each alternative and calculated several measures of forest spatial pattern. the total area of forest interior and the mean size of forest interior blocks declined more under the plan than under the plan, primarily because of the greater _ _" area dedicated to timber production in the plan. despite the % decrease in timber production in the plan, that plan produced almost as much forest edge as the plan due to heavy reliance on small harvests with large perimeter-to-area ratios. both plans resulted in an even distribution of age classes ' /j up to the age equal to rotation length, a dramatic decline in mid-age classes, and a ., ,_ / large area of mature forest. the restriction of harvest activity to more limited _, areas in the plan appeared to have a greater effect on forest spatial pattern than did the differences in harvest intensity of the two plans. harvest provides a tool to link planning alternatives with potential changes in landscape structure, commodity produotion, and other resource values that are spatially dependent. © academicpress limited keywordsi forest management planning, landscape ecology, simulation models,. habitat fragmentation_ forest interior, forest edge, clearcutting, age-class distribution, biological diversity, spatial pattern, scale. . introduction provisions of the national forest management act of (nfma) require the u.s. department of agriculture forest serviceto completea plan for each national forest . - / / + $ . / © academicpress limited • ° l" . ) \. ' • [] ' -.- .. • _ . . . . • . . : .... , • ,, .. simulating forest management strategies '\ " that constitutes a comprehensive statement of management direction. the contents of a forest plan include forest-wide, multiple-use goals and objectives, standards and guidelines for fulfilling these goals and objectives, projections of timber harvest quant- ities, andprocedures for monitoring management impacts. the forest plan is a guiding document, covering a -year period, that provides the broad framework for more site- specific analysis and project implementation. nfma requires the preparation of an environmental impact statement as part of the planning process, including the de- iineation and evaluation of multiple management alternatives (usda forest service, a). because there are ecological consequences of creating different landscape patterns _'_: ' , , ', in space and time, quantitative, spatial tools are needed to assess the long-term _: _ .: spatial consequences of alternative management strategies (franklin and forman, ; -_:_,:_:_'_ ...._,'_....".. hemstrom and cissel, ; li et al., ; thompson, ). to this end, we have ::_:_::_:::_;',i:_'_:_: constructed a timber harvest allocation model (harvest) that provides a visual and •-i:!_:;;:%ii::_i.'.. _" .!i__._ . , quantitative means to predict the spatial pattern of forest openings produced under _-_,_.....:__ . alternative harvest strategies (gustafson and crow, ). the model allows simulation :_i¢!-_.:_ '..-_, . of differences in the size of timber harvest units, the total area harvested, rotation _:_:::_:_:*_:_ _- lengths, and the spatial distribution of harvested areas. in this paper, we present a case _::_.... =_.t_i):::......_, study in which we used harvest to predict the effects of published management ;i,;_ _/- alternatives for the hoosier national forest (hnf) on the spatial pattern of forest "_ ' __ openings, and we assessed the relative importance of variation in ha_est intensity and ;_. ,_!i_:_i!:_ ' _" management area configuration. 'i_..... _y_ _, in' , the hnf in southern indiana released its forest plan. even-age manage- _}_'._ _:__- ment, primarily clearcutting, was emphasized in this plan, and reaction of some segments i :_::_i!_ _;i__ of the public was swift and clear: clearcutting on % of the land base was not an _ _ _' . acceptable option. as a result, the forest plan was amended in changing the management emphasis from even to uneven-age management and reducing the timber : _ ' base to about % of the forest. the amended plan proposed that timber harvest and . _ . ' vegetation management be applied "within the context of perpel_uating and enhancing ._ _ .._, biological diversity at different spatial scales and of differing desired conditions in different management units" (usda forest service, ). a primary silvicultural tool for implementing this new management direction was group selection (i.e. the harvesting "- ,. /,_ of small forest patches no more than . ha in size). furthermore, in the amended plan, / emphasis was placed on creating interior forest conditions and reducing the amount of __ _" forest edge. _ however, these goals are not always easy to achieve because of existing conditions. the current composition and structure of a landscape reflect the physical environment, .past land uses, and the interaction between the two (crow, ; baker, ). past . forest treatments often create age structures that are fragmented and distributed in small patches (ripple et al., ; mladenoff et al., ). federal ownership, especially _ . in the eastern u.s., is often fragmented and interspersed with commonly up to % private ownership, creating management problems for both ownerships and restricting some management options. furthermore, the effectiveness of the change in management strategy in achieving the desired future condition is difficult to predict without spatial information and without the ability to project change in spatial patterns with time. .. in our study, we used harvest to simulate management alternatives for the hnf, using stand maps compiled by the hnf for initial conditions. we used the standards and guidelines provided in the hoosier forest plan (primarily clear- cutting) and the amended plan (primarily group selection) as rule-bases for the i_ .... _%_ii_:_\i_=_:_i_ _ ......... e. j. gustafsonandt. r. crow [ \ model. our objectives were to ( ) characterize the differences in landscape structure over time that result from the implementation of these fundamentally different management v_._ ......., strategies, and ( ) determine the relative effects of differences in harvest intensity and the spatial extent of harvested areas. . methods . . _w_st' :_'__:i_._ __:i_':_ harvest was constructed to allow the input of specific rules to allocate forest stands _.:_' :_:_:_:_"=::_ for even-age harvest (clearcuts and shelterwood) and group selection, using parameters _. - commonly found in national forest plan standards and guidelines. the model produces ' _ ._ landscape patterns that have spatial attributes resulting from the initial landscape " :...... .. conditions and the proposed management activities. the model is simplistic in that it does not attempt to optimize timber production or quality, nor does it predict the , _ _:_ _ . specific locations of future harvest activity, as it ignores many considerations such as .....r:._ ._::' .... . " visual objectives and road access. instead, the model mimics stochastically the allocation '_ _ '_,: _" of stands for harvest by forest planners within the constraints of the broad management • - :_ili_:_;::. . • strategies. modelling this process allows experimentation to link variation in man- :_=:_ _. _.. agement parameters with the resulting landscape patterns. in our study, we considered [:-:i_i__ ' _:i:_ -only harvests that produce openings within the forest. harvest by single-tree selection .......was not modelled because such methods usually maintain an essentially closed canopy .... _ _ ,_. harvest. was constructed to run within erdas v. . (erdas, inc., atlanta, ga, u.s.a.) geographic information system (gis) software using erdas toolkit __:_;!' "_ (fortran) routines l%r input and output. erdas is a grid-cell gis that allows :_ _:_:_'_;_/i_ flexible display and manipulation of digital maps. timber harvest allocations were made by the. model using a digital stand map, where grid-cell values reflect the age (in _. years) .of each timber stand. the model allows control of the size distribution of • harvests, the total area of forests to be harvested, the rotation length (by specifying the minimum age on the input stand map where harvests may be allocated), and the ._ a, width of buffers left between adjacent harvests and between harvests and non-forest habitats. harvest selects stands randomly for harvest, checking first to ensure that _ _ the stand is old enough to meet rotation length requirements. the assumption that ." past _harvest allocations have been spatially random was tested, and that test is described , , / below. grou p selection is implemented on the hnf such that one-sixth of a stand is /i_: cut on each entry, and entries occur every - years (usda forest service, ; t. thai(e, pers. comm.). the model randomly selects stands for group selection from those stands with an age greater than the prescribed rotation length, and tracks the stands managed by group selection, ensuring re-entry at year intervals and preventing• . . other treatments in those stands. within group-selected stands, openings are placed randomly, with at least m ( pixel) between openings. _ " . . svtroy areas• simulations '_ere conducted at two spatial scales. input data at m resolution allowed simulation of openings within stands and enabled detailed, stand-level analysis to be " made of the ecological pattern produced by the management alternatives. data with m resolution allowed analysis of how the management alternatives would produce landscape patterns across an entire national forest. areas to be studied at m simulating forest management strategies '\ resolution-were selected in three of the four administrative units of the hnf [pleasant run study area (prun, ha), lost river (lriv, ha), and tell city (tell, ha)]. stand age maps of national forest land within the three study areas were digitized from paper-based and mylar planning maps and gridded to m cells, and ages were calculated as of . the two management plans included maps delineating management area (ma) boundaries. each ma has a specific management objective, the ma boundaries encompass tracts to be managed to meet that objective, n and several disjunct polygons of a particular ma may be designated within an _,:_ ,: .... administrative unit. the objectives of each ma and the decimal designations (e.g. . ) );,iii,_:,:i:..... are used consistently among all national forests. ma boundaries were manually /_ '_":_''. " -b transferred from the forest plan maps (approximate scale, " ) to " ., _. " scale u.s. geological survey (usgs) maps and digitized. land use on non-forest _... service land was derived from landsat thematic mapper (tm) imagery collected in _: :., , as described in gustafson and crow ( ). the lost river study area en- !i_i•_+_:!_:_i. compassed most of martin state forest (msf), and limited group selection ( . ha/ ......:_,, ., . decade) was simulated on msf land. "_-i__y simulations were conducted using maps with m resolution on a rectangular _ _,_ .,, - study area ( ha) that included the entire hnf purchase area. ownership :i'_ _: _!_,,_ boundaries were digitized from : scale paper maps produced by usgs for the _ • _ :-• forest service. because it was not feasible to digitize stand maps for the entire forest, :u_ _, _:i_/. , .- :we generated a forest cover map by resampling usgs-land use bata acquisition i!! ;_ _ii!i!!i_.i_ (luda) data ( -m cells) to produce a map with m resolution which allowed _iii_i -- allocation of harvests as small as ha. we assumed that the luda data represented forest cover at 'the start of the simulations ( ), even though they were derived from _._:..:._i_,__,,_,._!.......... aerial photography acquired in . we assumed that stand ages were distributed _!_'_ •_ ''_ • randomly across the forest (see test of this assumption below), and we began the o simulations with a homogeneous stand age map (i.e. all stands = years). " ' because the mean size of group selection openings under both plans was less than the cell size ( ha), we attempted to mimic the pattern produced by group selection by . _, allocating pixels ( x ) to represent each group selection stand. nine pixels ( ha) reflects the average size of hnf stands. because the prescription calls for one-sixth of a stand to be harvested on each entry ( pixels/ = . pixels) and fractions of pixels /,_ . could not be allocated, the model allocated either or pixels on each entry. each , _ ./ pixel allocated represented a group of groups (i.e. pixel dominated by group openings). ;_ _: this algorithm thus modelled the location of group selected stands and produced small openings at -year intervals in those stands, but some of the fine-grain spatial detail of group selection was lost at this scale. management on the msf was not simulated ' at this scale. • . . we examined the assumption of random distribution of stand ages by conducting • a nearest neighbor analysis (davis, ) on ten subareas of the ( m) stand age maps _ " in all three study areas, randomly selected from areas where there were few gaps in the stand maps (i.e. little privately-owned land). we conducted two analyses: ( ) to determine the distribution of stands allocated by hnf planners, we analyzed stands between - years on five of the subareas; and ( ) to determine the distribution of stands reaching rotation age in the most recent decade, we analyzed stands between - years on the remaining five subareas. to eliminate edge effects, nearest neighbors outside the subareas were included in the analysis. this analysis required point data, so we identified the centroid of each stand, and used fragstats . (mcgarigal :." , • . . h e. j. gustafson and t. r. crow and marks, ) to calculate the mean nearest neighbor distance (_r) of these centroids. the expected mean distance between nearest neighbors is: __,,_ _ , _=_ (i) with standard error: - :_: n .. . a ss=q n ( ) ,> ,+,_!_i":',,.:,. v ) where a is the area of the subset and n is the number of stand centroids. the ratio: -- . _ i) r=d / _ ( ) _............ is the nearest neighbor statistic and ranges from " for a distribution where all points _ " ' .... . .., ; , . are in the same location to . for a random distribution, to a maximum of . for a ....." ' . " -• ,r regular hexagonal pattern. we constructed a two-tailed z-test for each subset to test --, ... the null hypothesis that the stands are distributed randomly (ho: r = . ): "_:_" " z=(d-_)/s_, ( ) " rejecting.the null with a confidence coefficient ( -a) of " when ]z]> . . the results . '_,, ' _- of this simple analysis (table ) suggest that a random allocation of treatments is consistent with historical allocations, and that random allocations for the m " simulations should approximate those that would beproduced if actual stand age maps ._,_ ,,.............. were available, .-, (-,-?_!"_ e . i) . . experimental design we simulated specific-management alternatives on each study area: (i) the forest .plan (usda forest service, b), ( ) the forest plan amendment - _)" (usda forest service ), ( ) the management area (ma) boundaries of the plan wi,th the harvest intensity of the amended plan ( intensity- map), ( ) • , .. //the ma boundaries of the plan amendment with the harvest intensity of the plan ( map- intensity), and ( ) no harvest (i.e. no openings produced or maintained , _ /" on national forest land). the map- intensity scenario used the parameters of _: ma "!, which is the most intense of the plan harvest scenarios (table ), to provide the greatest contrast with the plan harvest intensity (ma . ). we simulated years of management under each alternative and produced three replicates .. of each simulation. the variability of the spatial pattern of the simulations was very low, and three replicates were deemed adequate to ensure robust results. because the harvesting rules differed among mas under the plan, each ma was simulated separately, and the maps output by the model were combined to produce a complete stand age map for each decade. wildlife openings were maintained throughout the simulations except in the "no-harvest" scenario where conversion to forest was simulated and closed canopy forest was achieved in two decades. to avoid confounding our investigation of the effects of alternative management strategies on'public lands, we did not simulate harvest activities on privately-owned inholdings. the existing pattern of forest openings on private land was derived from• ._ the tm imagery, and these conditions were maintained throughout the simulations. -.:r_¸__'_'.....,_ " ......... ' ,<:_<ii_'_._i_._<', ' ....i!i_:i;_:_¸ i,' e. j. gustafson and ".r. crow \' " table . harvest intensities as derived from the hoosier national forest plan and the° forest plan amendment ma ma model parameter . - . . total - mean clearcut opening size (ha) - . . . i mean group opening size (ha) - m m - maximum opening size (ha) - . . . m . total harvested/decade (ha) . " . - . . - _:i_:, harvest rate/decade (%) . - . . " . : _, '_..... , timber yield/decade (mbf) .... /:_i___'.__' * rotation length (years)- ,/' : ....."_ bufferwidth (m) " ....._..........._ forest plan management areas. the decimal designation of mas is that used by the forest service. !i_'i_';'__:i:_:.:i_ . forest plan amendment management areas. , ............ • " " "_i!;:_! _i_:_.__ . - • - represents harvest activity by management area across the entire forest. :_,._,,:_._ ._ .... represents percent of forest within the management area that is harvested each decade. the vfidth of buffers left between harvest allocations and other harvests, streams, and openings. :i!"-'_+_u_...... .. _,__., , , . spatial analysis :_:'_ ....__:....'*_ to assess the pattern of forest openings produced by the management alternatives, we :_ __.... - assumed that forest pixels _< years old functioned as openings, and all other forest _::,......... _,:-_:,,_.... .pixels were considered closed-canopy forest. because the management alternatives ,i__'_:_remained constant throughout the -year simulations, the relative difference in the , equilibrium pattern among alternatives would hold even if openings were defined ] _,:_i_,_ differently. using a gis proximity function, we produced maps of forest interior habitat ! _•"__?_!::_ [forest pixels > m from an opening (dellasalla and rabe, ; andren and ,. angelstam, )] and calculated the area of forest interior, the mean size of contiguous blocks of forest interior and linear forest edge at each decade for each alternative. at , _ m resolution,, forest interior was defined as forest pixels > m from an opening. ._ _ __, each index of spatial pattern was plotted as a function of time. to establish the pre-simulation trend, we successively recoded the stand age maps by decrementing the. age value (a) at time t for each pixel j by to produce the stand age map of the "' : _" previous decade (i.e. aj(t_lo)=aj(t)- ). this was done four times to establish the ; /: pattern of forest openings since . stands that reached an age of zero (i.e. harvested) _ duri.ngthis process were assumed to have been mature forest (> years) before that time. also. tabulated throughout the simulations were the stand age-class distributions •on:hnf land. to illustrate long-term age-class distribution trends clearly, we simulated " the. and plan scenarios for an additional years (a total of years) on the tel! city study.area. _ .• , to evaluate the relative effects of harvest intensity and the zonation of harvest activity, a repeated measures anova was used to test for treatment effects reflecting harvest, intensities (intensity), management area boundaries (map) and time • (decade)., the time periods were included in the analysis to account for potential autocorrelation among spatial pattern measures across successive decades. . • . results • the replications of the simulations produced little variability. harvest levels in all mas simulating forest managementstrategies were high enough that most stands >rotation age were allocated each decade. variability would be likely to be higher if harvest rates in timber production management areas were lower, allowing stochasticity in harvest placement to have more of an effect on the resulting pattern of openings. even the plan had a high intensity of harvest in ma . , because the use of groups of small openings required the allocation of manystands to achieve the specified timber production. although harvest intensity was _;__i high within ma " , much less of the land base was dedicated to timber production. alternatives. with lower harvest intensities within mas would result in greater spatial :'_i_ _,_......._,._:,_,,, pattern variability. the variability in our results was too low to show clearly with error bars on histograms and line graphs, so error bars are not shown. the standard deviation _!_i:_i!l!_ii:' from the mean of interior area produced by three replicates never exceeded . ha in . -. " any combination of treatments, and the standard deviation from mean linear edge never exceeded . km. _:_:_.:__+_ . . spatial pattern of forest openings :::',_-!-;"' _': " ",, ,_" , scenarios that used the plan ma boundaries ( plan, map- intensity) _ ......_/" - showed a decline in the total area of forest interior habitat (forest > m from an i _¢" " • edge) on all study sites, and the decline was greater than for those scenarios that used _ i_ _. :-_:, the plan boundaries ( plan, map- intensity, figure ). this was due .:.:iii.......... n_............... primarily to a greater proportion of the total area dedicated to tinaber production _ '_:_it_i_!_._ " under the plan (see table ). the total area of forest interior habitat decreased _;!_ _'_'- under "scenarios that used the plan boundaries, except on the tell site (figure :i!_:_!i,!i!ii_-ii_i_i _,- ). this site was different because timber harvest activity had been greater than on the .-,,_i:_:,,_ ...... other sites in recent decades and because it included the greatest proportion of area ,i_.....,_!_: _!_' . in management areas where timber harvest was excluded. the relationship among ° management scenarios was similar when simulated at m resolution over the entire • ' hnf area (figure ) and suggests that the prun and lriv sites are more representative of the forest as a whole .than the tell site in terms of forest interior characteristics. -_ . _, • the decline in forest interior at the start of the m simulations was steeper than in the m resolution simulations because the simulations produced a pattern of openings with a finer grain than the original m resolution map. , /,_ the mean size of forest interior blocks (not plotted) changed through time in a _ / pattern very similar to that of total area of forest interior (figure ) although variability . _ wassomewhat greater. even though the total area of forest interior habitat remained _ . fairly constant after the (figure ), forest interior habitat on these landscapes ' '. •occurred as a shifting mosaic through time. the location and size of blocks of forest , interior habitat were dynamic as harvest activity removed some blocks and re-growth • " replaced others .(figure ). all scenarios except "no-harvest" showed an increase in the amount of linear forest ' edge (figure ). the amount of edge produced by the plan was nearly as high as that produced by the plan due to a heavy reliance on group selection in the .plan, a practice that produces great amounts of edge relative to the area harvested (gus.tafson and crow, ). note the very large increase in forest edge of the map- intensity scenario as a result of implementing group selection over a greater • proportion of the forest. forest edge over the entire hnf area ( m resolution) increased much more under the scenarios using the plan management boundaries than scenarios using the, plan boundaries (figure ). this result is different from the results of the simulation at m resolution (figure ), because the model was not '." , - • , . e. j. gustafson and t. r. crow \' • . " prun - " "---" --"• *- "- -" "-- -" "- "-- -" ooo gm m m m m m m m m n mm m - !::__._ •_ . i :. _i_'" , , , i i , , i i , , i i , , i , , i .- ¢ . " lrfv ooo :{ •,., .wl. ..... • _ _'_', ., "t_ .... . " "._ : " . . • _ . • ', • i ., .. i _ tell " "- -" "- -" "- "- -" "- -" "- "- -" " " ' " . ioooo - _.,'z-:_lt>.,, ...........,_ ........ ' _ i m i _ _ i _ _ _ m i i t- .- t , -- "",, -,- i i i ,. _. year figure . changes in the amount of forest interior habitat (forest > m from an edge) over time resulting .' /,_'_ from simulation of management strategies at m resolution. simulations began at year . prun, lriv, _ /._ and tell are study areas. • = no harvest; i = plan; • = plan; o = map- intensity; f- = , map- intensity. ;l_: , •able to adequately simulate at m resolution the fine-scale pattern of openings " associated with group selection. • , . . statisticalanalysis the restriction of harvest activity to a more limited area in the plan appeared to • have a greater effect on the measures of forest spatial pattern than did the differences in harvest intensity of the two plans. although all of the main effects were highly significant in the anova models, examination of the sums of squares showed that the zonation of fiarvest activity (map) explained - % of the variance of forest interior, while differences in harvest intensity (intensity) explained only < - % and decade explained - %, depending on the site (tables and ). figures , , .._,:_,_:,-_ _._%, _::_::_;_,_;._?_:.,!c :¢_::_,_!_,.':.?_:i!_?,i_:_:. • simulating forest management strategies " entire hnf • _,-. "° <�~�:.. year _ " figure . changes in the amount of forest interior habitat (forest > m from an edge) over time resulting _ _ " _ _. from simulation of management strategies across the entire hnf at m resolution. simulations began :_: .... _:_: assuming a homogeneous year-old forest. symbols as for figure . _: :_, !_.... ._ ... and show that the curves representing a common plan map are clustered. the main " ' -" effects are less important in explaining the length of forest edge, but the zonation of harvest activity (map) explained - % of the variance, while harvest intensity explained only - % and decade explained - %. the curves fn figures and _ are not clustered as much by plan map, but the use of group selection in the f plan consistently increased the amount of forest edge. ,. . . stand age class distributions . e simulation of the plan resulted in an even distribution of young age classes (figure • ,_ ) related to the rotation lengths specified in the plan. approximately two-thirds of the timber base was to be managed with a rotation length of years, and the remainder .. _, was to. be managed with a rotation length of years (table ). in the mas managed for timber, most stands were harvested at rotation age which, over time, caused a dramatic drop in the abundance of mid-age classes. a significant number of very old . stands resulted in areas where timber harvesting was restricted. it should be noted here that we maintained a static management strategy throughout our simulations, while __/ /' national forest managers are required by law to review their management plans every _ years. our results show that static management would in fact produce striking . consequences for age class distribution. the plan showed a great reduction in even-aged stands, and a preponderance of uneven-aged and old-growth stands (figure ). recallthat our simulations did not consider single-tree selection, and a few of these older stands may in fact have undergone multiple single-tree selection treatments. _ . • ' however, most of these stands are in areas set aside from any timber harvest, and these would be true old-growth stands. . discussion harvest is a versatile model to simulate timber harvest management alternatives, when 'combined with the user interface of the erdas gis system. using harvest, we simulated two published management alternatives for the hnf and three variants to assess how those plans might be expected to impact forest interior habitat and forest • • o [ __l....i _....._ii,. ,, . e. j. gustafson and t. r. crow l ! i i " km a- '_i";'_ _ i,',_,_._--__'___ m _ • .. ": :.> • ,, ,_".e' .k_:¢< _'. • - ,; _ " "_ / •"_.-, _ . ,_.. "_. ,_ "/ • _ • ._. , n figure maps of.forest interior habitat (forest > m from an edge) during decades , , and (a, b, and ¢, respectively) of simulation of the plan on the tell study area, demonstrating the shifting mosaic nature of interior habitat through time. i- = non-forest; _ = edge; _ = interior. [ i " -- "_ ...... j__....... i .... ;-- . _ [_jljl.j '\ . simulating forest managementstrategies " prun __. "-_ .......... _, . ii ,-iy • "x _ , _dlil .. ,, '_ :/_ lriv_ i - 'rd_ , * t,,, :p . . ' ' j :'_!':_ _:_" " ___:: '".... tell " : ":_:...., ,-,_:_,_:?_,:_. _._".-., o e .. . lip, year figure . changes in the amount of linear forest edge over time resulting from simulation of management ' /_ strategies at m resolution. simulations began at year . prun, lriv, and tell are study areas. symbols as for figure . - _ / j edge over time, and to evaluate the relative effects of timber harvest intensity and zonation of timber harvest on forest spatial pattern. " projections of landscape structure and commodity outputs using the plan and •the amended plan as the rule-bases for the model suggest that the nearly % reduction in timber production under the plan would not improve the spatial components of habitat for interior-forest species as much as might be expected. only .small reductions in forest edge and relatively small increases in forest interior occurred -during the -year simulation, due to the change in management approach of the plan. widely distributed small harvest units, with their large perimeter-to-area ratios, result in large amounts of forest edge even when it is assumed that edges are ephemeral. using the harvest model, gustafson and crow ( ) found that the amount of forest edge increased dramatically and amount of forest interior decreased .dramatically when the size of the cutting edge unit was reduced to ha or less. this • , _, . w, , "" "_ _ " _ '" t . • _ ' _ " l ' ..... ..... :i ii: e. j. gustafson and t. r. crow \ - ° . • /, , ", z. -, -, -i -, , , i , , , h :ii_i::i_:::_ "_ e _. year .. -. • .... : v figure . changes in the amount of linear forest edge over time resulting from simulation of management _:_:,(_. _;;_. strategies across the entire hnf at j m resolution. symbols as for figure ....... _:_i.i_:_" ' .r , table . analysis of variance comparing the effect of harvest intensity (intensity), the ,......._.;_::_;:: . . zonation, of harvest activity by management area boundaries (map), and the time period . -:_:-.............'_:"_:_"_",:._ '" simulated (decade) on the area of forest interior and linear forest edge maintained on the _- _'_ . landscape. analysis included three replicates of simulations conducted on the entire hnf at -;_{_ i_}i " " " m resolution for decades •:ii....... _-,,...... • forest interior (ha) forest edge (km) , _,?_. source df ss, f prob>f r ss f prob>f r _._:,.<_;,_: ...... map . e+ . " . e+ . " -- intensity . . e + . . . e + . - -. decade . . e+ . " . e+ " . _ * error . e + _ . e + _ _ , " total . e+. _ . . e+ _ . ,, trend held even with significant reductions in total harvested area and clustering of _ / cutting units. - ¢ consolidation of management activities appears to have a more important effect '_:• , on spatial pattern than changes in intensity of harvest. forest interior is especially sensitive .to the management area boundaries (identified as map in tables and ). this,ltrendis confirmed by figures and , where projections of forest interior area • ' are clustered by common management area boundaries (map). much of the gain in forest interior related to the plan resulted from withdrawing lands from timber _ . production. very rtle interior forest remained or was created in production areas where harvesting occurred under either plan. the similarity in forest interior over time -y between the. plan and the map- intensity scenario indicates that high intensity • timber production can be sustained and yet maintain levels of forest interior similar to that of a lower intensity, group-selection strategy if timber production is concentrated . • in only a part. of the landscape. comparing the effects of harvest intensity and zonation of harvest intensity activity by management area boundaries also showed zonation to be more important than harvest intensity in terms of linear forest edge maintained on the landscape, but not to the same extent as forest interior. the increase in edge .' ,. , ¢, . simulating forest management strategies table . analysis of variance comparing the effects of harvest intensity (intensity), the zonation of harvest activity by management area boundaries (map), and the time period simulated (decade) on the area of forest interior and linear forest edge maintained on the landscape. analysis included three replicates of simulations conducted on the three study areas within the hnf at m resolution for- decades forest interior (ha) forest edge (km) / sourc ss ss ..... (a) lriv - . . . . _ :: _:_:_i , map ' ,. intensity . . . . - _ • " decade . - . . . _ i_i • • . ,.,--: _, error _ _ - _ _ i_ _ i _: total ---- - . _ . ii:_,i!_.ii_+_: (b) ppxun " " _ " " " _ ._:::_ _: " map _," 'intensity - . . - . _' __ :?; decade . - " - " _ _+_:::::_:_:_:_," :" " error _ _ - _ _ _,,., , total . -- -- - . -- . - _ i_ .... _!?_ _ (e) tell " " _ " : - f_! intensity - " _ - " " decade " - _ " " " '_ error . _ ...._ :_..... total " _ . . -- . t, . • \ ,: .-_-" produced by group selection under the plan appears to have offset the reduced area where harvesting occurred. if minimizing forest fragmentation across the entire forest is a goal, these results • /,,,i" suggest that the most effective approach to reaching this goal is to establish areas of __ ./" undisturbed forest with continuous canopy adjacent to forests with relatively high j harvest intensity, rather than simply reducing cutting intensity across the entire forest. • . •the map-- intensity scenario illustrates the effect on the area of forest interior of a reduction in cutting intensity across a large area, and it differs little from the high- . intensity plan scenario (figures and ). for the lake states region, solheim et al. ( ) advocate establishing large, contiguous blocks of undisturbed forest (i.e. no . harvest) to minimize edge effects and to protect sensitive species. our results do not " " address the question of how large such blocks should be, but do suggest that protecting blocks from harvesting is the most effective means for increasing forest interior and decreasing forest edge even when more intensive harvesting results on lands outside the protected area. indeed, various management strategies on different parts of the land base are required to meet the needs of species with disparate life histories and habitat requirements (freemark and merriam, ; hansen et al., ). harvest could be used to evaluate alternative management area boundaries and harvest strategies over long time periods to optimize the allocation and temporal rotation of multiple uses on public lands. • , o. e. j. gustafson and i'. r. crow _ '\ " years iii. ii _ .__ iiiiiiiiiiiiiiiiiiiiiiill °°°t. | .. ..... _ years _ years ::_,::__._,,/,_,:_v_ . . ¢:_:_:'_!ii_:..... -, . -_. . " years .._ _ . ., _:_ • : _ _. ,., e,l stand age classes (years) . , i " • figure. . changes in stand age. class distribution during simulation of years of implementation of the plan on the tell city unit of the hoosier national forest. the distributions on the other study areas _, were similar, and are not shown. _ ' " // /. the three study areas provided different initial conditions related to past man- _,_ / agement history. the study areas differed in stand age-class distribution and in the _: temporal trends in amounts of forest edge and forest interior before the baseline _ year. in spite of these differences, the resulting measures of forest spatial pattern produced over time by the simulations were similar for all three areas. small error terms were associated with estimates of total edge and total forest interior among the three replicates run for each study area. low variability in these parameters does not _ . suggest static conditions. the simulated landscapes were indeed dynamic, with shifting mosaics, as seen in figure . variance also serves as a source of error propagation that may be serious if the variance is large, and multiple time steps are run (gardner et al., ). the low variability .associated with our simulations produced robust results with relatively few (n= ) replicates. scale is also an important determinant of predictability (costanza and maxwell, : ). possible effects of resolution on predictability can be analyzed by comparing the simulations for the three study areas ( m resolution) with those conducted at m resolution for the entire hoosier national forest. for all three ,-. .. , simulating forest managementstrategies '\, " - • [ years / _- . i li'!i_:_ ___!_ _ ' ,. , , " " illlll in__ ., _ ; " ,i i i i_l i-i i i i i i i i i i i i i i i i i i i i i • , c_ . ,:_:::;:::_._,= _::'_':' "__""_'_ _ ..... i years ............. - _ :_': _ _ _ • . , o nmm,,m f. ............. innl/i ui i i,i ui i i i i i i i i i i i i i i i i i i i i i i i i ui i i i i i i i i i ui i i i ui i i i i stand age classes (years) " ... figure . changes in stand age class distribution during simulation of years of implementation of the plan on the tell city study area of the hnf. note the change in scaling of the y-axis for years .l_,• \ ._ . and . the distributions on the other study areas were similar and are not shown. ..... parameters (total forest interior, mean interior patch size and total linear edge), the _: /model indicated that rates of change with time were higher at the coarser resolution. differences in rates of change were due to differences in measurable edge because of larger cells at the m resolution, and the fact that smaller openings could not be • , detectedl the general trends, however, were similar between the two resolutions, . suggesting that these trends are scale invariant. . as with all models, simplifying assumptions were made in the simulations that _ , • affected the outcome. in this case, holding the treatments constant over a long period created extremely skewed age class structures (figures and ). although management always involves adjustments, the extreme cases presented here provide useful com- iparis. ns. another simplifying assumption, that of holding private lands in a constant state of canopy coverage throughout the simulation, probably reduced fragmentation .. at the landscape level. studies involving broader questions of context and interaction between adjacent public and private lands have been conducted or are underway (e.g. gustafson and crow, ). disturbances other than harvesting were not included in our-simulations to avoid confounding our analysis with the effects of management ,i _.,....l_t_.... • \ , , e. j. gustafson and t. r. crow • ",_ • . . alternatives. the stochastic allocation of stands is a simplistic representation of a complex decision-making process. based on our analysis of past allocations (table ), lit is reasonable to expect that the spatial patterns produced by our simulations provide reliable general principles about the effect of these management alternatives on forest spatial pattern through time. in the case study presented here, substituting uneven-age management in the form of group selection for even-age management and reducing harvest levels produces both economic as welt as ecological consequences. yet the substantial reductions in harvest -!___ levels do not necessarily result in substantial increases in forest interior or substantial decreases in forest edge. furthermore, oak-hickory forests may be difficult to maintain ._;i;!i'i__'_' '!_'_ii:_:i_ _ , without clearcutting (johnson, ). public land managers must also consider many .... - " other ecological concerns as well, such as water quality, disturbance-dependent com- : ' munities, backcountry recreational opportunities and invasion of exotics. it is the ;_! ....!_.i_' balancing of.various ecological and ec nomic values that is the challenge to public : _.- ¢ land managers. if the public limits management options in the hnf, this social ; ....... . . (and thus polhical) consideration may override the ecological and economic factors. _. "" ," harvest provides a tool to link policy (planning) with changes in landscape char- " acteristics (structure), commodity prodt_tion, and other potential benefits and values __ that are spatially dependent. such a tool may enable managers to better understand " management consequences and to defend their decisions to the public and other ?._ , professionals. " i_!_ _!_::i.... - we thank monica schwalbach, ellen jacquart, libby rice, mary stafford, tom thake and e:_,,?_i_;:, steve olson of the hoosier national forest for their assistance in providing data. thanks also to janet eger for producing a stand age map of the martin state forest and to kyle forbes for ':_ '_ _ digitizing stand maps. thanks to ellen jacquart, regis terney, cid morgan, louis iverson and lucy burde for their critical comments on earlier drafts of the manuscript, o .. .. references ., . a, andren,h. and angelstam,p. ( ).elevatedpredationrates asan edgeeffectin habitat islands:experimental evidence.ecology , - . baker,.w_l. ( ).effectsof settlementand firesuppressionon landscapestructure.ecology , - . ., / constanza, r. and maxwell,t. ( ).resolution and predictability:an approach to the scalingproblem. . / landscapeecology , - . __ " crow, t. (i ). landscapeecology:the big picture approach to resourcemanagement.in challengesin the _i conservationof biologicalresources,a practitioner'sguide(d. j. decker, m. e. krasny, g. r. goff, _, c. r.smith and d. w. gross, eds), pp. - . boulder, co: westviewpress. davis, j. c. ( ). statisticsand data analysisin geology,pp. - .new york: john wileyand sons. dellasalla,' d. a. and rabe, d. l. ( ). response of least flycatchersempidonax minimus to forest disturbances.biologicalconservation , - . " franklin, j. f. and forman, r. t. t. ( ). creating landscape patterns by forest cutting: ecological consequencesand principles.landscapeecology , - . .freemark,k. e. and merriam,h. g. ( ).importanceofarea and habitat heterogeneityto birdassemblages in temperateforest,fragments. biologicalconservation , - . gardner,r. h., dale, v.h. and o'neill,r. v.( ).error propagation and uncertaintyin processmodeling. in processmodelingof forest growthresponsesto environmentalstress (r. k. dixon, r. s. meldahl, g. a. ruai'k and w. g. warren, eds), pp. - . portland, oregon:timber press. gustafson, e. j.'and crow, t. r. ( ). modeling the effectsof forest harvestingon landscapestructure and the spatial distribution of cowbird brood parasitism. landscapeecology , - . hansen, a. j., garman, s. l., marks, b. and urban, d. l. ( ).an approach for managingvertebrate diyersityacross multiple-uselandscapes.ecologicalapplications , - . hemstrom, m. and cisse!,j. ( ).evaluating alternativelandscapepatterns over time--an examplefrom the willamettenational forest in oregon. in proceedingsof the society of americanforesters nationalconvention,safpublication - ,pp. - bethesda,md: societyof americanforesters. simulating forest management strategies johnson, e s, ( ). perspectives on the ecology and silviculture of oak-dominated forests in central and eastern states. usda forest service, general technical report nc- , pp. - . st. paul, mn: north central forest experimental station. v li, h., franklin, j. f., swanson, f. j. and spies, t. a. ( ). developing alternative forest cutting patterns: a simulation _approach. landscape ecology , - . mcgarigal, k. and marks, b. j. ( ). fragstats: spatial pattern analysis program for quantifying landscape structure. unpublished software. corvallis, or, u.s.a. department of forest science, oregon state university. mladen ff, d. j., white, m. a., pastor, j. and crow, t. r. ( ). comparing spatial pattern in unaltered ill old-growth and disturbed forest landscapes. ecological applications , - . ripple, w.j., bradshaw, g. a. and spies, t. a. ( ). measuring forest landscape patterns in the cascade range of oregon, usa. biological conservation , - . sotheim, s. l., alverson, w. s. and waller, d. m. ( ). maintaining biotic diversity in national forests: "_:._!__ -the necessity for large blocks of mature forests. endangered species: technical bulletin report , - . ii.... .-_, ........ , • ann ar. bor, mi: the school of natural resources, university of michigan. •......... thompson, f. r. iii. ( ). simulated responses of a forest-interior bird population to forest management ' ' options in central hardwood forests of the united states. conservation biology , - . _":_:._/._.li_)!_,ii_._,, usda forest service. ( a). final environmental impact statement, land and resource management plan, _'ii:_ _-i_:_.;l_;: hoosier national forest, pp. . - - . bedford, in: usda forest service, eastern region, hoosier " ,_::"":_:_ :. , . " national forest. : !_!_!,,,_:_:;__:_:.__:.... usda forest service. ( b). land and resource management plan, hoosier national forest, pp. - . :_:_!_!::)_::_, _, • ., bedford, in: usda forest service, eastern region, hoosier national forest. [ !_',,::i_.i_i_'.... "i_isda forest service. ( ). plan amendment, land and resource management plan, hoosier national forest, pp. - . bedford, in: usda forest service, eastern region, hoosier national forest .... [ __, " r . t, ° / / ' . ,,, i • . impact of the asco presentation of hog lun - /uso- on the prescribing plans of american medical oncologists for patients with stage iiib non-small cell lung cancer original article impact of the asco presentation of hog lun - / uso- on the prescribing plans of american medical oncologists for patients with stage iiib non-small cell lung cancer mark r. green, md,* howard west, md,† mark a. socinski, md,‡ joanne willey, rn, ocn,* laura daniels, ba,* kristine lemke, mba,* george rafferty, ba,* and lou iovino, ma* introduction: nonoperative treatment of stage iii non-small cell lung cancer has evolved over the past years. the current approach in the unites states most often includes concurrent chemoradiotherapy. methods: we have used live, case-based research events to docu- ment prescribing plans among american medical oncologists for first-line therapy in patients with n stage iiib non-small cell lung cancer. changes in prescribing plans documented before and after the american society of clinical oncology (asco) presen- tation of a hoosier oncology group trial testing the role of consol- idation docetaxel chemotherapy in this setting are presented. results: data from show a post-asco shift away from plans for docetaxel consolidation, increased use of concurrent chemora- diotherapy alone, and stable to increased plans for concurrent chemoradiation followed by additional cycles of the chemotherapy used during concurrent management ( %). preliminary data from confirm the durability of these changes. conclusions: the findings of the hoosier oncology group trial support a transition away from docetaxel consolidation. a trend in this direction among american medical oncologists is clear from our data. however, nearly % of oncologists studied in still plan to use docetaxel consolidation. furthermore, a majority of those studied after asco continue to report plans to use more than two cycles of chemotherapy as part of their preferred treatment recommendation despite no level i evidence to support this approach. key words: non-small cell lung cancer, stage iii, chemoradio- therapy, consolidation chemotherapy, phase iii trial. (j thorac oncol. ; : – ) the most recent figures from the american cancer societysuggest that we should expect , new cases of lung cancer in . approximately % of these new cases will have nonsmall cell histology and about one third will have regional (n –n ) disease at the time of diagnosis. based on these figures, nearly , individual americans should be identified with stage iii non-small cell lung cancer (nsclc) during this calendar year. nonoperative treatment of stage iii nsclc has evolved during the past years from largely palliative radiation therapy, through sequential chemotherapy followed by radiation, to the current use of concurrent chemoradiother- apy. in many cases, physicians have used two to four additional chemotherapy cycles as either induction or consol- idation therapy with the expectation that the added chemo- therapy would improve survival. in , data from a cancer and leukemia group b randomized phase iii trial failed to demonstrate a survival advantage for the addition of two cycles of full-dose carboplatin-paclitaxel followed by defin- itive radiation and concurrent weekly paclitaxel-carboplatin compared with definitive radiation and concurrent weekly paclitaxel-carboplatin alone. however, up until recently, phase ii data reported by multiple investigators suggested that immediate concurrent chemoradiotherapy followed by con- solidation chemotherapy produced excellent outcomes for nonoperative therapy in stage iii patients. one such approach developed by southwest oncology group (swog) investi- gators involved immediate concurrent etoposide-cisplatin and radiotherapy followed by three cycles of docetaxel. another approach using immediate concurrent carboplatin-paclitaxel radiation followed by two additional cycles of every -week paclitaxel-carboplatin was reported by belani et al. data on american medical oncologists studied by investigators at network for medical communications and research (nmcr) analytics between and before the amer- ican society of clinical oncology (asco) annual meeting *nmcr analytics, atlanta, georgia; †swedish cancer institute, seattle, washington; and ‡lineberger cancer center, university of north caro- lina, chapel hill, north carolina. mark r. green, laura daniels, kristine lemke, george rafferty, and lou iovino are employees of nmcr analytics, a market research data company working in the pharmaceutical space. this work was done without any requests or special funding from any pharmaceutical com- pany or related third party. all the aforementioned authors are full time employees of nmcr analytics and none has received any personal financial consideration for any aspect of this work. none has a conflict of interest to report and howard west has no conflicts to report concerning this work. address for correspondence: mark r. green, md, nmcr analytics, johnson ferry road, suite , atlanta, ga . e-mail: mgreen@ nmcr.com presented at the american society of clinical oncology annual meeting, abstract no . copyright © by the international association for the study of lung cancer issn: - / / - journal of thoracic oncology • volume , number , august showed that nearly % planned to use both concurrent che- moradiation and consolidation chemotherapy in this setting. at the asco meeting, investigators from the hoosier oncology group (hog) reported that there was no survival advantage but significant toxicity from the addition of docetaxel consolidation to immediate radiation and con- current etoposide-cisplatin (figure ). to assess the impact of these data, we documented the prescribing plans for nonop- erative therapy for patients with n stage iiib nsclc among several hundred additional american medical oncolo- gists. here, we report those prescribing plans and discuss the impact of new phase iii data on treatment strategies planned for patients with stage iiib nsclc. methods since , nmcr analytics has studied prescribing plans of american medical oncologists for patients with stage iii nsclc using individual live research events. physi- cians participating in these research programs self-register through direct invitation or by means of the nmcr analytics website. attendees are reimbursed for research-related ex- penses including any travel costs incurred and specific out- of-pocket expenses. attendees do not receive additional pay- ments for their participation. responses to research queries are treated confidentially, and no identifying information about individual participants is disclosed. attendees may participate in only one lung cancer research event within a calendar year. some of the prescribing plans recorded in the first two research events of may come from individuals also included in the prescribing plans tallied pre-asco or post-asco . research programs are designed to mimic real-life clinical scenarios. the hypothetical case history is projected on the screen at the front of the meeting room and read aloud by the research event moderator. treatment plan options (up to ) are then offered, and physician selections are acquired anonymously and contemporaneously by keypad and elec- tronic data capture before any discussion of relevant data supporting one or more of the possible prescribing plans offered. as soon as the participants have completed their selection (documented by real-time counter of responses), the results are tallied and projected in bar graph format. demographic information is collected from the partic- ipants at the beginning of each research session. the charac- teristics polled include practice venue (academic versus com- munity and geographic region), number of partners, years from completion of training in oncology, number of new lung cancer patients seen each month, sex, and participation in a group purchasing organization. these features are entered into the wireless keypad by each attendee in response to specific questions. this information then gives each keypad a demographic character that can be used to analyze the re- spondent choices by subset while maintaining the anonymity of the individual participants. to study current approaches in a patient with n stage iiib nsclc, we used a core case history and a “menu” of preset treatment plans (figures , ). oncologists are very familiar with this approach. case-based decision making is the standard format for tumor board or clinical case confer- ence discussions of potential management options for indi- vidual patients. although this format cannot encompass all possible treatment plans among the options offered (� ), we include a broad range of relevant options including all relevant evidence-based strategies. an option of “other” is also included to avoid a forced choice. not all attendees respond to every research question posed, but overall participation of more than % is achieved for each question. figure . schema for hoosier oncology group (hog) trial evaluating the role of docetaxel consolidation in pa- tients with a new diagnosis of stage iii non-small cell lung cancer (nsclc). a -year-old former smoker has a progressive cough. after an episode of hemoptysis (actually only a small amount of bloody sputum) he is seen in the er where a chest x-ray and chest ct are done. there is a partially cavitated, medially located, rul mass with enlarged hilar and n and n mediastinal nodes. a pulmonary medicine physician sees him and arranges an urgent bronchoscopy (tbb specimen positive for nsclc nos) and wang needle aspiration of subcarinal nodes (also + for tumor). a pet scan is positive in these areas as well as the right hilum and the right and left paratracheal areas. a head mri is negative for metastatic disease. figure . case scenario for assessing prescribing plans of american medical oncologists. no changes were made in the case scenario during the entire period reported. . immediate trt followed by (f/by) ct . induction ct f/bytrt . cycles of induction ct f/by ct/trt (same drugs as in induction) . ct/trt (no induction or consolidation ct) . ct/trt f/by added cycles of the same ct used during radiation . ct/trt f/by doses of docetaxel . ct/trt f/by doses of docetaxel and then single agent erlotinib. . ct alone . other trt = thoracic radiotherapy; ct = chemotherapy; ct/trt = concurrent chemoradiotherapy figure . prescribing plans offered as selection options for the research study participants. the option “other” is rou- tinely included to avoid driving research participants to a “forced choice” selection. green et al. journal of thoracic oncology • volume , number , august copyright © by the international association for the study of lung cancer results between march and may , medical oncologists participated in live research events studying prescribing plans for different clinical settings in lung cancer. these individuals and those queried after the asco meeting were � % self-identified as “community based” and displayed a geographic spread reflective of the broad american oncology workforce. less than % were in solo practice, and a majority of all respondents were more than years out from completion of their oncology training. in the assessment of prescribing plans for the stage iiib setting, a substantial majority of the responding physi- cians planned to use concurrent chemoradiotherapy followed by additional chemotherapy (figure ). for % of physi- cians, the planned therapy was immediate chemoradiation followed by three doses of docetaxel consolidation. for %, the plan included the same chemoradiation followed by three doses of docetaxel consolidation and then additional mainte- nance therapy with an oral antiepidermal growth factor re- ceptor tyrosine kinase inhibitor. an additional % of re- spondents favored the use of immediate chemoradiotherapy followed by two additional cycles of the same chemotherapy used during the concurrent phase of treatment. in , we studied an additional physicians (pre- set participation goal � ) in five live research events held between march and september. demographic characteristics of these research participants were similar to those of the earlier physician cohort. these research events again used the same clinical scenario of stage iiib disease and offered the same panel of therapy options. in figure , the aggregate findings of the five research events of are shown, whereas in figure , the data are broken down by pre-asco and post-asco . the aggregate data do show a shift away from the use of immediate chemoradio- therapy followed by docetaxel consolidation ( %), a rise in the use of immediate concurrent chemoradiotherapy alone ( %), and stable plans for use of immediate concurrent chemoradiation followed by two additional cycles of the same chemotherapy that was used in the concurrent phase of therapy ( %). however, the aggregate data (figure ) ob- scure a more dramatic time-dependent movement away from docetaxel consolidation ( % %; � p � . ) and toward immediate concurrent chemoradiotherapy without consolidation, the strategies compared directly in the hog phase iii trial. in figure , the data from each of the five research events are displayed individually. presenting the data in this fashion further highlights the change driven by the asco presentation of the hog trial data. immediately after asco , a shift in prescribing plans occurred. its dura- bility and further amplification are evident in the data from late september. after the acquisition of these data, we con- tinued to monitor prescribing plans in (table ). in two research events in february and april , plans for use of figure . prescribing plans for a patient with a new diag- nosis of stage iiib non-small cell lung cancer (nsclc) as- sessed in american medical oncologists who partici- pated in network for medical communications and research (nmcr) live research events between march and may . figure . prescribing plans for a patient with a new diag- nosis of stage iiib non-small cell lung cancer (nsclc) as- sessed in american medical oncologists who partici- pated in network for medical communications and research (nmcr) live research events between march and september . figure . prescribing plans for a patient with a new diag- nosis of stage iiib non-small cell lung cancer (nsclc) as- sessed in american medical oncologists who partici- pated in network for medical communications and research (nmcr) live research events between march and september . prescribing plans shown for two time frames, before and after american society of clinical oncology (asco) . choices of “radiation followed by chemotherapy,” “chemotherapy alone,” and “other,” totaling % are not shown here. journal of thoracic oncology • volume , number , august impact of the asco presentation of hog lun - /uso- copyright © by the international association for the study of lung cancer consolidation docetaxel have fallen to less than % of research respondents. the largest single group now plans to use immediate concurrent chemoradiotherapy without either induction or consolidation. however, the ambiva- lence of the treating oncology community is clear, with a majority of all respondents planning some treatment strat- egy that includes more than just two cycles (or equivalent) of chemotherapy during radiation as total therapy in the stage iiib setting. discussion individuals diagnosed with a contralateral-mediastinal, n , stage iiib nsclc are seeking the best therapy option to provide them with the longest duration of quality life. they recognize that there may be some individual potential to achieve long-term disease control, although the best strategy for optimizing this goal is unclear. for fit individuals with a diagnosis of n , stage iiib nsclc, the treatment recom- mendations of american medical oncologists almost always involve both chemotherapy and radiation. these two modal- ities are designed to both maximize control of the macro- scopic locoregional disease and eradicate micrometastatic foci that are also highly likely to be present. currently, phase iii trial data from numerous sources suggest that use of concurrent chemoradiotherapy can increase overall survival when compared directly with the same therapies given se- quentially. the survival advantage is modest however, and there are additional toxicities associated with concurrent rather than sequential management. the potential of chemotherapy to effectively eradicate micrometastases and improve survival when used in the adjuvant setting has been demonstrated among patients with many different types of solid tumors including nsclc. in almost all these settings (testis cancer being a notable solid tumor exception), the adjuvant chemotherapy prescription has included more than two cycles of chemotherapy. thus, when approaching an individual with more disease (e.g., stage iiib nsclc) and the expectation that the burden of micrometastatic disease is likely to be at least as great if not greater than in the traditional adjuvant setting, investigators and treating physicians have tended to favor strategies using more than two cycles of chemotherapy as part of definitive management. yet, no randomized phase iii data are avail- able to endorse this approach in stage iii nsclc. however, as demonstrated here, before asco the use of more than two cycles of chemotherapy was the overwhelming plan of american medical oncologists treating patients with n , stage iiib disease. for nearly %, the additional chemother- apy was planned as consolidation and for most of those the consolidation involved use of single-agent docetaxel. this approach was supported by impressive multiinstitutional phase ii data provided by investigators from the swog. in their phase ii work (swog ), the median survival achieved with immediate concurrent etoposide-cisplatin and thoracic radiation followed by three cycles of docetaxel consolidation was months and the - and -year survival rates were % and %, respectively. in a previous swog study in patients with stage iii disease using the same initial chemoradiotherapy but followed by consolidation with two added cycles of etoposide and cisplatin, the median survival was months and - and -year survival rates were %. these phase ii findings and informal comparison generated major enthusiasm for the swog approach. after lengthy discussions with representatives from the national cancer institute’s cancer therapy evaluation program, the follow-up phase iii trial proposed by swog investigators to confirm the benefit of consolidation docetaxel that was im- plied by their sequential phase ii trials was not approved. instead, an alternative design that presumed a benefit from consolidation docetaxel and tested the value of maintenance gefitinib (swog s ) was moved forward. at asco , findings from the hog phase iii (albeit modest in size) trial testing immediate concurrent figure . prescribing plans for a patient with a new diag- nosis of stage iiib non-small cell lung cancer (nsclc) as- sessed in american medical oncologists who partici- pated in network for medical communications and research (nmcr) live research events between march and september . prescribing plan data are shown for each specific re- search event with the inflection point coinciding with the hoosier oncology group (hog) data presentation at ameri- can society of clinical oncology (asco) . table . prescribing plan data from research events held in february and april demonstrate the persisting shift away from use of docetaxel consolidation after asco regimen february , , n � (%) april , , n � (%) ct/trt ct-ct/trt ct/trt (hog) ct/trt-same ct (lamp) ct/trt-doc (swog ) ct/trt-doc-erlotinib ( ) ct alone other ct/trt, chemotherapy/thoracic radiotherapy; hog, hoosier oncology group; lamp, locally advanced multi-modality protocol; swog, southwest oncology group. green et al. journal of thoracic oncology • volume , number , august copyright © by the international association for the study of lung cancer chemoradiotherapy as used by swog with or without do- cetaxel in patients with stage iii nsclc were presented by hanna et al. although some of the patients included in this study had baseline characteristics, especially less optimal pulmonary function, that were different from the swog study participants, the core comparison of ct/trt with or without consolidation addressed the critical question of whether the addition of consolidation chemotherapy would significantly improve the survival outcome in patients with stage iii nsclc. the data from the presentation were clear. in every relevant treatment-efficacy comparison there was no improvement, let alone no significant benefit, for the addition of docetaxel consolidation. at asco , further follow-up of the hog trial again showed no benefit from docetaxel consolidation. however, the presentation did highlight the toxicity risks associated with docetaxel consolidation. these included a . % rate of grade or neutropenia, a . % rate of febrile neutropenia, an . % rate of grade pneumonitis, and a . % treatment related death rate. the findings of the hog trial support a transition away from docetaxel consolidation, and the trend in this direction among american medical oncologists is clear from our data. however, two surprising results of our research are evident. first and foremost, nearly % of those oncologists studied in are still planning a swog approach with docetaxel consolidation. this seems difficult to justify in light of the toxicity produced by this strategy and the failure of the phase iii comparison to suggest any benefit. whether the swog lung commit- tee will decide to retest a chemoradiotherapy followed by consolidation strategy remains to be seen. more globally, a majority of all the oncologists studied after asco continue to report plans to use more than two cycles of chemotherapy as part of their preferred treatment recom- mendation for fit patients with stage iii nsclc. for the largest single subgroup of those oncologists, the preferred recommendation now involves immediate induction che- moradiotherapy followed by two added cycles of the same chemotherapy in the consolidation setting. although this decision is consistent with the model of adjuvant chemo- therapy widely applied in patients with completely re- sected solid tumors, there are no phase iii data in the lung cancer setting to support it, and the toxicity risks and economic costs associated with it are formidable. cur- rently, there are no plans among the american national cancer institute sponsored cooperative groups to retest the role of chemotherapy consolidation, e.g., immediate con- current chemoradiation with or without two added cycles of the same chemotherapy as consolidation. although such a trial would add to the evidence base in stage iii disease, there are scant to no compelling phase ii data that support such a resource-intensive undertaking. the changing approach to definitive nonoperative management of patients with stage iii nsclc is clearly demonstrated by our research. we will continue to monitor prescribing plans proposed by american medical oncolo- gists for patients with regionally advanced nsclc. it is hoped that the trends away from docetaxel consolidation will continue. this is a critical setting in which patients with a new diagnosis of stage iii nsclc and their treating oncologists should come together to participate in well- designed clinical trials with the potential to generate new leads or definitive data concerning better approaches to achieve the jointly shared goal of a longer duration of quality life and an enhanced opportunity for long-term disease control. acknowledgments the authors acknowledge the kind statistical advice of dr. stephen george, professor of biostatistics, duke university. references . jemal a, siegel r, ward e, et al. cancer statistics . ca cancer j clin ; : – . . farray d, mirkovic n, albain ks. multimodality therapy for stage iii non-small cell lung cancer. j clin oncol ; : – . . vokes ee, herndon je ii, kelley mj, et al.; cancer and leukemia group b. induction chemotherapy followed by chemoradiotherapy com- pared with chemoradiotherapy alone for regionally advanced unresect- able stage iii non-small-cell lung cancer: cancer and leukemia group b. j clin oncol ; : – . . gandara dr, chansky k, albain ks, et al. long-term survival with concurrent chemoradiation therapy followed by consolidation docetaxel in stage iiib non-small-cell lung cancer: a phase ii southwest oncology group study (s ). clin lung cancer ; : – . . belani cp, choy h, bonomi p, et al. combined chemoradiotherapy regimens of paclitaxel and carboplatin for locally advanced non-small- cell lung cancer: a randomized phase ii locally advanced multi-modality protocol. j clin oncol ; : – . . green mr, west h, socinski ma, et al. management of n stage iiib nsclc: changes in us medical oncologists prescribing plans following asco . j clin oncol ; (suppl ): . . hanna nh, neubauer m, ansari r, et al. phase iii trial of cisplatin (p) plus etoposide (e) plus concurrent chest radiation (xrt) with or without consolidation docetaxel (d) in patients (pts) with inoperable stage iii non-small cell lung cancer (nsclc): hog lun - /uso- . j clin oncol ; (suppl s): . . pignon jp, tribodet h, scagliotti gv, et al.; lace collaborative group. lung adjuvant cisplatin evaluation: a pooled analysis by the lace collaborative group. j clin oncol ; : – . . albain ks, crowley jj, turrisi at iii, et al. concurrent cisplatin, etoposide, and chest radiotherapy in pathologic stage iiib non-small-cell lung cancer: a southwest oncology group phase ii study, swog . j clin oncol ; : – . . kelly k, chansky k, gaspar le, et al. phase iii trial of maintenance gefitinib or placebo after concurrent chemoradiotherapy and docetaxel consolidation in inoperable stage iii non-small cell lung cancer: swog s . j clin oncol : – . . mina la, neubauer ma, ansari rh, et al. phase iii trial of cisplatin (p) plus etoposide (e) plus concurrent chest radiation (xrt) with or without consolidation docetaxel (d) in patients (pts) with inoperable stage iii non-small cell lung cancer (nsclc): hog lun - /uso- — updated results. j clin oncol ; (suppl s): . . hanna n, neubauer m, yiannoutsos c, et al.; hoosier oncology group; us oncology. phase iii study of cisplatin, etoposide, and concurrent chest radiation with or without consolidation docetaxel in patients with inoperable stage iii non-small-cell lung cancer: the hoosier oncology group and u.s. oncology. j clin oncol ; : – . journal of thoracic oncology • volume , number , august impact of the asco presentation of hog lun - /uso- copyright © by the international association for the study of lung cancer 	impact of the asco presentation of hog lun - /uso- on the prescribing plans of american medical oncologists for patients with stage iiib non-small cell lung cancer 	methods 	results 	discussion 	acknowledgments 	references i n d i a n a m a g a z i n e o f h i s t o r y the sons of westwood: john wooden, ucla, and the dynasty that changed college basketball by john matthew smith (urbana: university of illinois press, . pp. . illustrations, notes, index. clothbound, $ . ; paperbound, $ . .) for good reason, historians revel in examining periods rife with clashes of cultural values, clashes that can stem from differences in generations, regions, religion, politics, and race. in the case of john wooden and the basketball dynasty he built at ucla from to , all of those fac- tors resulted in on- and off-the-court disagreements between coach and athletes. significantly, the disagree- ments never hindered the team’s success, and almost forty years since he coached his last game, wooden remains a popular figure. john matthew smith cautions that “in the popular culture, adulators have constructed a sanitized version of ucla’s past,” a narrative that excludes any reference to blemishes—includ- ing wooden’s reluctance to banish booster-bandit sam gilbert from the ucla program—in an effort to my- thologize the coach as the “wizard of westwood.” smith has, instead, pro- duced a thoroughly researched history that examines not only wooden, his depression-era values, and his often fabricated image, but the reactions of his players and some of his critics. smith quotes journalist herbert war- ren wind: “he is an anachronism, john wooden—an island of james whitcomb riley in a sea of ken kesey, the grateful dead, terry southern, and jerry rubin” (p. ). to that stormy sea of american culture in the late-sixties and mid- seventies, smith adds the stories of some of wooden’s most accomplished players, notably lew alcindor (now kareem abdul-jabbar), bill walton, and wooden’s first great african american player walt hazzard, a young man who left the ghetto of west philadelphia for what he believed was the racial paradise of ucla and southern california. while alcindor’s difficulties with his coach came from his own growing racial awareness and walton’s came from his opposition to the vietnam war, hazzard’s objections were more fundamental to the style of basketball wooden favored: “the hoosier coach held steadfast to… a philosophy founded on simplic- ity and structure. the philadelphian brought an urban east coast ethos to the west coast, a style shaped by his experiences competing in playground pickup games” (p. ). smith’s account of wooden and his tenure at ucla offers no information that would strike sport and cultural historians as especially new, but the author has succeeded in interweaving wooden’s story with those of his play- ers, and that narrative style reveals some of the thorniest and most de- structive issues of a turbulent period in american culture. that smith ably illustrates these issues within the con- reviews text of the most successful program in american intercollegiate sports is a noteworthy accomplishment. any review of the sons of west- wood would be incomplete if it failed to praise smith’s writing style. the book is interesting and readable largely because of the writer’s voice. smith begins his first chapter by adopting wooden’s point of view as he struggled to negotiate the alien los angeles freeway traffic in , the year he left indiana for ucla: “john wooden wanted to turn around, but it was too late. indiana was long gone in his rearview mirror” (p. ). two short sentences that effectively say it all. the rest is historical exegesis and a great read. dennis gildea is professor of com- munications at springfield college. detroit’s cold war: the origins of postwar conservatism by colleen doody (champaign: university of illinois press, . pp. . illustrations, notes, works cited, index. $ . .) more than twenty years after the collapse of the soviet union, few people beyond historians and other scholars remember the intensity of the cold war. however, its influence is still felt today whenever political conservatives rail against liberals and denounce civil rights initiatives and unions as un-american, anti-freedom, and socialistic. colleen doody’s detroit’s cold war: the origins of postwar conservatism is a brief but well-researched book that explores the origins of modern conservatism in the united states, and largely debunks the long-held notion that americans emerged from the new deal and world war ii with a liberal consensus regarding the political economy and society at-large. doody uses detroit, the industrial powerhouse of the nation for most of the mid-to late twentieth century, as a case study for her inquiry. union membership peaked during the s and s, and the american midwest was clearly the world’s industrial heart- land. many industrial cities in indiana and ohio were immersed in the same political forces as detroit during this era, but the latter city was the capital of the american labor movement and the fourth largest urban center in the u.s., with a large, growing african american population and an active civil rights movement. during these years, the author argues, the key elements of the modern conservative ethos coalesced: antipathy toward centralized “big government”; an embrace of religious conservatism, especially among the city’s large catholic population; a cel- ebration of laissez-faire capitalism; and militant anti-communism. ongy .qxd © s. karger ag, basel fax + access to full text and tables of contents, e-mail karger@karger.ch including tentative ones for forthcoming issues: www.karger.com www.karger.com/ocl_issues contents vol. , no. review how to interpret serum ca levels in patients with serosal involvement? a clinical dilemma sevinc, a.; camci, c.; turk, h.m.; buyukberber, s. (gaziantep) clinical studies alveolar soft part sarcoma in japan: multi-institutional study of patients from the japanese musculoskeletal oncology group ogose, a. (niigata); yazawa, y. (utsunomiya); ueda, t. (osaka); hotta, t.; kawashima, h.; hatano, h.; morita, t. (niigata) impact of overall treatment time on local control of anal cancer treated with radiochemotherapy graf, r.; wust, p.; hildebrandt, b.; gögler, h.; ullrich, r. (berlin); herrmann, r. (basel); riess, h.; felix, r. (berlin) pain after quadrantectomy and radiotherapy for early-stage breast cancer: incidence, characteristics and influence on quality of life. results from a retrospective study amichetti, m.; caffo, o. (trento) long-term prognosis of stage i ovarian carcinoma. prognostic importance of intraoperative rupture mizuno, m.; kikkawa, f.; shibata, k.; kajiyama, h.; suzuki, t.; ino, k. (nagoya); kawai, m. (toyohashi); mizutani, s. (nagoya) laboratory/clinical translational research increased expression of rela/nuclear factor-��b protein correlates with colorectal tumorigenesis yu, h.-g. (wuhan/bochum); yu, l.-l.; yang, y.; luo, h.-s.; yu, j.-p. (wuhan); meier, j.j.; schrader, h.; bastian, a.; schmidt, w.e.; schmitz, f. (bochum) a novel case of a cat to aat transversion in codon of the p gene in a supratentorial primitive neuroectodermal tumor harbored by a young girl. case report and review of the literature postovsky, s.; weyl ben arush, m.; elhasid, r.; davidson, s.; leshanski, l.; vlodavsky, e.; guilburd, j.n. (haifa); amikam, d. (haifa/tel-hai) prognostic role of serum sialyl lewisx (cd s) in colorectal cancer paganuzzi, m.; bobbio, b.; marroni, p.; filiberti, r.; secco, g.b.; grossi, c.e. (genoa) molecular biologic characteristics of seven new cell lines of squamous cell carcinomas of the head and neck and comparison to fresh tumor tissue welkoborsky, h.-j. (hannover); jacob, r.; riazimand, s.h.; bernauer, h.s.; mann, w.j. (mainz) expression of c-met tyrosine kinase receptor is biologically and prognostically relevant for primary cutaneous malignant melanomas cruz, j. (porto); reis-filho, j.s. (porto/braga); silva, p.; lopes, j.m. (porto) fas ligand expression is correlated with metastasis in colorectal carcinoma nozoe, t.; yasuda, m.; honda, m.; inutsuka, s.; korenaga, d. (fukuoka) abnormal fragile histidine triad expression in advanced cervical cancer and evaluation of its utility as a prognostic factor machida, s.; ohwada, m.; saga, y.; suzuki, m. (tochigi) book review no. clinical studies afp-producing gastric carcinoma: multivariate analysis of prognostic factors in patients adachi, y. (oita); tsuchihashi, j. (fukuoka); shiraishi, n.; yasuda, k.; etoh, t.; kitano, s. (oita) phase i study of combination chemotherapy using irinotecan hydrochloride and nedaplatin for advanced or recurrent cervical cancer machida, s.; ohwada, m. (tochigi); fujiwara, h.; konno, r.; takano, m.; kita, t.; kikuchi, y.; komiyama, s.; mikami, m. (saitama); suzuki, m. (tochigi) high incidence of angina pectoris in patients treated with -fluorouracil. a planned surveillance study with patients wacker, a.; lersch, c.; scherpinski, u.; reindl, l.; seyfarth, m. (münchen) features of second primary cancer in patients with gastric cancer ikeda, y.; saku, m.; kawanaka, h.; nonaka, m.; yoshida, k. (fukuoka) central nervous system involvement in children with sarcoma postovsky, s. (haifa); ash, s. (petach tikva); ramu, i.n. (jerusalem); yaniv, y.; zaizov, r. (petach tikva); futerman, b. (haifa); elhasid, r.; benbarak, a.; halil, a.; ben arush, m.w. (haifa) laboratory/clinical translational research quantitative analysis of thymidine phosphorylase and dihydropyrimidine dehydrogenase in renal cell carcinoma morita, t.; matsuzaki, a.; tokue, a. (tochigi) association between serum carcinoembryonic antigen and endothelial cell adhesion molecules in colorectal cancer roselli, m.; guadagni, f.; martini, f.; spila, a.; mariotti, s.; d’alessandro, r.; aloe, s.; gazzaniga, p.p.; basili, s.; cosimelli, m.; ferroni, p. (rome) clinical relevance of sialyltransferases st gal-i and st gal-iii in gastric cancer gretschel, s.; haensch, w.; schlag, p.m.; kemmner, w. (berlin) colorectal cancer relapse: allelic alterations associated with tumour marker overexpression vegh, i.; de-la-cruz, j.; navarro, s.; morales, c.; colina, f.; abad, a.; de-la-calle, a.; enriquez-de-salamanca, r.; moreno-gonzález, e. (madrid) expression of phosphatidylethanolamine n-methyltransferase in human hepatocellular carcinomas tessitore, l. (novara); marengo, b. (genoa); vance, d.e. (edmonton, alta.); papotti, m.; mussa, a. (turin); daidone, m.g.; costa, a. (milan) alteration of g cell cycle regulators occurs during carcinogenesis of the endometrium tsuda, h.; hashiguchi, y.; inoue, t.; yamamoto, k. (osaka) contents oncology vol. , iii laboratory investigations interleukin- �� enhances integrin �� �� expression and metastatic capability of human pancreatic cancer sawai, h.; funahashi, h.; yamamoto, m.; okada, y.; hayakawa, t.; tanaka, m.; takeyama, h.; manabe, t. (nagoya) adhesion and migration of hrt- colorectal carcinoma cells on extracellular matrix components typical for the desmoplastic stroma of colorectal adenocarcinomas hauptmann, s. (berlin); budianto, d. (aachen); denkert, c.; köbel, m. (berlin); borsi, l.; siri, a. (genova) case report successful treatment of human immunodeficiency virus-related castleman’s disease: a case report and literature review liberopoulos, e.; tolis, c.; bai, m.; efremidis, s.; pavlidis, n.; elisaf, m. (ioannina) erratum no. invited review gastrointestinal stromal tumors: biology and treatment duffaud, f. (marseille); blay, j.-y. (lyon) clinical studies elderly patients with advanced non-small cell lung cancer. a phase ii study with weekly cisplatin and gemcitabine berardi, r.; porfiri, e.; scartozzi, m.; lippe, p. (ancona); silva, r.r. (fabriano); nacciarriti, d.; menichetti, e.t. (senigallia); tummarello, d.; carle, f.; piga, a.; cellerino, r. (ancona) recurrence rate and prognosis of patients with hepatocellular carcinoma that developed after elimination of hepatitis c virus rna by interferon therapy. a closed cohort study including matched control patients ikeda, k.; kobayashi, m.; saitoh, s.; someya, t.; hosaka, t.; akuta, n.; suzuki, f.; tsubota, a.; suzuki, y.; arase, y.; kumada, h. (tokyo) effective combination chemotherapy with bimonthly docetaxel and cisplatin with or without hematopoietic growth factor support in patients with advanced gastroesophageal cancer schüll, b.; kornek, g.v.; schmid, k.; raderer, m.; hejna, m.; lenauer, a.; depisch, d. (vienna); lang, f. (neunkirchen); scheithauer, w. (vienna) phase ii study of gemcitabine plus docetaxel in advanced pancreatic cancer: a hoosier oncology group study schneider, b.p.; ganjoo, k.n.; seitz, d.e. (indianapolis, ind.); picus, j. (st. louis, mo.); fata, f. (goshen, ind.); stoner, c.; calley, c.; loehrer, p.j. (indianapolis, ind.) phase ii trial of weekly paclitaxel in previously untreated advanced non-small-cell lung cancer yasuda, k.; igishi, t.; kawasaki, y.; yamamoto, m.; kato, k.; matsumoto, s.; kotani, m.; sako, t.; shigeoka, y.; sugitani, a.; histuda, y.; shimizu, e. (yonago) changes in the activation and reconstitution of lymphocytes resulting from total-body irradiation correlate with slowed tumor growth miller, g.m.; kim, d.w.; andres, m.l.; green, l.m.; gridley, d.s. (loma linda, calif.) low-dose thalidomide treatment for advanced hepatocellular carcinoma hsu, c.; chen, c.-n. (taipei); chen, l.-t. (taipei/kaohsiung); wu, c.-y.; yang, p.-m.; lai, m.-y.; lee, p.-h.; cheng, a.-l. (taipei) prevention of intrahepatic metastasis by curcumin in an orthotopic implantation model ohashi, y.; tsuchiya, y.; koizumi, k.; sakurai, h.; saiki, i. (toyama) anaplastic oligodendrogliomas: prognostic factors for tumor recurrence and survival puduvalli, v.k.; hashmi, m. (houston, tex.); mcallister, l.d. (portland, oreg.); levin, v.a.; hess, k.r. (houston, tex.); prados, m. (san francisco, calif.); jaeckle, k.a.; yung, w.k.a. (houston, tex.); buys, s.s. (salt lake city, utah); bruner, j.m. (houston, tex.); townsend, j.j. (salt lake city, utah); davis, r. (san francisco, calif.); sawaya, r.; kyritsis, a.p. (houston, tex.) laboratory/clinical translational research forced expression of cytidine deaminase confers sensitivity to capecitabine morita, t.; matsuzaki, a.; kurokawa, s.; tokue, a. (tochigi) helicobactor pylori infection is closely associated with telomere reduction in gastric mucosa kuniyasu, h. (kashihara); kitadai, y.; mieno, h.; yasui, w. (hiroshima) letter to the editor comparison of low-dose versus standard-dose colony stimulating factor in patients with chemotherapy-induced febrile neutropenia burney, i.a.; salam, a.; jat, a.a.; iqbal, m.p.; siddiqui, t.; khan, m.a.; smego, r.a., jr. (karachi) no. clinical studies multiple signal pathways are involved in the mitogenic effect of (s)-hete in human pancreatic cancer ding, x.-z.; tong, w.-g.; adrian, t.e. (chicago, ill.) usefulness of mtc-tetrofosmin scintigraphy in different variants of kaposi’s sarcoma spanu, a.; madeddu, gr.; cottoni, f.; manca, a.; migaleddu, v.; chessa, f.; masala, m.v.; cossu, a.; falchi, a.; mura, m.s.; madeddu, g. (sassari) rituximab for treatment of advanced extranodal marginal zone b cell lymphoma of the mucosa-associated lymphoid tissue lymphoma raderer, m. (vienna); jäger, g. (graz); brugger, s. (st. pölten); püspök, a.; fiebiger, w.; drach, j. (vienna); wotherspoon, a. (london); chott, a. (vienna) role of mammography, ultrasound and large core biopsy in the diagnostic evaluation of papillary breast lesions puglisi, f.; zuiani, c.; bazzocchi, m.; valent, f.; aprile, g.; pertoldi, b.; minisini, a.m.; cedolini, c.; londero, v.; piga, a.; di loreto, c. (udine) impact of interval debulking surgery of clinical outcome in primary unresectable figo stage iiic ovarian cancer patients fanfani, f.; ferrandina, g.; corrado, g.; fagotti, a.; zakut, h.v.; mancuso, s.; scambia, g. (rome) laboratory/clinical translational research metal contents in the liver of patients with chronic liver disease caused by hepatitis c virus. reference to hepatocellular carcinoma ebara, m.; fukuda, h.; hatano, r.; yoshikawa, m.; sugiura, n.; saisho, h.; kondo, f.; yukawa, m. (chiba) changes in apoptosis and mitotic index, p and ki expression in various types of oral leukoplakia kövesi, g.; szende, b. (budapest) low collagenase- (mmp- ) and mt -mmp expression levels are favourable survival markers in advanced colorectal carcinoma bendardaf, r.; lamlum, h.; vihinen, p.; ristamäki, r.; laine, j.; pyrhönen, s. (turku) the supplementary diagnostic power of selected immunohistochemical, molecular genetic and infective parameters in epithelial hyperplastic laryngeal lesions kleist, b.; junghans, d.; lorenz, g.; bankau, a.; poetsch, m. (greifswald) contents oncology vol. , iv s. karger medical and scientific publishers basel � freiburg � paris � london new york � bangalore � bangkok singapore � tokyo � sydney drug dosage the authors and the publisher have exerted every effort to en- sure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. however, in view of ongoing research, changes in government regulations, and the constant flow of informa- tion relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any change in indications and dosage and for added warnings and precau- tions. this is particularly important when the recommended agent is a new and/or infrequently employed drug. all rights reserved. no part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher or, in the case of photocopying, direct payment of a specified fee to the copyright clearance center (see ‘general information’). © copyright by s. karger ag, p.o. box, ch– basel (switzerland) printed in switzerland on acid-free paper by reinhardt druck, basel skp and jab expression are associated with inverse expression of p kip and poor prognosis in oral squamous cell carcinomas shintani, s.; li, c.; mihara, m.; hino, s.; nakashiro, k.-i.; hamakawa, h. (ehime) seprase, a membrane-type serine protease, has different expression patterns in intestinal- and diffuse-type gastric cancer okada, k. (yamanashi/tokyo); chen, w.-t. (stony brook, n.y.); iwasa, s.; jin, x. (tokyo); yamane, t.; ooi, a. (yamanashi); mitsumata, m. (tokyo) loss of p expression is associated with poor prognosis in stage i–ii pancreatic cancer juuti, a.; nordling, s.; louhimo, j.; lundin, j.; von boguslawski, k.; haglund, c. (helsinki) letter to the editor taxol plus gemcitabine and vinorelbine given every other week in advanced chemo-naïve non-small cell lung cancer patients frasci, g.; comella, p.; de cataldis, g.; maiorino, l.; fiore, f.; de rosa, v.; lapenta, l.; licenziato, m.; comella, g. (naples) announcement acknowledgement to the reviewers author index vol. , subject index vol. , suppl. new aspects in the diagnosis and treatment of prostate cancer guest editors: k. miller; t. wiegel; w. hinkelbein (berlin) suppl. ifosfamide: a milestone drug international meeting, padua, february – , guest editor: g. cartei (padua) a comparison of stimulus ratings made online and in person: gender and method effects ps © the psychonomic society, inc.ps the internet has had a profound effect on how psycho- logical science is conducted, and recent years have seen an explosion in the use of the internet for data collection (e.g., birnbaum, ; skitka & sargis, ). research done online is faster, easier, and less expensive to conduct. it also allows for samples that are more diverse in terms of ethnicity, geographical location, education level, age, and socioeconomic status (e.g., birnbaum, ; gosling, vazire, srivastava, & john, ). internet administration of surveys is not without prob- lems (e.g., birnbaum, ; buchanan, ), but grow- ing numbers of researchers are converting to online meth- ods, with the assumption that the results they obtain will be interchangeable with those they would have obtained in the lab. however, this is an underexamined assumption that could lead to problems (cronk & west, ; hewson & charlton, ; see also skitka & sargis, ). there is a growing literature contrasting computerized/ internet and paper-and-pencil survey administration (for reviews, see, e.g., buchanan, ; richman, kiesler, weisband, & drasgow, ). this literature is only tan- gentially related to our present purpose, which is to ex- amine the comparability of data collected over the inter- net and those collected in the traditional way in cognitive psychology, because for several decades this traditional way of collecting data has been computerized but labora- tory based. the issue, therefore, is one of testing envi- ronment: independently of the computer program used to collect the ratings, are different ratings made when a participant must come to the lab to make them? some of the reasons this could be expected to make a difference will be discussed below. testing environment in a small number of studies, the testing environ- ment has been manipulated independently of the form of the survey. cronk and west ( ) assigned a moral- ity questionnaire to be completed via the internet, either in class or at home, and with paper and pencil, again either in class or at home. neither factor had a signifi- cant effect on the mean ratings, although variance and dropout were significantly higher when the internet task was performed at home (as compared with all the other conditions). chuah, drasgow, and roberts ( ) randomly as- signed participants to one of three conditions: paper-and- pencil (in groups of or more), internet (in a computer lab in groups of – ), or internet (alone, whenever and wherever the participant wanted to complete the experi- ment). they analyzed ratings on items from personal- ity inventories, using item response theory. significant differential item functioning was found for item com- parisons, but these were described as likely being due to the large number of comparisons made. analysis of mean differences showed no significant effects when bonfer- roni corrections were applied. the authors concluded that testing condition had no effect on the ratings. dandurand, shultz, and onishi ( ) contrasted two computerized versions of a problem-solving experiment. the internet version was modeled on a lab-based study reported in dandurand, bowen, and shultz ( ); some minor modifications were intended to make the task seem more “attractive” and reduce dropout. the task itself was a complex problem-solving situation in which the partici- pant had to figure out how to determine the lightest and a comparison of stimulus ratings made online and in person: gender and method effects diana a. barenboym, lee h. wurm, and annmarie cano wayne state university, detroit, michigan in experiment , separate samples rated nouns on danger, using either an online survey or the same survey in person. in experiment , a single sample rated words on familiarity, using both methods. women’s in-person and online ratings correlated significantly better than men’s. in-person ratings correlated significantly better with existing norms in of instances. there were significant effects of condition on mean ratings and comple- tion times. ratings from participants who withdrew from the experiment correlated significantly less well with existing norms than did ratings from those who completed the whole experiment, in of instances. analysis of existing data showed that a different statistical conclusion is reached depending on whether in-person or online ratings are used. furthermore, the categorization of . % (experiment ) and . % (experiment ) of the items as high or low depends on which ratings are used. ratings gathered in person and online cannot be freely substituted. behavior research methods , ( ), - doi: . /brm. . . l. h. wurm, lee.wurm@wayne.edu barenboym, wurm, and cano ditions. they examined negative affect as a function of state anxiety. they found a significant mean difference in computerized (not via internet, but in a laboratory room) and paper-and-pencil measures of state anxiety. this dif- ference was observed after (but not before) assignment of a grade for coursework. the results of these five studies are mixed, with three showing measurable meaningful effects of testing condi- tion even with equivalent samples. whitaker and mckin- ney ( ) and norris et al. ( ) demonstrated that even after completely removing what is often the largest single source of variance in psychological studies (i.e., individual differences), a method effect can be observed. however, in none of these studies was the form of the instrument held constant across testing conditions. this complicates their interpretation. although not the primary focus of the present study, we explored the issue of nonequivalence of samples in two experiments. one used different samples for two rating conditions. the other used a single sample. in the latter experiment, potential nonequivalence became relevant because of dropout. what is being rated? in nearly every study that we are aware of, researchers have investigated characteristics of the raters themselves (e.g., personality characteristics, job satisfaction, health behaviors, anxiety, memory). although this is not a short- coming of the existing research, it does leave a gap in the knowledge base in an area that is of central importance in many areas of psychology. in many areas of cognitive psy- chology, for example, it is common practice to ask partici- pants to rate various characteristics of stimuli that are to be used in subsequent experiments (objects, events, words, and so on). it is possible that differential administration is more relevant for ratings of items, because participants may implicitly believe that there are correct and incorrect answers. even if there is no a priori reason to expect that participants will perform differently in in-person versus online studies, one cannot rule out a possible method ef- fect on stimulus ratings, because the existing findings on self-ratings are quite mixed. we are aware of only two studies in which ratings of stimuli have been compared across different conditions. balota, pilotti, and cortese ( ) had three samples of participants rate the subjective familiarity of words: un- dergraduates from the university community, healthy older adults, and an internet sample that ranged in age from to years. each word was rated by either or par- ticipants. ratings for the undergraduates and healthy older adults were administered with paper and pencil, and under- graduates were tested in groups of – . all of the correla- tions between these samples’ ratings were high (all three rs . ), but the internet sample gave significantly higher ratings than did either of the other two (both ps . ). the ratings from the undergraduate sample have become a standard reference in psycholinguistics, and they were used for comparison in experiment of the present study. lahl, göritz, pietrowsky, and rosenberg ( ) gath- ered ratings of german nouns via the internet. each noun heaviest of a set of objects by using a scale a limited number of times. there were three conditions: some participants were told whether their answers were correct (reinforcement); some watched demonstrations of solutions (imitation); and some read instructions on how to solve problems of this type (explicit). in the study, the participants in the reinforcement condition performed significantly worse than those in the other conditions. the study replicated this, although there was also a significant main effect of testing environment: internet participants were less accurate than those in the lab ( % vs. %). they were also more likely to drop out of the study, despite the efforts to make the task seem more attractive. thus, there are mixed findings regarding the impact of the testing en- vironment on task performance. possible nonequivalence of samples if differences are observed between results collected via the internet and those collected in person, they might be due to nonequivalence of samples: most comparisons of testing methods have either accidentally or by design used noncomparable samples of participants, and internet surveys are not likely to produce representative samples (epstein & klinkenberg, ). some researchers have attempted to remedy the prob- lem. joinson ( ), for example, randomly assigned participants to an internet or pen-and-paper condition (crossed with a two-level anonymity factor). they then filled out several personality measures. on social anxiety and social desirability, there were significant differences between the mean scores for the paper-and-pencil version and the internet version. on self-esteem, the difference was marginally significant ( p . ). we will return to the anonymity manipulation below. denscombe ( ) assigned groups of -year-old stu- dents from the same school to complete either a web-based or a paper-and-pencil version of a survey of health-related behaviors. there was an unambiguously statistically differ- ent pattern of responding on only of questions. how- ever, the design was unbalanced ( % of the participants were assigned to the paper-and-pencil condition), and there may have been issues of statistical power as well. three studies have used precisely the same partici- pants in multiple administration conditions. silverstein et al. ( ) compared an internet-based neurocognitive test battery with an established computerized (but non- internet) battery, using the same participants in both conditions. the results suggested that the new battery was perfectly usable: “results indicated comparability across the two batteries” (p. ). whitaker and mckinney ( ) had participants com- plete internet and paper-and-pencil versions of a job satisfaction instrument. they found that job satisfaction ratings were quite comparable across methods (demon- strating measurement invariance), but they also found that a correlation between age and job satisfaction held only for the paper-and-pencil data. norris, pauli, and bray ( ) also made use of a single sample of participants in two measurement con- online versus in-person ratings in experiment , we used two separate samples that were not restricted on gender. in experiment , we used one approximately gender-balanced sample for both parts of the experiment. analytic strategy there are different things that might be meant when researchers speak of results being comparable across rat- ing methodologies or locations. the underlying factor structure of the data or other psychometric indices may be relevant if researchers are validating constructs or devel- oping instruments (e.g., buchanan et al., ; hewson & charlton, ; meyerson & tryon, ). these were not important for our purposes. the mean difference in the ratings may be assessed as a way of concluding whether or not the rating method has an effect. pearson correlation coefficients inform researchers about the extent to which items have the same standard score on two scales. this information is logically inde- pendent of the mean rating, and in fact, potential stimuli can switch their relative positions (and thus perhaps even end up in different “groups” if one were doing such cat- egorizing) without having any effect on the means (see, e.g., buchanan, ; meyerson & tryon, ). in the present study, we used assessments of scale means and correlations to compare the results from different rating conditions. we also compared our ratings with existing norms in experiment , assessing both mean differences and correlations where appropriate. we included participant gender in our statistical mod- els, since differences in the performance of subgroups of participants have not been explored extensively (whitaker & mckinney, ). there is some evidence of differen- tial gender effects in how participants make ratings such as these (e.g., hewson & charlton, ; ullman et al., ) and also in how they respond to them (e.g., wurm, whitman, seaman, hill, & ulstad, ). differences in motivation are often cited as a possible reason for differences between online and offline perfor- mance (buchanan, ). we did not include any direct measures of motivation, but we did have available what we believed to be a proxy in experiment : completion times. completion times are not often available in the published literature, but if they are found to depend on the rating condition, it could be of interest to researchers. finally, we analyzed data from any participants who failed to complete any of the sessions for possible differ- ences from the data of people who finished the experi- ment. dropout is a problem in many research projects, and although we did not expect it to be a major problem in these simple rating tasks, any patterns that began to emerge could inform future research efforts aimed at ex- amining participant motivation, nonequivalence of sam- ples, and so on. we analyzed the ratings (and completion times when available) with a multilevel linear mixed-effects ancova, with participant and item as crossed random effects (baayen, davidson, & bates, ). it simultane- ously included participants and items as random effects, replacing the separate by-subjects and by-items analy- was rated by approximately participants on one of three psycholinguistic variables: concreteness, valence, or arousal. they did not collect any in-person ratings them- selves but compared their results with two existing sets of norms. the authors concluded that there was “good agree- ment” between their obtained ratings and these norms. however, the correlations between the obtained ratings and those from these existing databases ranged from . to . , averaging . . unexplained variance thus ranged from % to %, averaging nearly %. correlations this low suggest that if one were using a factorial design and classifying items into groups (e.g., high and low arousal), one would end up with different sets of items depending on whether internet ratings or exist- ing norms were being used. in the present article, we will demonstrate that this happens. more and more commonly, though, this factorial approach is being abandoned in favor of using the mean item ratings as a continuous predictor variable in regression analyses of data from subsequent experiments that use the stimuli. differential rating perfor- mance in different settings is a potential problem for this approach, too; the correlation coefficients from lahl et al. ( ) indicate substantial variation in the estimation of the item means. we will provide evidence below that this can lead to differing statistical conclusions in regression analy- ses: we would reject the null hypothesis using one set of ratings and fail to reject it using the other set of ratings, even though the two sets of ratings were very highly correlated. one other study that may f it into this category was conducted by krantz, ballard, and scher ( ), who compared ratings of female attractiveness made via the internet with those gathered in the lab. although the study was not described as such, we can imagine a situation in which the images shown to participants would be poten- tial stimuli for subsequent studies. the authors found no significant effects of testing condition on mean ratings and obtained correlations above . between the internet ratings and the in-person ratings. the results of the small number of studies reviewed in this section are thus also mixed. complicating matters, in all of the studies, the form of the rating instrument varied along with the testing location. one additional factor that may have influenced the results is that, in each of the stud- ies, the internet participants (probably) completed the task alone, whereas the in-person participants were tested in groups (except for balota et al.’s [ ] sample of healthy older adults, who were tested at home). we will have more to say about this issue below. the present study adds to the literature in which stimu- lus ratings gathered in different settings have been com- pared. we collected ratings made in a typical laboratory setting, in which a single participant at a time rated a number of stimuli on some dimension. we also collected ratings using an identical rating program, but under what we assume to be typical internet procedures. we will refer to the latter condition as online. the laboratory testing condition will be called in person, even though it, too, was technically done online; the identical rating program was used, but participants had to come to the lab to participate (see the method section of experiment ). barenboym, wurm, and cano were seated at a computer and accessed the rating program, with the experimenter typing in the invitation code. after making sure that the program was running, the experimenter exited the testing room and closed the door. in both conditions, after reading the on-screen consent form, the participants were shown the following text: in this rating questionnaire, you will rate how dangerous for human survival several words are. numbered choices will ap- pear on-screen for you to use in making your rating. take as much time as you need and just use your own judgment. there are no right or wrong answers. if you do not recognize a word, press the key. words appeared to a screen, with each word having the rating scale appear just below it. the -point scale had end points labeled not at all dangerous for human survival ( ) and extremely danger- ous for human survival ( ). trials on which a rating of zero was given were not analyzed. the participants clicked a button to indicate that they were fin- ished with the current screen. they were then shown the next words (until the final group, which had only ). location of words in this list of items was randomized for each participant. credit was granted by the software as soon as the student closed the survey upon completion and went toward a course that they indicated at the time of signing up. results and discussion we calculated the mean rating for each item and con- dition. these means are available from the authors upon request. sixty-five of the items had a mean difference of at least . point in one direction or the other. the range of mean differences was . (ranging from . in one di- rection to . in the other, and representing % of the -point scale). thus, a particular item’s mean rating might differ by nearly a third of the rating scale, depending on the method used to obtain the ratings. this range of mean differences measure is intended to give some indication of the risk of misclassification if one were using ratings to create categories of items, as is often done in psycholinguistic studies. in the present study, we did not intend to classify or categorize items at all, because we were primarily interested in testing hypotheses related to method differences. however, by way of illustration, we used a median split to group the items into high- or low-danger categories, as is common. twenty-one of the items ( . %) had their classification change from high to low danger (or vice versa), depending on whether the in-person or online ratings were used to make the classifi- cation. thus, one’s choice of method for gathering ratings can produce differing lists of items. as will be seen below, one’s choice of method can also determine whether a re- searcher concludes that he or she has a significant effect in regression analyses. the mean ratings (logged because they were posi- tively skewed) did correlate highly, as would be expected [r( ) . for men; r( ) . for women; both ps . ]. a fisher r-to-z transformation showed that the correlation for women was significantly different from that for men (z . , p . ). we analyzed the log danger ratings with a multilevel linear mixed-effects ancova, with participant and item as crossed random effects (see table ). ratings given on- ses usually seen in psycholinguistic research. it is also more powerful than the traditional approach of collapsing observations across participants or across items. unless otherwise noted, all of the analyses in this study were of this type, in which there were multiple observations per participant. there is a debate about the appropriate df for these analyses. the p value typically produced by software packages is based on the upper bound of df, which is equal to the number of participants times the number of items, minus the number of parameters in the model. this p value has been shown to be somewhat anticonservative, and an alternate method for its calculation has been devel- oped (baayen et al., ). in the present study, we report the traditional p value, but all of our significant effects reached significance by both methods. in comparing correlation coeff icients, we usually computed a z score, using the method outlined by meng, rosenthal, and rubin ( ) for comparing correlated correlations. this method was not appropriate for all of our comparisons; those where we did not use it will be marked as fisher r-to-z transformations. experiments a and b danger ratings we have conducted several studies in which we cor- related participants’ mean ratings of the subjective danger and usefulness of word referents with subsequent perfor- mance on cognitive tasks such as naming and lexical de- cision (e.g., wurm, ; wurm & seaman, ; wurm & vakoch, ). the danger ratings have always been collected via a computerized, laboratory-based rating methodology. to determine whether online ratings would be comparable to those collected by our typical method, we ran two concurrent experiments. they were identical except that ratings were gathered online in experiment a and in the lab in experiment b. method participants. fifty-five students ( women) from the psychol- ogy participant pool at wayne state university participated. all were native speakers of english. extra credit in a psychology course was offered in exchange for participation. twenty-five ( women) par- ticipated online, and ( women) in person. none dropped out before completing the entire experiment. gender was independent of condition [ ( ) . , p . ]. materials. stimuli were first constituents of noun–noun compound words used as part of another study (e.g., the cow in cow- boy). these were all fairly common nouns ranging in length from three to eight letters. procedure. two separate studies were posted on a department web page, the only method by which studies may be advertised in our department. the ratings were made within the same online study environment whether the participants made the ratings online or in person. the rating program was, in fact, identical. for the online version, the participants logged onto the experiment web site, read the consent form and instructions, and completed the experiment. they were free to do this whenever and wherever they wanted to. the in-person session was identical, except that the participants had to come to the lab to participate. they called from a waiting room and were met and escorted to the lab by the first author. they then online versus in-person ratings experiment familiarity ratings in addition to using a single sample of participants, in experiment , we gathered survey completion times and more evenly balanced participant gender. we also chose a different variable for rating (familiarity). we had no rea- son to expect that this change would have any implications for the comparison of data between conditions. neither of the rating tasks had a great deal of personal relevance for the participants, so we did not expect a major influence of variables such as social desirability or online disinhibi- tion. any differences we observed would likely require a different explanation. we chose familiarity because it is an intuitive and easy judgment for participants to make. in addition, because familiarity is one of the most exten- sively studied variables in psycholinguistics, there were data available for comparison. our analyses proceeded along the same lines as in experiment . we computed correlation coefficients for the ratings themselves and made comparisons of mean ratings. we could also now compare mean completion times. from this, we might expect to learn something about the conscientiousness of the participants, or how much care and effort they put into the task in different conditions. other statistical tests were possible here that were not possible in experiment , including the relation of our data to established databases (in terms of correla- tions and means) and an analysis of data from partici- pants who did not complete both parts of the experiment (again in terms of correlations, mean ratings, and mean completion times). method participants. we were contacted by students from the psy- chology participant pool at wayne state university ( women and men) about enrollment in the study, of whom provided at least partial data. all were native speakers of english. extra credit in a psychology course was offered in exchange for participation. materials. we randomly selected words from among those that are common to a number of widely used databases containing familiarity norms: the mrc psycholinguistic database (wilson, ), the bristol norms (stadthagen-gonzalez & davis, ), the hoosier norms (nusbaum, pisoni, & davis, ), and the balota norms (balota et al., ). procedure. the experiment was posted on a department web page. the information posted made it clear that the participants had to complete two sessions, one in person and one online, in order to receive credit. students contacted the researcher directly and were randomly assigned to complete either the online or the in-person phase first. if a person was assigned to the online phase first, they were e-mailed a password by which to log in and complete ses- sion . if a person was assigned to the in-person phase first, they were given a lab appointment for session . the procedural specifics for the online and in-person conditions were identical to those in experiment , except for the change in what was being rated. all the ratings were made within the same online program. regardless of condition, after reading the on-screen consent form, the participants saw the following text: words differ in how commonly or frequently they are encoun- tered. some words are encountered very frequently, whereas other words are encountered infrequently. the purpose of this study is to have you rate a list of words with respect to this di- mension. we believe that your ratings will be important to future line were significantly higher than those given in person (m . vs. . on the -point scale; sem . and . , respectively). although small in absolute terms, this is a difference of %. participant gender was quite im- balanced ( % women), so we reran the analysis without it in the model. the main effect of condition remained significant (b . , se b . ; t . , p . ). as was mentioned in the introduction, differing item means could also affect the outcomes of regression analy- ses. as one example of this, we took the existing data from fischer ( ), a study that gathered data on a number of variables of psycholinguistic interest for of the items in the present experiment. we examined the ques- tion of whether the morphological family size (the num- ber of derived words and compound words that contain a particular morpheme; see schreuder & baayen, ) of the first constituent of a compound word (e.g., the cow in cowboy) can be predicted by the rated subjective danger of that constituent. the answer depends on which ratings are used (for in-person ratings, b . , se b . , t . , p . ; for online ratings, b . , se b . , t . , p . ). it is important to note that the correlation between the in-person and online ratings for these items is very high (r . ). nevertheless, this demonstrates very clearly that the statistical conclusion one reaches can depend on whether one uses ratings gath- ered online or in person. we thus have clear evidence that participant ratings of word characteristics can differ substantially as a function of the setting in which they are given. we also have evi- dence that the correlation between online and in-person ratings depends significantly on rater gender. the sample of men, though, was small (n ). we have also seen that the classification of a word as high or low on danger (using a median split) depends on which ratings are used, in nearly one out of every five instances. finally, we have demonstrated that the decision about whether a particular effect is significant in a regression analysis can also de- pend on which ratings are used. perhaps the biggest shortcoming of experiment was the use of two separate participant samples in the two ad- ministration modes, which leaves us unable to conclude that testing condition per se has an effect on ratings (for an extreme example of why this issue is so crucial, see birnbaum, ). table summary of the multilevel ancova for variables predicting log danger ratings regression standard variable coefficient (b) error of b t main effects gender (m) . . . condition (ol) . . . * interaction gender condition . . . note—the label in parentheses shows which level of the two-level fac- tor is to have the coefficient added to the ratings estimate. the default level of these factors (f, ip) is determined alphabetically by the statistical software. “ip” stands for in person, and “ol” stands for online. m, male; f, female. *p . . barenboym, wurm, and cano both ps . ]. although both of these values were ex- tremely high, a fisher r-to-z transformation showed that they were significantly different from each other (z . , p . ). in experiment , too, we saw that the women’s correlation was stronger than the men’s. a multilevel linear mixed-effects ancova with par- ticipant and item as crossed random effects showed that all the predictors had significant main effects on the famil- iarity ratings (table ). the gender coefficient was sub- stantially larger than either of the other main effects. this underscores the importance of assessing gender effects, which is typically not done in research of this type. men’s and women’s ratings correlated very highly [r( ) . , p . ], but the means differed significantly: men gave significantly higher ratings than did women ( . vs. . ; sems . and . , respectively). familiarity ratings were significantly higher in session than in session ( . vs. . ; sems . and . , respectively). as in experiment , condition (online vs. in-person) had a significant main effect. however, whereas in ex- periment the in-person danger ratings were significantly lower than those given online, in experiment familiarity ratings given in person were significantly higher than those given online ( . vs. . ; sems . and . , respec- tively). the apparent discrepancy could be due to differ- ences in the dimension rated (danger vs. familiarity), but analyses presented below suggest a different explanation. condition also interacted with gender: the difference between in-person and online mean ratings was more than twice as large for women as for men. note that this interac- tion is over and above the main effects, including the sub- stantial main effect of gender. post hoc analyses showed that the condition effect was significant for women (b . , se b . ; t . , p . ), but not for men (b . , se b . ; t . , p . ). the online condition produced both the shortest and the longest completion times. for online ratings, the range of times was – min ( . – . in the log units that were analyzed), whereas for in-person ratings, the range of times was – min ( . – . in log units). we analyzed these completion times, after logging them to achieve normality. this analysis was similar to the one studies involving word recognition. the rating scale you should use will be displayed on-screen at all times. words appeared to a screen, with each word having the rating scale appear just below it. the rating scale used was from balota et al. ( ). the participants were asked to rate “how frequently you encounter” each word on a scale from to ( never, once a year, once a month, once a week, once every days, once a day, and several times a day). the participants clicked a button to indicate that they were fin- ished with the current screen. they were then shown the next words. location of the words in this list of items was random- ized for each participant. credit was granted by the software as soon as the student closed the survey upon completion and went toward a course that they indicated at the time of signing up. the software stored not only the rating given, but also the time (in minutes) taken to complete the task and participant gender. timing began when the participants saw the consent form and ended when they clicked to submit their final page of ratings. one week from the date of completion for session , the partici- pants received an e-mail telling them how to perform session . the participants who had completed session in the lab were e-mailed a password to complete session wherever and whenever they wanted to. the participants who had completed session online made an appointment to come to the lab for session . results and discussion we obtained complete data for participants ( % women) and incomplete data from another participants. table presents a summary of the participant contacts. a test of independence showed that the distribution for women differed from that for men [ ( ) . , p . ]. women were much more likely to agree to participate but to fail to complete session (. vs. . ), whereas men were more likely to decline to participate at all (. vs. . ). we will begin with data from the participants who com- pleted both sessions, before presenting analyses of some of the incomplete data. we calculated the mean rating for each item and condition. these means are available from the authors upon request. the range of mean differences across items was . (ranging from . in one direction to . in the other direction and representing % of the -point scale). although this was smaller than the analo- gous effect in experiment , it nevertheless resulted in items ( . %) having their categorization as high or low on familiarity (again using a median split) depend on which ratings were used to make the categorization. this was a smaller percentage than the . % observed in ex- periment but is still rather impressive when we consider that exactly the same people provided the two ratings in the present experiment. as was expected, mean ratings correlated extremely highly [r( ) . for men; r( ) . for women; table number of participants contacting the authors about the study and their relative frequency women men category no. freq. no. freq. completed both phases . . completed only one phase . . inquired but declined to participate . . total contacts . . table summary of the multilevel ancova for variables predicting familiarity ratings regression standard variable coefficient (b) error of b t main effects gender (m) . . . ** session (two) . . . *** condition (ol) . . . *** interactions gender session . . . gender condition . . . * session condition . . . three-way . . . note—the label in parentheses shows which level of the two-level fac- tor is to have the coefficient added to the ratings estimate. the default level of these factors (f, one, ip) is determined alphabetically by the statistical software. “ip” stands for in person and “ol” stands for online. m, male; f, female. *p . . **p . . ***p . . online versus in-person ratings from a number of other databases. the rating scale and instructions for the present study were based on the balota et al. ( ) study, and that is the study with which our ratings correlated most strongly. note that all the correla- tions with the hoosier database were quite low. the famil- iarity scale used by nusbaum et al. ( ) had different anchor points and labels, and as a result, of our words got a rating of (on their -point scale), and none had a mean of less than . . the by-condition differences in these correlations ap- pear small, but several were significant. for both men and women, the in-person ratings correlated significantly bet- ter with the balota norms than did the online ratings. for women, there were no other differences, but for men, two of the other three comparisons reached significance, also favoring in-person ratings. because we used the instructions and rating scale of balota et al. ( ), it is appropriate to compare our mean ratings with theirs. in all four cases (in person and online, for both women and men), t tests showed that our mean ratings were significantly lower than balota et al.’s mean of . (men, in person, t . , p . ; men, online, t . , p . ; women, in person, t . , p . ; women, online, t . , p . ). to summarize this subsection, for both men and women, the in-person familiarity ratings correlated significantly better with the balota norms than the online ratings did. for men, in-person ratings correlated significantly better for two of the other existing databases as well. for both genders and for both online and in-person ratings, the mean ratings for our participants were significantly lower than the mean rating in the balota norms. comparison of completers versus noncompleters. twenty-four participants ( women) completed the on- line phase but failed to keep their subsequent lab appoint- ments. their data may inform us about the performance of nonequivalent samples, since a goodness-of-fit test showed that the gender composition of the completers ( % women) differed from that of the noncompleters ( % women) [ ( ) . , p . ]. the groups did not differ on age [m . for completers, sem . ; for the ratings, except that it was not possible to include item as a random effect. table shows the results of this analysis. the mean log completion time was significantly shorter for the in-person condition and for session . condition also interacted significantly with gender. as with the rat- ings, post hoc analyses showed a significant condition ef- fect for women (b . , se b . ; t . , p . ), but not for men (b . , se b . ; t . , p . ). the results of experiment thus far demonstrate that the correlation between in-person and online ratings was significantly higher for women than for men, even though both were extremely high. furthermore, mean ratings were lower when given online, but only for women. mean completion times were longer online, but again only for women. the differential condition effect for women on the ratings in particular has implications for how one chooses to collect such data. relation to other familiarity databases. the solid boxes in tables (women) and (men) show the correla- tions between the online and in-person ratings and those table summary of the multilevel ancova for variables predicting log completion times regression standard variable coefficient (b) error of b t main effects gender (m) . . . session (two) . . . ** condition (ol) . . . * interactions gender session . . . gender condition . . . * session condition . . . three-way . . . note—the label in parentheses shows which level of the two-level factor is to have the coefficient added to the time estimate. the default level of these factors (f, one, ip) is determined alphabetically by the statisti- cal software. completion times were log transformed prior to analysis. “ip” stands for in person and “ol” stands for online. m, male; f, fe- male. *p . . **p . . table bivariate correlation matrix for several measures of familiarity, female participants online in person (withdrew)a balota bristol hoosier mrcb online . . . . . . in person . . c . . . online (withdrew)a . d . d . . balota . . . bristol . . hoosier . note—boxes show the correlations between the online and in-person ratings and those from other databases. athese data are from women who did not complete the whole experiment (see the comparison of completers versus noncompleters section). bmrc familiarity ratings were inverse and square root transformed to achieve normality (tabachnick & fi- dell, ). ccoefficient differs from the “online” coefficient ( p . ) directly above it. d coefficient differs from both the “in person” and “online” coefficients directly above it (both ps . ). barenboym, wurm, and cano pant and item as crossed random effects (table ). the gender difference observed in the full data set above was seen again here, with men giving higher ratings. comple- tion status was not significant, nor was the interaction of these two factors. we log transformed the completion times to achieve normality and analyzed them with an ordinary multiple regression analysis (table ). note that in this analysis, there was only one observation per participant in the data object (i.e., the person’s completion time for the only ses- sion being considered), and so the traditional p value was not anticonservative here. m . for noncompleters, sem . ; t( ) . , p . ]. we compared the data from these participants with the data from participants who completed both phases of the experiment, using only the online data from those who completed the online phase first. the columns and rows labeled “online (withdrew)” in tables and show the correlations between each item’s mean rating for those participants who withdrew and the other mean ratings discussed above. for women (table ), ratings from the participants who withdrew correlated . with the ratings given by the participants who completed both phases of the experiment, regardless of condition. both of these cor- relations were significantly weaker than the . correla- tion between online and in-person ratings for women who completed the experiment (z . , p . ). for men (table ), ratings from the participants who withdrew cor- related much more weakly (. and . ) with the ratings given by the participants who completed both phases of the experiment. both of these correlations were signifi- cantly weaker than the . correlation between online and in-person ratings for men who completed the experiment (the smaller z was . , p . ). we will have more to say about these very low correlations, and the small num- ber of participants on which they are based, below. several significant differences emerged in the correlations with existing databases, too. the correlations for ratings made by women who withdrew (shown in the dotted box in table ) were significantly weaker than the correlations for ratings made in person, for two of the four databases (smaller z . , p . ). in these same two cases, the correlations in the dotted box were also significantly weaker than those for ratings made online by women who completed the ex- periment (smaller z . , p . ). the results for male noncompleters were even more dramatic. the correlations for ratings made by men who withdrew (shown in the dotted box in table ) were astonishingly weak. for all eight com- parisons, the coefficients in the dotted box are significantly weaker (smallest z . , p . ). as in our analysis above, we analyzed the ratings with a multilevel linear mixed-effects ancova, with partici- table summary of the multilevel ancova for variables predicting familiarity ratings regression standard variable coefficient (b) error of b t main effects gender (m) . . . * withdrew (yes) . . . interaction gender withdrew . . . note—the label in parentheses shows which level of the two-level factor is to have the coefficient added to the ratings estimate. the default level of these factors (f, no) is determined alphabetically by the statistical software. m, male; f, female. *p . . table summary of the regression analysis for variables predicting log completion times regression standard variable coefficient (b) error of b t main effects gender (m) . . . ** withdrew (yes) . . . * interaction gender withdrew . . . * note—the label in parentheses shows which level of the two-level factor is to have the coefficient added to the time estimate. the default level of these factors (f, no) is determined alphabetically by the statistical soft- ware. completion times were log transformed prior to analysis. m, male; f, female. *p . . **p . . table bivariate correlation matrix for several measures of familiarity, male participants online in person (withdrew)a balota bristol hoosier mrcb online . . . . . . in person . . c . . c . c online (withdrew)a . d . d . d . d balota . . . bristol . . hoosier . note—boxes show the correlations between the online and in-person ratings and those from other databases. athese data are from men who did not complete the whole experiment (see the comparison of completers versus noncompleters section). bmrc familiarity ratings were inverse and square root transformed to achieve normality (tabachnick & fi- dell, ). ccoefficient differs from the “online” coefficient ( p . ) directly above it. d coefficient differs from both the “in person” and “online” coefficients directly above it (all ps . ). online versus in-person ratings completers finished the task significantly more quickly than completers; this effect was especially dramatic in the case of men, but it was significant for women, too. can we learn anything from participants? we must be careful not to overinterpret the results of the completers versus noncompleters analyses, because there were so few men in the noncompleter group. this small group is an intriguing puzzle, though. these participants were not identifiable as outliers during data screening. their com- pletion times were all within min of each other (note the sems in table ), but these were not the shortest times in the data set. the distribution of logged completion times was quite normal (skew . , kurtosis . ; both ps . ). nor could these participants be identified as outliers on the basis of their means or variances: none had either the highest or the lowest mean rating, and none had either the highest or the lowest variance. the concern about a sample size of remains, but ad- ditional analyses argue that the responses of male non- completers truly were different from those of other subsets of participants and that the results were not simply due to the small sample size. the first row of table shows the correlation coefficients for ratings from our male non- completers and existing norms (copied from table ). for row of the table, we took random samples of size n from the men who completed the experiment. for each of these random samples, we calculated the correla- tion between the online ratings and existing norms. the values shown are the averages of those coefficients. rows and were calculated in an analogous way for the female participants. for all databases, the coefficient shown in row dif- fered significantly from all of the others (smallest z . , p . ). in fact, not a single one of the new cor- relation coefficients computed in constructing table was as weak as its corresponding value in row . it is thus not the case that any sample of size n will produce the extremely poor correlations that our male noncompleters did. there must be some other explanation. prudence dic- tates that these findings be replicated with larger samples, but this is an intriguing avenue for future research. general discussion the present study adds to the small literature looking at the effects of testing condition on ratings of potential stimuli, rather than on people’s ratings of themselves. we both main effects were significant. the participants who withdrew performed the task more quickly than those who returned to complete the experiment, and men performed the task more quickly than women. over and above the main effects of gender and completion status, there was also a significant interaction between the two. men who withdrew performed session in less than half the time, on average, as compared with the other partici- pants (table ). post hoc analyses showed that the effect of completion status on completion time was significant both for women (b . , se b . ; t . , p . ) and for men (b . , se b . ; t . , p . ). because of the small number of men in the noncom- pleters group, we performed a nonparametric test on the ranks of the completion times, using the mann–whitney u test. this test makes no assumptions about the normality of the underlying data or the relative sizes of the samples, but it does account for imbalances in these relative sizes. this analysis, too, showed that the effect of completion status on completion times was significant for men (z . , p . ). finally, t tests showed that for women, the noncom- pleters’ mean was significantly lower than the balota et al. ( ) mean [t( ) . , p . ]. for men, the noncompleters’ mean was significantly higher than the balota et al. mean [t( ) . , p . ]. to summarize this subsection, noncompleters and com- pleters differed in their gender composition but not in their ages. noncompleters’ ratings correlated significantly less well with existing norms than did completers’ ratings, in of the comparisons. female noncompleters had a significantly lower mean rating than the balota norms, but male noncompleters had a significantly higher mean. non- table mean completion times (with standard errors of the means) as a function of participant gender and completion status (in minutes) completed completed both phasesa online only gender m sem m sem women . . . . men . . . . note—completion times were log transformed prior to analysis. aonly the data from the online phase are included, and only if that phase was session . table comparison of bivariate correlations for very small random samples of participants balota bristol hoosier mrcb online, men who withdrew . c . c . c . c online, men who did not withdrawa . . . . online, women who withdrewa . . . . online, women who did not withdrawa . . . . acorrelations shown are averages computed over random samples of size n . bmrc familiarity ratings were inverse and square root transformed to achieve normality (tabachnick & fidell, ). ccoefficient differs from all three coefficients below it ( p . ). barenboym, wurm, and cano in particular) has implications for how one chooses to col- lect rating data. in experiment , we were able to assess the correlations between our data and existing norms. correlations varied as a function of testing condition in half of the compari- sons. for both men and women, ratings made in person correlated with the balota et al. ( ) database signifi- cantly better than did ratings made online. for men, two of the other three comparisons were also significant, both again favoring the ratings made in person. these findings make sense, since none of these other norms were col- lected via the internet. these results suggest that one can- not simply assume equivalence across testing conditions. more widespread dissemination of norms gathered online would help researchers make more appropriate compari- sons of studies. our study also has something to say about nonequiva- lence of samples because noncompleters differed signifi- cantly from the completers in terms of gender composi- tion. this matters because gender had a main effect in three of the five analyses, and when it was significant, it tended to have a very large coefficient. this makes sense in that gender is necessarily a between-subjects effect, whereas the other effects were usually within-subjects ef- fects (as was noted above, individual differences constitute the single largest source of variation in many psychology studies). gender often interacted, too, over and above its significant main effect. the participants who completed both phases of ex- periment and those who dropped out produced dramati- cally nonequivalent data. noncompleters’ correlations with existing databases were significantly lower than the corresponding correlations for completers, in of instances. further analyses showed that noncompleters (especially men) made their ratings significantly more quickly than completers, too. male noncompleters were also the only subgroup to have a significantly higher mean rating than the balota et al. ( ) database. as was mentioned above, we must be careful about these noncompleter results, because there were so few men in the noncompleter group. supplemental analyses showed, though, that a small sample size does not by itself produce the pattern of results seen for our male noncom- pleters (see table ). this intriguing issue awaits addi- tional research. several effects in experiment appear to be numeri- cally small even when statistically significant. for exam- ple, although significant at p . , the mean difference in ratings as a function of testing condition was only about one tenth the size of the mean difference in experiment . this is not altogether surprising, given that the data in experiment came from two different samples of par- ticipants, whereas the data in experiment came from one sample, measured twice. correlation differences, too, were larger in experiment than in experiment , even when significant in the latter. we echo buchanan ( ), who concluded in a review of the generally small method effects in studies like ours, that “the practical significance of these differences will vary depending on how the tests did not have strong reasons to believe that testing condi- tion would be more relevant for ratings of items than for self-ratings, except insofar as the participants might be- lieve that there were correct and incorrect answers for rat- ings of words. however, we still were not able to rule out a possible condition effect on stimulus ratings, because even the existing findings on self-ratings are so equivocal. in experiment , with two separate samples, we found a high correlation between mean online and in-person rat- ings for women and a significantly weaker correlation for men. in experiment , with a single sample, the correla- tions were extremely high—but even here, the correlation for women was significantly stronger than that for men. we thus have evidence that the relationship between on- line and in-the-lab performance is stronger for women, whether these comparisons are between subjects (experi- ment ) or within subjects (experiment ). even with the strong correlations, there was substantial variation in the means obtained in the two conditions for particular items (representing % of the rating scale in experiment and % of the scale in experiment ). such differences could easily result in misclassifications if one were constructing groups of items in different ranges on the scale (e.g., high vs. low). we showed that even in ex- periment , with identical participants making the ratings at both times and with the correlation between the two sets of ratings being nearly perfect (r . ), this happened more than % of the time. in experiment , which we think more nearly approximated real-world testing situations (see below), items were differentially categorized as high versus low nearly % of the time, even though the two sets of ratings correlated . . a less-than-perfect correla- tion, no matter how strong it appears, is thus no guarantee that one set of ratings can be used in place of another. if the mean values are instead to be used as continu- ous predictor variables in regression models, mean dif- ferences such as these are likely to affect the conclusions of regression analyses. in our analysis of the data from fischer ( ), we showed that by using one set of rat- ings, it is possible to predict morphological family size, but if we use the other set of ratings, it is not possible. unfortunately we did not have access to a similar data set with which to test the experiment ratings in this way, but our finding will nevertheless be of interest to researchers who use ratings to classify items for further use or analy- sis. clearly, additional research is needed to determine the extent to which shifting item classification occurs across different item types and methods, but these results suggest that researchers should consider how their method might influence classification even if the ratings derived from different methods are highly correlated. we also found interactions between gender and testing condition in experiment . we do not have a compelling explanation for these interactions based on gender roles, technophobia, or personality characteristics. neverthe- less, these findings will be of interest to researchers who recruit from undergraduate psychology participant pools, which typically consist of a majority of women. the dif- ferential condition effect for women (on the mean ratings, online versus in-person ratings ing these like the supervised/proctored conditions in the studies above). the only thing that differed was the in- struction that the participants were given. the manipula- tion had strong effects: social anxiety and social desirabil- ity scores were significantly higher in the nonanonymous condition, whereas self-esteem was significantly higher in the anonymous condition. in the present study, we opted not to introduce any ma- nipulations having to do with supervision or anonymity. instead, both testing conditions closely approximated cur- rent operating procedures. in between-subjects designs like our experiment , we believe that perceived supervi- sion and accountability are much higher in person, espe- cially given the fact that our participants took part in the experiment one at a time. in experiment , with a within- subjects design, we observed a smaller effect of testing condition; but even here, the participants knew that they would have to do half of the study in person, so perceived supervision was probably higher for the in-person condi- tion. perceived accountability may have been comparable in the two conditions, insofar as the participants knew that they were expected to participate in both phases of the study and that we must, therefore, have had some way of keeping track of which ratings belonged to which partici- pant even in the online condition. however, we did not as- sess the perceptions of supervision or accountability, and neither did cronk and west ( ) or chuah et al. ( ). also, like joinson ( ), we did not assess perceptions of anonymity. these should be investigated more systemati- cally in future research comparing identical computerized administrations in different settings. conclusions and directions for future research the question of whether the traditional laboratory- based administration or internet administration is to be preferred does not have a straightforward answer. in the context of gathering ratings to be used in a subsequent psycholinguistic study, it seems likely to us that the tradi- tional in-person setting would lead to a more conscientious effort by participants, who may believe that their behavior is being observed, at least indirectly, by the experimenter or that their data can be identified as theirs (e.g., smyth, dillman, & christian, ). mean completion time was longer online (for women), which may appear to argue that online performance is more conscientious, but we must exercise caution in interpreting completion times because of the possibility of a curvilinear relationship. as was mentioned above, the online participants in experi- ment provided not only the shortest completion times, but also the longest (perhaps they actually walked away from the task for minutes at a time or took a phone call). internet administration of surveys is valid under many circumstances, and in some cases, it is clearly preferable. joinson ( , ) noted that there may be less serious social desirability effects with online administration and that participants may be more honest or forthcoming with information. in our study, because the participants were rating words rather than themselves, issues such as these seem unlikely to have been behind our significant condi- tion effects. are to be used” (p. ). assessment of practical signifi- cance is made even harder because ours is the first study to compare stimulus ratings, holding the form of the in- strument constant, with a within-subjects design. the larger effects might be the more realistic estimates, because our experience in recruiting participants for these two experiments suggests the presence of two virtually nonoverlapping populations of participants: those who do not mind coming to the lab and those who would strongly prefer not to. not everyone who refused to participate stated a reason, but those who did invariably said that they did not want to come for an in-lab session. in addition, despite the study description saying that there were two re- quired phases, several of these participants asked whether they could do just the online phase. the present study also adds to the small literature in which the same computer administrations have been com- pared in different settings. as was discussed in the intro- duction, cronk and west ( ) found that mean ratings were the same for an internet version of a morality ques- tionnaire whether administered in class or at home, and chuah et al. ( ) similarly concluded that ratings on personality inventories were equivalent whether collected via the internet in a computer lab in groups of – or alone at home. these conclusions contrast with our own, in that we found a significant effect of testing condition in both experiments. epstein and klinkenberg ( ) noted that “researchers lose almost all hope at controlling the experimental envi- ronment when they decide to collect data via the internet” (p. ; see also krantz et al., ). in the laboratory set- ting, there is much more control over the environment, but even here there can be variation along dimensions such as perceived supervision and accountability. in our view, this variation offers the most likely explanation for the con- trasting conclusions of the present study and earlier work. perceived supervision and accountability cronk and west ( ) and chuah et al. ( ) included a manipulation of supervision or proctoring in their stud- ies. both studies concluded that the factor did not affect the data in important ways. this methodology sounds similar to that in the present study, except that, both in cronk and west and in chuah et al., the presence or ab- sence of an experimenter was confounded with the pres- ence or absence of other participants: in the unsupervised/ unproctored conditions, participants were assumed to be alone while completing the survey, but the supervised/ proctored conditions always took place in groups. the presence of an experimenter in these studies can be pre- sumed to have increased perceived accountability, but the presence of (sometimes many) other participants would seem to have worked against this manipulation. perhaps the effect of supervision could be diluted by the presence of these other participants. a related but separate issue is perceived anonymity. joinson ( ) told half of his participants that their re- sponses were anonymous and could not be linked to them. testing of all participants was done in groups, in the same physical setting, with an experimenter present (thus mak- barenboym, wurm, and cano in a between-subjects design. psychological methods, , - . doi: . / - x. . . birnbaum, m. h. ( ). human research and data collection via the internet. annual review of psychology, , - . doi: . / annurev.psych. . . buchanan, t. ( ). personality testing on the internet: what we know, and what we do not. in a. n. joinson, k. mckenna, t. postmes, & u.-d. reips (eds.), oxford handbook of internet psychology (pp. - ). oxford: oxford university press. buchanan, t., ali, t., heffernan, t. m., ling, j., parrott, a. c., rodgers, j., & scholey, a. b. ( ). nonequivalence of on-line and paper-and-pencil psychological tests: the case of the prospective memory questionnaire. behavior research methods, , - . chuah, s. c., drasgow, f., & roberts, b. w. ( ). personality as- sessment: does the medium matter? no. journal of research in per- sonality, , - . doi: . /j.jrp. . . cronk, b. c., & west, j. l. ( ). personality research on the internet: a comparison of web-based and traditional instruments in take-home and in-class settings. behavior research methods, instruments, & computers, , - . dandurand, f., bowen, m., & shultz, t. r. ( ). learning by imitation, reinforcement, and verbal rules in problem-solving tasks. in j. triesch & t. jebara (eds.), proceedings of the third international conference on development and learning: developing social brains (pp. - ). la jolla: university of california, san diego, institute for neural computation. dandurand, f., shultz, t. r., & onishi, k. h. ( ). comparing online and lab methods in a problem-solving experiment. behavior research methods, , - . doi: . /brm. . . denscombe, m. ( ). web-based questionnaires and the mode effect: an evaluation based on completion rates and data contents of near- identical questionnaires delivered in different modes. social science computer review, , - . doi: . / epstein, j., & klinkenberg, w. d. ( ). from eliza to internet: a brief history of computerized assessment. computers in human be- havior, , - . doi: . /s - ( ) - fischer, k. ( ). constituent usefulness effects on the recognition of compound words. unpublished honors thesis, wayne state university. gosling, s. d., vazire, s., srivastava, s., & john, o. p. ( ). should we trust web-based studies? a comparative analysis of six preconcep- tions about internet questionnaires. american psychologist, , - . doi: . / - x. . . hewson, c., & charlton, j. p. ( ). measuring health beliefs on the internet: a comparison of paper and internet administrations of the multidimensional health locus of control scale. behavior research methods, , - . joinson, a. ( ). social desirability, anonymity, and internet-based questionnaires. behavior research methods, instruments, & comput- ers, , - . joinson, a. ( ). knowing me, knowing you: reciprocal self- disclosure in internet-based surveys. cyberpsychology & behavior, , - . doi: . / krantz, j. h., ballard, j., & scher, j. ( ). comparing the results of laboratory and world-wide web samples on the determinants of female attractiveness. behavior research methods, instruments, & computers, , - . lahl, o., göritz, a. s., pietrowsky, r., & rosenberg, j. ( ). using the world-wide web to obtain large-scale word norms: , ratings on a set of , german nouns. behavior research methods, , - . doi: . /brm. . . meng, x. l., rosenthal, r., & rubin, d. b. ( ). comparing corre- lated correlation-coefficients. psychological bulletin, , - . doi: . / - . . . meyerson, p., & tryon, w. w. ( ). validating internet research: a test of the psychometric equivalence of internet and in-person samples. behavior research methods, instruments, & computers, , - . norris, j. t., pauli, r., & bray, d. e. ( ). mood change and com- puter anxiety: a comparison between computerised and paper mea- sures of negative affect. computers in human behavior, , - . doi: . /j.chb. . . nusbaum, h. c., pisoni, d. b., & davis, c. k. ( ). sizing up the hoosier mental lexicon: measuring the familiarity of , words researchers doing pilot studies can make educated guesses about variables such as personal relevance, like- lihood of socially desirable responding, and whether anonymity will be desired. investigations with stronger anonymity manipulations (and assessments of their effec- tiveness) will be very useful. teasing apart anonymity and supervision effects will also be important. the full ben- efits of internet-based research (e.g., testing thousands of participants from all over the world) are available only in conditions in which supervision is impossible. anonym- ity, on the other hand, can be varied to a certain extent even with internet studies. if psychology participant pools are to be used, gender will be a thorny issue. whenever such participants are al- lowed unrestricted freedom to volunteer for a study, the sample is likely to be overwhelmingly female. we saw this in experiment ( % women) and in the self-selected noncompleters in experiment ( % women). research- ers have known for a long time that random samples from psychology participant pools are not likely to be repre- sentative of the population to which they usually want to generalize. it could be, though, that artificially creating representativeness on gender brings with it other prob- lems. as the present study has shown, gender is a factor that can have a very large main effect and can interact with other variables. future research should determine whether different par- ticipants volunteer for lab-based and online studies. this relates to our observation of differences between com- pleters and noncompleters. noncompleters can be con- ceptualized in at least two non-mutually-exclusive ways. they might be representative of less responsible partici- pants, or they might be representative of participants who favor online studies to in-person studies. willingness to travel to a lab is almost certainly a graded dimension, and our noncompleters probably fall closer to the avoid in- person studies end than our completers do. researchers could easily enough develop a measure of this willing- ness and look at how it relates to performance in a variety of tasks and settings. additional work on this question, including more direct assessments of conscientiousness and motivation, will increase our understanding of the complex issues involved. author note we thank sean seaman for a valuable discussion about aspects of our experimental design. correspondence concerning this article should be addressed to l. h. wurm, department of psychology and program in linguistics, wayne state university, woodward avenue, th floor, detroit, mi (e-mail: lee.wurm@wayne.edu). note—accepted by the previous editorial team, when john h. krantz was editor. references baayen, r. h., davidson, d. j., & bates, d. m. ( ). mixed-effects modeling with crossed random effects for subjects and items. journal of memory & language, , - . doi: . /j.jml. . . balota, d. a., pilotti, m., & cortese, m. j. ( ). subjective fre- quency estimates for , monosyllabic words. memory & cogni- tion, , - . birnbaum, m. h. ( ). how to show that : collect judgments online versus in-person ratings pancheva, r., ozawa, k., et al. ( ). sex differences in the neuro cognition of language. brain & language, , - . whitaker, b. g., & mckinney, j. l. ( ). assessing the measure- ment invariance of latent job satisfaction ratings across survey ad- ministration modes for respondent subgroups: a mimic modeling approach. behavior research methods, , - . wilson, m. ( ). mrc psycholinguistic database: machine-usable dictionary, version . . behavior research methods, instruments, & computers, , - . wurm, l. h. ( ). danger and usefulness: an alternative framework for understanding rapid evaluation effects in perception? psychonomic bulletin & review, , - . wurm, l. h., & seaman, s. r. ( ). semantic effects in naming and perceptual identification, but not in delayed naming: implica- tions for models and tasks. journal of experimental psychology: learning, memory, & cognition, , - . doi: . / - . . . wurm, l. h., & vakoch, d. a. ( ). the adaptive value of lexical connotation in speech perception. cognition & emotion, , - . doi: . / wurm, l. h., whitman, r. d., seaman, s. r., hill, l., & ulstad, h. m. ( ). semantic processing in auditory lexical decision: ear- of-presentation and sex differences. cognition & emotion, , - . doi: . / (manuscript received february , ; revision accepted for publication october , .) (research on speech perception progress report no. ). blooming- ton: indiana university, department of psychology, speech research laboratory. richman, w. l., kiesler, s., weisband, s., & drasgow, f. ( ). a meta-analytic study of social desirability distortion in computer- administered questionnaires, traditional questionnaires, and inter- views. journal of applied psychology, , - . doi: . / - . . . schreuder, r., & baayen, r. h. ( ). how complex simplex words can be. journal of memory & language, , - . doi: . / jmla. . silverstein, s. m., berten, s., olson, p., paul, r., williams, l. m., cooper, n., & gordon, e. ( ). development and validation of a world-wide-web-based neurocognitive assessment battery: web- neuro. behavior research methods, , - . skitka, l. j., & sargis, e. g. ( ). the internet as psychological laboratory. annual review of psychology, , - . doi: . / annurev.psych. . . smyth, j. d., dillman, d. a., & christian, l. m. ( ). context ef- fects in internet surveys: new issues and evidence. in a. n. joinson, k. mckenna, t. postmes, & u.-d. reips (eds.), oxford handbook of internet psychology (pp. - ). oxford: oxford university press. stadthagen-gonzalez, h., & davis, c. j. ( ). the bristol norms for age of acquisition, imageability, and familiarity. behavior re- search methods, , - . tabachnick, b. g., & fidell, l. s. ( ). using multivariate statis- tics ( th ed.). boston: allyn & bacon. ullman, m. t., estabrooke, i. v., steinhauer, k., brovetto, c., reviews and notices prehistoric antiquicies of indiana. by eli lilly. indiana his- torical society, indianapolis, . pp. xiii, , il- lustrated, $ . . one is seldom privileged t o review a publication which combines so many outstanding innovations. this d e zuze vol- ume will serve as a lasting memento t o the author; bring superlative commendations to the indiana historical society ; and establish a unique standard for archeological publications. it will give to the present generation in the hoosier state a re- view of past archeological records, and enable everyone to grasp the full significance of the prehistoric heritage in the state. the author has very modestly set forth in the preface the real purpose of this book: “the object in writing . . . was to interest more of the people of indiana in the relics of our vanished predecessors, and to stimulate inquiry into the prehistory and archaeology of our state. the pursuit of this subject has led the author along such pleasant paths that he is desirous of sharing them with others.” the reviewer was privileged t o participate in the fascinating work of recovering a very small part of the prehistory of indiana. he feels it a duty, therefore, t o point out that the science of anthropology, especially archeology, has grown by leaps and bounds not only in indiana, but in the entire northern mississippi valley be- cause of mr. lilly’s enthusiastic response t o the problems con- fronting scientists in this region. indiana has now passed the initial step of outlining the important prehistoric problems within the state and we can look forward to the fruitful phase which will follow under the expert guidance of the archaeological staff. much is yet to be accomplished, but with the highest hurdle cleared and inches t o spare, the remaining problems will be taken in a natural stride. there are two methods now being followed by the pro- fessional archaeogolist in obtaining a true picture of the pre- history of a state. one method is t o sample as many of the outstanding sites as possible in order t o obtain an outline o r “bird’s-eye view” of the prehistoric picture. another method is t o survey systematically the state by county units o r stream valleys, making sufficient excavations t o obtain a cross sec- indiana magazine of history tion within these units of the aboriginal material cultures. the latter requires planned research, often unlimited patience on the part of the investigator and sponsor, but in the end will give the most accurate history of the first inhabitants. it is a pleasure to report that indiana has willingly responded t o the latter approach. prehistoric antiquities of indiana presents in a form unsurpassed a compilation based on the work of various investigators whose programs of research have been based on systematically surveying the more im- portant river valleys, using the county as a unit. this may serve as a model for other state organizations, especially when the dry initial reports are so entertainingly sum- marized. members of the indiana historical society will cherish the volume even though they have no interest in the re- creation of prehistory. this quarto volume is beautifully printed by the lakeside press on ivory wove linweave text paper with pt. baskerville type. it contains eighty-eight pages of illustrations produced by the donnelley deeptone process, a recent development of deep-etched lithography. the end papers carry a panoramic view of the great circle at mounds park, now well protected as a state park near anderson, indiana. the illustrations have been faithfully reproduced from prints made by an expert photographer. the artistic arrangement of the specimens is quite unusual, especially for archeological reports. the mounting on such backgrounds as white sand, bear robes, tree sections, old survey maps of archeological sites are quite effective. but the arrangement of discoidals on an illustration of a carved shell gorget depicting the use of discoidals and carved stone tobacco pipes mounted on a pressed specimen of nicotiana attenuuta produces very pleasant, attractive, and novel il- lustrations when compared t o the more usual black and white reproductions. the chapter initials very appropriately show design elements from carved bones recovered in some of the ohio hopewell mounds. all of this is handsomely bound be- tween green and brown buckram cloth boards, with gold- stamped front cover and backbone. one must see the book to appreciate it fully. the story told by prehistoric antiquities of indiana. is just as entertainingly written as the format suggests. it com- bines mcdern viewpoints derived from the sound theories reviews and notices upon which they are based. any difference of opinion that might exist between the author and other specialists in the field is of such a nature that even the professionals are di- vided as to the correct solution. mr. lilly touches upon con- troversial subjects and sublimates the theories of all in n o t e worthy style. the abundant use of footnotes not only clarifies details, but credits everyone who has contributed anything t o the archaeology of the state. the first three chapters have the following headings : “the original antiquity of the american indian” “a tentative outline of the prehistory of indiana” (to the reviewer this is the most important chapter and should be read not only by the archeologist, but by everyone interested in the history of this state. the outline is one of the most clear cut and up-to-date presentations of these problems.) “some striking prehistoric sites in indiana” (herein is reviewed all previous research. it can serve as an excellent guide for the expansion of state parks and recreation areas which would preserve these remnants of a prehistoric civilization.) the remaining seven chapters deal primarily with de- scriptions, classifications and illustrations of the various types of artifacts from the entire state. another important contribution and a subject which first brought mr. lilly t o the attention of archeologists is an archaeological biblography of indiana, which he began many years ago and published in may, . the bibliography in- cluded in this volume is revised and considerably enlarged. a publications, after the more general ones, have been al- phabetically arranged according t o author and the counties within the state. in closing, the reviewer wishes t o offer his sincere con- gratulations t o mr. lilly for establishing this high standard for archaeological publications and t o express complete satis- faction in the methods used by him t o place the study of in- diana archaeology before the public and scholars in such a handsome form. frank m. setzler. research open access novel alternative splicing isoform biomarkers identification from high-throughput plasma proteomics profiling of breast cancer fan zhang , *, mu wang , , tran michael , renee drabier * from the international conference on intelligent biology and medicine (icibm ) nashville, tn, usa. - august abstract background: in the biopharmaceutical industry, biomarkers define molecular taxonomies of patients and diseases and serve as surrogate endpoints in early-phase drug trials. molecular biomarkers can be much more sensitive than traditional lab tests. discriminating disease biomarkers by traditional method such as dna microarray has proved challenging. alternative splicing isoform represents a new class of diagnostic biomarkers. recent scientific evidence is demonstrating that the differentiation and quantification of individual alternative splicing isoforms could improve insights into disease diagnosis and management. identifying and characterizing alternative splicing isoforms are essential to the study of molecular mechanisms and early detection of complex diseases such as breast cancer. however, there are limitations with traditional methods used for alternative splicing isoform determination such as transcriptome-level, low level of coverage and poor focus on alternative splicing. results: therefore, we presented a peptidomics approach to searching novel alternative splicing isoforms in clinical proteomics. our results showed that the approach has significant potential in enabling discovery of new types of high-quality alternative splicing isoform biomarkers. conclusions: we developed a peptidomics approach for the proteomics community to analyze, identify, and characterize alternative splicing isoforms from ms-based proteomics experiments with more coverage and exclusive focus on alternative splicing. the approach can help generate novel hypotheses on molecular risk factors and molecular mechanisms of cancer in early stage, leading to identification of potentially highly specific alternative splicing isoform biomarkers for early detection of cancer. introduction a biomarker as defined by the national cancer institute is “a biological molecule found in blood, other body fluids, or tissues that is a sign of a normal or abnormal process, or of a condition or disease [ ].” it is a charac- teristic that is objectively measured and evaluated as an indicator of normal biological processes, pathogenic pro- cesses, or pharmacologic responses to a therapeutic intervention [ ]. the field of biomarkers has grown extensively over the past decade in many areas such as medicine, cell biology, genetics, geology and astrobiol- ogy, and ecotoxicology etc. and biomarkers are currently being studied in many academic centers and in industry. in recent years, functional genomics studies using dna microarrays have been shown effective in identify- ing markers differentiating between breast cancer tissues and normal tissues, by measuring thousands of differen- tially expressed genes simultaneously [ - ]. however, early detection and treatment of breast can- cer is still challenging. one reason is that obtaining tissue samples for microarray analysis can be still difficult. another reason is that genes are not directly involved in any physical functions. on the contrary, the proteome, are the real functional molecules and the keys to * correspondence: fan.zhang@unthsc.edu; renee.drabier@unthsc.edu department of academic and institutional resources and technology, university of north texas health science center, fort worth, usa full list of author information is available at the end of the article zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s © zhang et al.; licensee biomed central ltd. this is an open access article distributed under the terms of the creative commons attribution license (http://creativecommons.org/licenses/by/ . ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. the creative commons public domain dedication waiver (http:// creativecommons.org/publicdomain/zero/ . /) applies to the data made available in this article, unless otherwise stated. mailto:fan.zhang@unthsc.edu mailto:renee.drabier@unthsc.edu http://creativecommons.org/licenses/by/ . http://creativecommons.org/publicdomain/zero/ . / http://creativecommons.org/publicdomain/zero/ . / understanding the development of cancer. moreover, the fact that breast cancer is not a single homogeneous dis- ease but consists of multiple disease status, each arising from a distinct molecular mechanism and having a dis- tinct clinical progression path [ ] makes the disease diffi- cult to early detect. alternative splicing isoforms represent a new class of diagnostic biomarkers [ ]. the chance of success with alternative splicing isoforms would be higher than the conventional approach [ , ]. alternative splicing occurs in % human genes and works by selecting specific exons and sometimes even intronic regions of the gene into mature mrnas [ ]. alternative splicing accounts for approximately % of all protein isoforms which is any of several different forms of the same protein and have three types: alternative splicing, snp, and posttran- slational modification (ptm). recent scientific studies have shown that diseased cells may produce many types of splicing variants of common regulatory proteins, e.g., protein kinase c, - - , p , and vgfr, which could provide novel insights into complex disease diagnosis and management, particularly in cancers [ - ]. alternative mrna splicing is an important source of achieving molecular functional diversity. it is often regulated in a temporal or tissue- specific fashion, giving rise to different protein isoforms in different tissues or developmental states mediated by extracellular signaling mechanisms [ , ]. splicing reg- ulation is a key mechanism to tune gene expression to a variety of conditions and its dysfunction may often be at the basis of the onset of genetic disease and cancer [ ]. in cancer, many examples of alternative splicing iso- forms were reported [ - ]. for example, julian et al. used a high-throughput reverse transcription-pcr-based system for splicing annotation to monitor the alternative splicing profiles of cancer-associated genes in a panel of normal and cancerous breast tissues. they found that alternative splicing events signifi- cantly differed in breast tumors relative to normal breast tissues and that most cancer-specific changes in splicing that disrupt known protein domains support an increase in cell proliferation or survival consistent with a func- tional role for alternative splicing in cancer. compared to normal mrna splicing events, alternative splicing mechanisms and patterns in complex diseases such as cancer can be quite complex. finding alternative splicing isoforms or patterns of their development, therefore, have been promising in helping develop high-quality biomarkers and targets for disease management [ ]. discovering disease biomarkers in the human plasma has been met with both enthusiasm and criticisms in recent years. on one hand, it is expected that disease conditions such as cancer may be diagnosed early by ana- lyzing complex protein mixtures in easily accessible human blood (serum or plasma), in which proteins induced by cancer may be differentially cleaved, secreted, leaked out, and therefore differentially detected from nor- mal healthy conditions. on the other hand, the bulk of proteins circulating in human blood vary in different con- ditions without cancers and across different individuals, and changes in protein expressions are often diluted in the blood – extremely challenging for biological interpre- tation of protein quantification changes [ , ]. the use of alternative splicing isoforms as potential biomarkers, therefore, offers a new opportunity to use the detectabil- ity instead of quantification of biomarker peptides, for the peptides that may map to unique alternative splicing isoforms specific to cancer. however, systematic, comprehensive, proteome-scale experimental and computational characterization of pro- tein isoforms directly at the protein/peptide level and with exclusive focus on alternative splicing has never been reported. compared with the “indirect” transcrip- tome-level characterizations such as est sequencing [ ], exon array[ ], exon-exon junction array[ ], and next-generation sequencing of all mrna transcripts [ , ], “direct” proteome-level characterization of alter- native splicing isoforms can address the wide-spread concern that mrna expressions and protein expression do not correlate well in a clinical proteomics [ , ]. in , we developed the peptidomics protein isoform database (peppi [ ], http://bio.informatics.iupui.edu/ peppi), a database of computationally-synthesized human peptides that can identify protein isoforms derived from either alternatively spliced mrna tran- scripts or snp variations. although the peppi database is the first peptidomics protein isoform database, it is not exclusive to alternative splicing and has poor cover- age of alternative splicing. therefore, based on the peppi [ ], we presented a new peptidomics approach to searching novel alternative splicing isoform in proteomics data and demonstrated that the approach can help researchers identify and characterize novel alternative splicing isoform from experimental proteomics data and discover diagnostic value and clinical significance. materials and methods human plasma samples plasma protein profiles were collected by the hoosier oncology group (hog) (indianapolis, in, usa). each sample was analyzed in a single batch by mass spectro- metry. in the study, plasma samples were collected ( samples collected from women with breast cancer and from healthy volunteer woman who served as controls). an independent validation dataset of sam- ples which contains samples collected from women diagnosed with breast cancer and from healthy zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of http://bio.informatics.iupui.edu/peppi http://bio.informatics.iupui.edu/peppi volunteer woman who served as controls were collected by the hoosier oncology group (hog) (indianapolis, in, usa) too and is comparable to the study in the demography and clinical distribution of breast cancer stages/subtypes. for example, most of patients involved in the two studies were diagnosed with an early stage breast cancer (stage i or ii), fell into age group between and , and had mean tumor size of . . protein identification and quantification for protein identification, tryptic peptides were ana- lyzed using thermo-fisher scientific linear ion-trap mass spectrometer (ltq) coupled with a surveyor hplc system. peptides were eluted with a gradient from to % acetonitrile developed over minutes and data were collected in the triple-play mode (ms scan, zoom scan, and ms/ms scan). the acquired raw peak list data were generated by xcalibur (version . ) using default parameters and further analyzed by the label-free identification and quantitative algorithm using default parameters described by higgs et al [ ]. ms database searches were performed against the combined protein data set from international protein index and the non-redundant ncbi-nr human protein database, which totaled , protein records. carious data pro- cessing filters for protein identification were applied to keep only peptides with the xcorr score above . for singly charged peptides, . for doubly charged peptides, and . for triply charged peptides. these xcorr scores were set according to linear discriminant analysis similar to that described in dtaselect (version . ) to control false-positive rate at below % levels. for protein quantification, first, all extracted ion chro- matograms (xics) were aligned by retention time. each aligned peak were matched by precursor ion, charge state, fragment ions from ms/ms data, and retention time within a one-minute window. then, after align- ment, the area-under-the-curve (auc) for each indivi- dually aligned peak from each sample was measured, normalized, and compared for relative abundance–all as described in [ ]. here, a linear mixed model general- ized from individual anova (analysis of variance) was used to quantify protein intensities. in principle, the lin- ear mixed model considers three types of effects when deriving protein intensities based on weighted average of quantile-normalized peptide intensities: ) group effect, which refers to the fixed non-random effects caused by the experimental conditions or treatments that are being compared; ) sample effect, which refers to the random effects (including those arising from sam- ple preparations) from individual biological samples within a group; ) replicate effect, which refers to the random effects from replicate injections from the same sample preparation. peptidomics approach to searching novel alternative splicing isoform in proteomics data our peptidomics approach to identifying novel alterna- tive splicing isoform in proteomics data includes three steps (figure ): ) building an synthetic database of figure a peptidomics approach to identifying novel alternative splicing isoforms in proteomics data. zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of alternative splicing isoforms for proteomics experiments; ) identification, characterization of alternative splicing isoform using proteomics; and ) validation of alterna- tive splicing isoform. step : building a synthetic database of alternative splicing isoforms for proteomics experiments briefly, we developed the synthetic alternative splicing database (sasd) [ ] in three steps: )extracting infor- mation on gene structures of all genes in the human genome and incorporating the ipad database [ ], ) compiling artificial splicing transcripts, and ) translat- ing the artificial transcripts into alternative splicing peptides. in the first step, we use the biomart to extract infor- mation on all human genes in the ensemble [ ] from the homo sapiens genes dataset (grch .p ) in the ensembl genes database. we then extract informa- tion on each human gene’s position, name, exon/intron coordinates, exon phase, sequences, and annotation. in the second step, we generate artificial splicing tran- script (ast), which is an exhaustive compilation of two categories of peptides. the first category is the peptides translated from all single exons and introns, and the second category is the peptides that cover all theoreti- cally possible exon/intron junction regions of all genes in the human genome. the first category contains two types of alternative splicing: single exon(exon_nm) and single intron (intron_as). the second category contains four types of alternative splicing: intron-exon (i_e_as, left intron retention junction), exon-intron (e_i_as, right intron retention junction), neighboring exon-exon (e_e_nm, normal splicing junction) and non-neighboring exon-exon (e_e_as, exon skipping junction). in the third step, we directly use the phase to translate the sequence for the exons with the phase information in ensemble transcript. for the exons without the phase information in ensemble transcript, three open reading frames are used in translating into three peptides and the longest peptide crossing the splicing site is reserved as alternative splicing peptide for sasd. step : identification and characterization of alternative splicing isoform using proteomics the most popular three types of search algorithms are ) correlating acquired ms/ms with theoretical spec- trum, counts the number of peaks in common, such as: sequest [ ] and x!tandem [ ], ) modeling the extent of peptide fragmentation, then estimates the probability that an assignment is incorrect due specifi- cally to a random match, such as mascot [ ] and omssa [ ], and ) de novo sequencing such as lute- fisk[ ] and peaks[ ]. we first run omssa against sasd to identify peptide from ms data. then we per- form preliminary data analysis. last, we extract informa- tion about alternative splicing. omssa reports hits ranked by e-value. an e-value for a hit is a score that is the expected number of ran- dom hits from a search library to a given spectrum, such that the random hits have an equal or better score than the hit. for example, a hit with an e-value of . implies that one hit with a score equal to or better than the hit being scored would be expected at random from a sequence library search [ ]. the e-value is calculated to report the expected frequency of observing scores equivalent to or better than the one for the reported peptide if the results were to take place randomly. the lower the e-value is, the more significant the score for the identified peptide by the peptide search using sasd database is. one-sided wilcoxon signed-rank test is used to per- form the preliminary statistical analysis in order to iden- tify peptides with significant occurrence differences in the health and breast cancer samples. step : validation of alternative splicing isoform we present two kinds of methods to validate isoforms in proteomics data, which are ) literature curation of alternative splicing isoforms and ) cross-validation of multiple studies. first, we perform an extensive litera- ture curation to determine the constituents of alterna- tive splicing isoform. then, we validate results using independent proteomics datasets derived from other study. we believe that such an integrative systems approach is essential to development and validation of panel alternative splicing isoform that may withstand rigorous testing for the future steps. pathway analysis pathway analysis is performed using the following data- bases: integrated pathway analysis database (ipad) (http://bioinfo.hsc.unt.edu/ipad/) [ ]. results the breast cancer plasma samples with samples from women diagnosed with breast cancer and from healthy volunteer women as controls were searched by omssa[ ] against the sasd database. after omssa searching, preliminary statistical analysis, and alternative splicing, we identified the eight alternative splicing iso- form biomarkers using the peptidomics approach to searching novel alternative splicing isoform in the proteo- mics data (table ). the peptides with e-value greater than . were filtered out. the p-value was calculated by performing one-sided wilcoxon signed-rank test to examine the probability that the median difference between two groups of samples is greater than zero. the number of such peptide found in health (h) and cancer zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of http://bioinfo.hsc.unt.edu/ipad/ (c) samples are listed separately in the table. the tested peptides are more likely to exist in cancer samples than in healthy samples when p-value is small. bold text is the left part of the junction and italic text is the right part. splicing site is marked by ^ or (). ‘()’ means the splicing site is shared by the left region and right region. for example, the second peptide qtpkhiseslgaevdp dmswssslatpptlsstvli(g)llhssvk is a synthetic product of the enst in gene brca when its eighth, ninth, and tenth exons are skipped and its seventh exon is combined together with its eleventh exon. the glycine is the shared splicing site between the seventh exon and the eleventh exon. none of the eight peptides are reported to have ever been detected in the peptide atlas database, which con- tains a comprehensive catalogue of all peptides derived from published proteomics experiments. they are not found with normal splicing mechanism (table ). the first sequence is a single intron alternative spli- cing. it was observed in patient samples and healthy samples. triple play mode of the annotated thermo-finnegan lcq-deca ion-trap ms/ms spec- trum is shown in figure [ ]. the triple play mode includes a) primary mass spectrum; b) zoom scan mass spectrum; c) ms/ms mass spectrum and d) protein identification from ms/ms). due to space limit, the spectrums of other seven alternative splicing sequences are omitted. the eight peptides, identified by omssa, have signifi- cant difference in the numbers of hit samples between healthy women and breast cancers (pvalue < . , table ). a screen shot from the ucsc genome browser [ ] in the region of these peptides are also shown in figure . it shows that these peptide sequences are not found in est sequences and mrna from genbank and one refseq gene(figure ). table shows that our peptidomics approach has sig- nificant potential in enabling discovery of new types of high-quality alternative splicing isoform biomarkers. further literature search found that there are no any lit- erature reports for the eight peptides. moreover, a cross-validation found that identification of the eight peptides is supported by the independent study which contains samples collected from women diagnosed with breast cancer and from healthy volunteer woman who served as controls. pathway analysis shows the pathways linked with the eight alternative splicing isoforms are transcription fac- tor, signaling, cancer, and synthesis (table ). discussions we described the peptidomics approach to searching novel alternative splicing isoform in proteomics data, especially artificial alternative splicing and snp. we can use it to identify two types of common alternative spli- cing events: exon skipping and intron retention. exon skipping is an alternative splicing mechanism in which exon(s) are included or excluded from the final gene transcript leading to extended or shortened mrna var- iants. and intron retention is an event in which an intron is retained in the final transcript. other types of alternative splicing events such as alternative ’ splice site and ’ splice site are not included in our method but can be derived indirectly from the two basic types: exon skipping and intron retention. the current protein sequence databases used by tan- dem mass spectra search engines, for example ipi, uni- prot, and ncbi nr, are designed to be useful as possible to as many researchers as possible. as such, they are a less than ideal substrate for tandem mass spectra search. protein sequence databases typically represent only “full-length” protein sequences and attempt to collapse protein variants to a single “consensus” entry. tandem mass spectra search engines, however, chop up the pro- tein sequence using an in-silico enzymatic digestion (such as trypsin), so full-length proteins are not needed in order to identify experimentally observed peptides; and the currently available search engines require the experimental peptides’ sequences be explicitly present in the sequence database in order to identify them, so explicit sequence variants are very important. the sasd is in fact a complete peptide sequence database, which table novel alternative splicing isoform candidate biomarkers for breast cancer in plasma peptide sequence gene transcript mode type pvalue h c peptide atlas swggrpqrmgavpggvwsavlmggar erbb enst i intron_as . e- no qtpkhiseslgaevdpdmswssslatpptlsstvli(g)llhssvk brca enst e _e e_e_as . e- no slwlqsqphfccfwltvtfppplq^threlaqsshaqr ntrk enst i _e i_e_as . e- no wglllallppgaastq(a)vwtwmtr erbb enst e _e e_e_as . e- no lswnhvaraltltqslvssvtsgk ntrk enst i intron_as . e- no cq(g)epyhdirfnlmavvpdr bap enst e _e e_e_as . e- no qvlp^vgvlgppgqqapppypgphpagppviqqpttpmfvapppk pbrm enst e _e e_e_as . e- no dhlacw^dydlcitcyntknhdhk ep enst e _e e_e_as . e- no zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of figure triple play mode spectrum of swggrpqrmgavpggvwsavlmggar in patient c_ . figure ucsc genome browser screen shot of genomic region for the novel peptide. zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of table pathway analysis for the eight alternative splicing isoforms. pathwayid pathwayname molecule p transcription factor network brca ;ep foxm transcription factor network ep ;brca validated targets of c-myc transcriptional repression ep ;erbb hsa adherens junction erbb ;ep hsa pancreatic cancer erbb ;brca grb events in erbb signaling erbb downregulation of errb :erbb signaling erbb host interactions with influenza factors pbrm virus assembly and release pbrm fusion and uncoating of the influenza virion pbrm entry of influenza virion into host cell via endocytosis pbrm packaging of eight rna segments pbrm viral rnp complexes in the host cell nucleus pbrm vrna synthesis pbrm crna synthesis pbrm erbb receptor signaling network erbb influenza virus induced apoptosis pbrm fusion of the influenza virion to the host cell endosome pbrm release pbrm budding pbrm uncoating of the influenza virion pbrm vrnp assembly pbrm homologous recombination repair of replication-independent double-strand breaks brca traf -dependent irf activation pathway ep homologous recombination repair brca presynaptic phase of homologous dna pairing and strand exchange brca homologous dna pairing and strand exchange brca h_hifpathway hypoxia-inducible factor in the cardiovascular system ep h_pitx pathway multi-step regulation of transcription by pitx ep h_ppargpathway role of ppar-gamma coactivators in obesity and thermogenesis ep h_relapathway acetylation and deacetylation of rela in the nucleus ep hsa pathways in cancer erbb ;brca ;ep h_carm pathway transcription regulation by methyltransferase of carm ep h_melanocytepathway melanocyte development and pigmentation pathway ep h_pelp pathway pelp modulation of estrogen receptor activity ep h_vdrpathway control of gene expression by vitamin d receptor ep hsa prostate cancer erbb ;ep h_il pathway il- signal transduction ep viral messenger rna synthesis pbrm grb events in erbb signaling erbb fanconi anemia pathway brca double-strand break repair brca sema d induced cell migration and growth-cone collapse erbb h_g pathway cell cycle: g /m checkpoint ep h_mef dpathway role of mef d in t-cell apoptosis ep h_atrbrcapathway role of brca , brca and atr in cancer susceptibility brca h_her pathway role of erbb in signal transduction and oncology ep h_nthipathway nfkb activation by nontypeable hemophilus influenzae ep h_p hypoxiapathway hypoxia and p in the cardiovascular system ep h_tgfbpathway tgf beta signaling pathway ep zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of includes majority of all occurrence of alternative spli- cing. it also provides alternative splicing for each pep- tide, such as splicing mode, splicing type, splicing site, starting position, ending position, and peptide sequence. moreover, the current protein sequence databases and some alternative splicing database such as astd and eid are not ideal or enough for identifying alternative splicing isoform from tandem mass spectrometry. there are either no or very few isoform information in these databases. for example, astd only includes occurrences of intron isoforms and occurrences of exon isoforms. even for cassette exons event, the num- ber of occurrences is only [ ]. in contrast, the sasd database in our method includes , , alternative splicing peptides covering about , genes (ensembl gene ids), , transcripts (ensembl transcript ids), pathways, diseases, drugs, and organs. its comprehensive coverage means better sensitivity in identifying novel alternative splicing isoforms than the peppi. and its exclusive focus on alternative splicing can definitely increase the specificity of the identification of alternative splicing. alternative splicing isoform biomarkers are apparently important and can serve as an alternative to traditional bio- markers. we can use quantitative information such as p- value to determine the significance of the marker. we can also use the qualitative information such as: splicing type, splicing mode, peptide sequence etc. to further analyze the alternative splicing’s mechanism. we think that combina- tion of traditional biomarkers with the alternative splicing isoform biomarkers will definitely help us better under- stand the treatment, diagnosis, and prognosis of cancer. competing interests statement the authors declare that they have no competing interests. authors’ contributions rd and fz conceived the initial work and designed the method. fz and rd developed the alternative splicing database and method, and performed the computational analyses. mw provided experimental data. tm performed literature search. all authors are involved in the drafting and revisions of the manuscript. acknowledgements we thank hoosier oncology group for collecting breast cancer plasma samples. the proteomics study for biomarker discovery was supported by the national cancer institute clinical proteomics technology assessment for cancer program (u ca ). we also thank the support of bioinformatics program at university of north texas health science center. declarations the publication costs for this article were funded by the bioinformatics program in university of north texas health science center. this article has been published as part of bmc systems biology volume supplement , : selected articles from the international conference on intelligent biology and medicine (icibm ): systems biology. the full contents of the supplement are available online at http://www. biomedcentral.com/bmcsystbiol/supplements/ /s . authors’ details department of academic and institutional resources and technology, university of north texas health science center, fort worth, usa. department of forensic and investigative genetics, university of north texas health science center, fort worth, usa. department of biochemistry and molecular biology, iu school of medicine, indianapolis, in , usa. indiana center for systems biology and personalized medicine, indianapolis, in , usa. published: december references . national cancer institute dictionary. . . biomarkers and surrogate endpoints: preferred definitions and conceptual framework. clinical pharmacology and therapeutics , ( ): - . . hu x, zhang y, zhang a, li y, zhu z, shao z, zeng r, xu lx: comparative serum proteome analysis of human lymph node negative/positive invasive ductal carcinoma of the breast and benign breast disease controls via label-free semiquantitative shotgun technology. omics , ( ): - . . zeidan ba, cutress ri, murray n, coulton gr, hastie c, packham g, townsend pa: proteomic analysis of archival breast cancer serum. cancer genomics proteomics , ( ): - . . lebrecht a, boehm d, schmidt m, koelbl h, schwirz rl, grus fh: diagnosis of breast cancer by tear proteomic pattern. cancer genomics proteomics , ( ): - . . polyak k: breast cancer: origins and evolution. j clin invest , ( ): - . . roberts tk, eugenin ea, morgello s, clements je, zink mc, berman jw: prpc, the cellular isoform of the human prion protein, is a novel biomarker of hiv-associated neurocognitive impairment and mediates neuroinflammation. am j pathol , ( ): - . . brinkman bm: splice variants as cancer biomarkers. clinical biochemistry , ( ): - . . garcia-blanco ma, baraniak ap, lasda el: alternative splicing in disease and therapy. nat biotechnol , ( ): - . . kornblihtt ar, schor ie, allo m, dujardin g, petrillo e, munoz mj: alternative splicing: a pivotal step between eukaryotic transcription and translation. nature reviews molecular cell biology , ( ): - . . anensen n, oyan am, bourdon jc, kalland kh, bruserud o, gjertsen bt: a distinct p protein isoform signature reflects the onset of induction chemotherapy for acute myeloid leukemia. clin cancer res , ( ): - . . assender jw, gee jm, lewis i, ellis io, robertson jf, nicholson ri: protein kinase c isoform expression as a predictor of disease outcome on endocrine therapy in breast cancer. j clin pathol , ( ): - . table pathway analysis for the eight alternative splicing isoforms. (continued) shc events in erbb signaling erbb signaling events mediated by hdac class iii ep hsa homologous recombination brca retinoic acid receptors-mediated signaling ep sema d in semaphorin signaling erbb traf mediated irf activation ep zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of http://www.biomedcentral.com/bmcsystbiol/supplements/ /s http://www.biomedcentral.com/bmcsystbiol/supplements/ /s http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract . cressey r, wattananupong o, lertprasertsuke n, vinitketkumnuen u: alteration of protein expression pattern of vascular endothelial growth factor(vegf) from soluble to cell-associated isoform during tumourigenesis. bmc cancer , : . . di fede g, giaccone g, limido l, mangieri m, suardi s, puoti g, morbin m, mazzoleni g, ghetti b, tagliavini f: the epsilon isoform of - - protein is a component of the prion protein amyloid deposits of gerstmann- straussler-scheinker disease. j neuropathol exp neurol , ( ): - . . thanaraj ta, stamm s, clark f, riethoven jj, le texier v, muilu j: asd: the alternative splicing database. nucleic acids res , (database):d - . . matter n, herrlich p, konig h: signal-dependent regulation of splicing via phosphorylation of sam . nature , ( ): - . . lixia m, zhijian c, chao s, chaojiang g, congyi z: alternative splicing of breast cancer associated gene brca from breast cancer cell line. j biochem mol biol , ( ): - . . zhu z, xing s, cheng p, zeng f, lu g: modification of alternative splicing of bcl-x pre-mrna in bladder cancer cells. j huazhong univ sci technolog med sci , ( ): - . . ku th, hsu fr: mining colon cancer specific alternative splicing in est database. amia annu symp proc , . . ogawa t, shiga k, hashimoto s, kobayashi t, horii a, furukawa t: apaf- - alt, a novel alternative splicing form of apaf- , potentially causes impeded ability of undergoing dna damage-induced apoptosis in the lncap human prostate cancer cell line. biochem biophys res commun , ( ): - . . venables jp, klinck r, bramard a, inkel l, dufresne-martin g, koh c, gervais- bird j, lapointe e, froehlich u, durand m, et al: identification of alternative splicing markers for breast cancer. cancer res , ( ): - . . skotheim ri, nees m: alternative splicing in cancer: noise, functional, or systematic? int j biochem cell biol , ( - ): - . . rifai n, gillette ma, carr sa: protein biomarker discovery and validation: the long and uncertain path to clinical utility. nat biotechnol , ( ): - . . kulasingam v, diamandis ep: strategies for discovering novel cancer biomarkers through utilization of emerging technologies. nature clinical practice oncology , ( ): - . . adams md, kelley jm, gocayne jd, dubnick m, polymeropoulos mh, xiao h, merril cr, wu a, olde b, moreno rf, et al: complementary dna sequencing: expressed sequence tags and human genome project. science , ( ): - . . kapur k, xing y, ouyang z, wong wh: exon arrays provide accurate assessments of gene expression. genome biol , ( ):r . . johnson jm, castle j, garrett-engele p, kan z, loerch pm, armour cd, santos r, schadt ee, stoughton r, shoemaker dd: genome-wide survey of human alternative pre-mrna splicing with exon junction microarrays. science , ( ): - . . pan q, shai o, lee lj, frey bj, blencowe bj: deep surveying of alternative splicing complexity in the human transcriptome by high-throughput sequencing. nat genet , ( ): - . . pando mp, kotraiah v, mcgowan k, bracco l, einstein r: alternative isoform discrimination by the next generation of expression profiling microarrays. expert opin ther targets , ( ): - . . lee ps, shaw lb, choe lh, mehra a, hatzimanikatis v, lee kh: insights into the relation between mrna and protein expression patterns: ii. experimental observations in escherichia coli. biotechnol bioeng , ( ): - . . mehra a, lee kh, hatzimanikatis v: insights into the relation between mrna and protein expression patterns: i. theoretical considerations. biotechnol bioeng , ( ): - . . zhou a, zhang f, chen jy: peppi: a peptidomic database of human protein isoforms for proteomics experiments. bmc bioinformatics , , suppl : s . . higgs re, knierman md, gelfanova v, butler jp, hale je: comprehensive label-free method for the relative quantification of proteins from biological samples. j proteome res , ( ): - . . zhang f, drabier r: sasd: the synthetic alternative splicing database for identifying novel isoform from proteomics. bmc bioinformatics , , suppl : s . . zhang f, drabier r: ipad: the integrated pathway analysis database for systematic enrichment analysis. bmc bioinformatics , , suppl : s . . flicek p, ahmed i, amode mr, barrell d, beal k, brent s, carvalho-silva d, clapham p, coates g, fairley s, et al: ensembl . nucleic acids res , (database):d - . . eng jk, mccormack al, yates iii jr: an approach to correlate tandem mass spectral data of peptides with amino acid sequences in a protein database. journal of the american society for mass spectrometry , ( ): - . . bjornson rd, carriero nj, colangelo c, shifman m, cheung kh, miller pl, williams k: x!!tandem, an improved method for running x!tandem in parallel on collections of commodity computers. j proteome res , ( ): - . . koenig t, menze bh, kirchner m, monigatti f, parker kc, patterson t, steen jj, hamprecht fa, steen h: robust prediction of the mascot score for an improved quality assessment in mass spectrometric proteomics. j proteome res , ( ): - . . geer ly, markey sp, kowalak ja, wagner l, xu m, maynard dm, yang x, shi w, bryant sh: open mass spectrometry search algorithm. j proteome res , ( ): - . . johnson rs, taylor ja: searching sequence databases via de novo peptide sequencing by tandem mass spectrometry. mol biotechnol , ( ): - . . ma b, zhang k, hendrie c, liang c, li m, doherty-kirby a, lajoie g: peaks: powerful software for peptide de novo sequencing by tandem mass spectrometry. rapid commun mass spectrom , ( ): - . . zhang f, drabier r: ipad: the integrated pathway analysis database for systematic enrichment analysis. bmc bioinformatics , ( ). . wang m, witzmann fa, lyubimov av: proteomics in drug discovery and development. encyclopedia of drug metabolism and interactions john wiley & sons, inc; . . kent wj, sugnet cw, furey ts, roskin km, pringle th, zahler am, haussler d: the human genome browser at ucsc. genome res , ( ): - . doi: . / - - -s -s cite this article as: zhang et al.: novel alternative splicing isoform biomarkers identification from high-throughput plasma proteomics profiling of breast cancer. bmc systems biology (suppl ):s . submit your next manuscript to biomed central and take full advantage of: • convenient online submission • thorough peer review • no space constraints or color figure charges • immediate publication on acceptance • inclusion in pubmed, cas, scopus and google scholar • research which is freely available for redistribution submit your manuscript at www.biomedcentral.com/submit zhang et al. bmc systems biology , (suppl ):s http://www.biomedcentral.com/ - / /s /s page of http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract http://www.ncbi.nlm.nih.gov/pubmed/ ?dopt=abstract 	abstract 	background 	results 	conclusions 	introduction 	materials and methods 	human plasma samples 	protein identification and quantification 	peptidomics approach to searching novel alternative splicing isoform in proteomics data 	step : building a synthetic database of alternative splicing isoforms for proteomics experiments 	step : identification and characterization of alternative splicing isoform using proteomics 	step : validation of alternative splicing isoform 	pathway analysis 	results 	discussions 	competing interests statement 	authors’ contributions 	acknowledgements 	declarations 	authors' details 	references techniques and technology note landscape habitat suitability index software william d. dijak, united states department of agriculture forest service, northern research station, anheuser-busch natural resources building, columbia, mo , usa chadwick d. rittenhouse, department of fisheries and wildlife sciences, university of missouri, anheuser-busch natural resources building, columbia, mo , usa michael a. larson, department of fisheries and wildlife sciences, university of missouri, anheuser-busch natural resources building, columbia, mo , usa frank r. thompson, iii, united states department of agriculture forest service, northern research station, anheuser-busch natural resources building, columbia, mo , usa joshua j. millspaugh, department of fisheries and wildlife sciences, university of missouri, anheuser-busch natural resources building, columbia, mo , usa abstract habitat suitability index (hsi) models are traditionally used to evaluate habitat quality for wildlife at a local scale. rarely have such models incorporated spatial relationships of habitat components. we introduce landscape hsimodels, a new microsoft windowst (microsoft, redmond, wa)–based program that incorporates local habitat as well as landscape-scale attributes to evaluate habitats for species of wildlife. models for additional species can be constructed using the generic model option. at a landscape scale, attributes include edge effects, patch area, distance to resources, and habitat composition. a moving window approach is used to evaluate habitat composition and interspersion within areas typical of home ranges and territories or larger. the software and sample data are available free of charge from the united states forest service, northern research station at http://www.nrs.fs.fed.us/hsi/. ( journal of wildlife management ( ): – ; ) doi: . / - key words habitat, index, landscape, model, software, suitability. habitat suitability index (hsi) models have been used since the early s to determine wildlife habitat quality at the local scale based on field measurement data (u.s. fish and wildlife service , ). habitat suitability ranges from (low quality) to (high quality) based on wildlife– habitat relationships such as vegetation type (e.g., grass, forb, woody), structure (e.g., stem density), canopy cover, and the presence or absence of known foods. many existing hsi models are field-labor intensive and difficult to apply to large areas due to the fine-scale nature of required data. further, advances in our understanding of wildlife–habitat relationships, especially in regard to area- or edge-sensitivity and interspersion of life requisites (morrison et al. ), necessitates incorporation of spatial requirements in hsi models. we developed landscape hsimodels software to provide wildlife biologists and managers a tool to evaluate wildlife habitat quality across large landscapes. habitat suitability is basically evaluated at scales: the pixel and the landscape surrounding the pixel. the landscape hsimodels software allows the user to incorporate landscape-scale habitat components such as area sensitivity, edge effects, distance to specific resources, and habitat composition within areas typical of animal home ranges or territories or larger. we accomplish this through the use of a moving window that systematically processes portions of the landscape representing home ranges or territories to determine resources immediately accessible to an individual. this moving window moves across the landscape until the entire landscape has been processed. the size of the moving window can be changed by altering its dimensions or by varying the resolution of the input data; however, users should keep in mind ecological processes (e.g., edge sensitivity) or species home-range size when selecting the appropriate resolution. we combined landscape information with existing hsi models or through the creation of new hsi models to build models available in the software. users can evaluate habitat quality at larger geographic extents than was typically done in the past. for example, the software may be used to evaluate maps of alternative forest management practices on wildlife, assesses habitat conditions prior to management activities, or identify focal areas for conservation or management (fig. ). habitat suitability index models are available in the software for wildlife species. we developed hsi models through literature review of studies using empirical data, through adaptation of existing hsi models, and from expert opinion. species models within the software include oven- bird (seiurus aurocapillus), prairie warbler (dendroica dis- color), henslow’s sparrow (ammodramus henslowii), hooded warbler (wilsonia citrina), cerulean warbler (dendroica cerulea), yellow-breasted chat (icteria virens), pine warbler (dendroica pinus), worm-eating warbler (helmitheros vermi- vorus), wood thrush (hylocichla mustelina), wild turkey (melagris gallopavo), ruffed grouse (bonasa umbellus), american woodcock (scolopax minor), northern bobwhite (colinus virginianus), gray squirrel (sciurus carolinensis), black bear (ursus americanus), bobcat (lynx rufus), red bat e-mail: wdijak @fs.fed.us present address: minnesota department of natural resources, forest wildlife populations and research group, east highway , grand rapids, mn , usa the journal of wildlife management � ( ) (lasiurus borealis), northern long-eared bat (myotis septen- trionalis), indiana bat (myotis sodalis), red backed salaman- der (plethodon cinereus), and timber rattlesnake (crotalus horridus). the user can modify the assumed habitat relationships for each species or build models for new species using the functions in the software. each species model includes suitability indices that represent habitat components and spatial relationships. input data layers for suitability indices include tree age, dominant tree species, ecological land type, land cover, and fire history. input and output data formats are american standard code for information interchange (ascii) rasters, which may be created in arcview .x by exporting a data source, in arcgis using arctoolbox, and in arcinfo by issuing the gridascii command. the ascii rasters created in other geographic information system software packages need to follow the environmental services research institute format for header lines. the software guides users through the calculation of each suitability index (si) and finally through the overall hsi by combining the previously calculated sis (fig. ). expressions for si and hsi calculations may include numerical constants and mathematical (e.g., , exponents, natural logarithm), comparison (e.g., ., ,, �), and logical (e.g., if, and, or) operators (larson et al. ). measures of many habitat components (e.g., vegetation structure) are not available at a landscape scale. so, the models rely on ecological relationships and inference from ecological attributes that may be available and mapped at landscape scales, such as ecological land type or forest size- class. therefore, knowledge of how forest structure is related to tree species, forest age, and ecological land type is required for many of the models. much of that information can be garnered from the united states forest service, forest inventory and analysis data (miles et al. ) or previously conducted studies. users can customize models to adapt them to their data values, geographical area of interest, and understanding of species–habitat relationships. these modifications can be saved in a file that can later be opened to restore the user’s own parameters. a generic model also is provided so that users can select individual sis from different species to create models for new species. the minimum computer system recommended is a personal computer with a . gigahertz (ghz) processor and megabytes of random access memory (ram). we also recommend using a -inch or larger monitor. computers with faster processors and more ram will reduce model processing time. a computer with the above configuration was successfully used to process a , row by , column landscape. a landscape with , rows and , columns was modeled on a computer with a . ghz processor and gigabytes of ram. the maximum size of a landscape that can be processed will vary from model to model based on the number of individual suitability indices incorporated into the model and the complexity of the calculations that need to be processed within the model. some models using large moving windows on large land- scapes take several hours to complete. microsoft windowst (microsoft, redmond, wa) does not allow any single computer application to access . gigabytes of processor memory so the software will not take advantage of computers figure . cerulean warbler suitability index map for the patoka management unit of the hoosier national forest in south-central indiana, usa. habitat values are based on forest patch size and range from (light grey, poor-quality habitat) to (black, high-quality habitat). dijak et al. � landscape evaluation software with . gigabytes of ram. this limitation is the ultimate limiting factor in the size of the landscape that can be successfully processed. management implications managers need to evaluate current and future habitat conditions for various species when developing management plans. through the use of spatially explicit forest-simu- lations models such as landis (mladenoff et al. ) managers can determine temporal cumulative effects of management decisions on forested landscapes. applying landscape hsimodels to forest-simulation model output provides managers a tool to contrast the effects of forest management practices with current habitat quality for a variety of wildlife species (shifley et al. ). the software, which includes sample data, is available free of charge at: http://www.nrs.fs.fed.us/hsi/. acknowledgments we wish to thank n. sullivan for his contribution in the development of mast equations. we thank m. w. hubbard, d. a. hamilton, s. k. amelon, t. a. g. rittenhouse, c. basile, j. perez, the hoosier national forest planning staff, and forest planning participants for their contributions in species model development. literature cited larson, m. a., w. d. dijak, f. r. thompson, iii, and j. j. millspaugh. . landscape-level habitat suitability models for wildlife species in southern missouri. u.s. department of agriculture forest service, north central research station, general technical report nc- , st. paul, minnesota, usa. miles, d. p., g. j. brand, c. l. alerich, l. f. bednar, s. w. woudenberg, j. f. glover, and e. n. ezell. . the forest inventory and analysis database description and users manual version . . u.s. department of agriculture forest service, north central research station, general technical report nc- , st. paul, minnesota, usa. mladenoff, d. j., g. e. host, j. boeder, and t. r. crow. . landis: a spatial model of forest landscape disturbance, succession, and management. pages – in m. f. goodchild, l. t. parks, b. o. parks, c. johnston, d. maidment, m. crane, and s. glendining, editors. gis and envi- ronmental modeling. gis world books, fort collins, colorado, usa. morrison, m. l., b. g. marcot, and r. w. mannan. . wildlife– habitat relationships. second edition. university of wisconsin press, madison, usa. shifley, s. r., f. r. thompson, iii, w. d. dijak, m. a. larson, and j. j. millspaugh. . simulated effects of forest management alternatives on landscape structure and habitat suitability in the midwestern united states. forest ecology and management : – . u.s. fish and wildlife service. . habitat evaluation procedures (hep). u.s department of interior fish and wildlife service, division ecological services manual , washington, d.c., usa. u.s. fish and wildlife service. . standards for the development of habitat suitability index models for use in the habitat evaluation procedure. u.s. department of interior fish and wildlife service, division ecological services manual , washington, d.c., usa. associate editor: white. figure . cerulean warbler final habitat suitability index map for the patoka management unit of the hoosier national forest in south-central indiana, usa. final map is based on the combination of suitability index maps. habitat quality ranges from (light grey, poor-quality habitat) to (black, high-quality habitat). the journal of wildlife management � ( ) computer science and information technology ( ): - , http://www.hrpub.org doi: . /csit. . a discovery of the relevance of eastern four-valued (catuskoti) logic to define modular transformations when there are multiple ways of representing the same modular transformation madanayake r s*, dias g k a, kodikara n d school of computing, university of colombo, sri lanka copyright © by authors, all rights reserved. authors agree that this article remains permanently open access under the terms of the creative commons attribution license . international license abstract there were two methods of doing modular transformations from entity relationship diagrams to class diagrams according to international researchers. in order to establish which method would best suit software engineers, we conducted a survey by giving a group of students in the computer science field, whom we considered potential future software engineers. the results we got were valid, but did not match those of any of the international researchers. we found that this situation could only be explained using eastern four-valued logic, also known by such names as catuskoti and tetralemma. keywords catuskoti, four-valued logic, tetralemma, modular transformations, erd to class diagram, boolean logic, both affirmation and negation, neither affirmation nor negation . the initial motivation for this experiment the research question for the mphil which the first author of this paper was involved in was “how could the different types of transformations of software engineering models be combined to expedite the software engineering process?”. for a solution to this question, we first tried to get a group of students to do an exercise to find out which of the two methods given in the international publications they will use to transform an erd diagram into a class diagram. . objectives of the survey/experiment of the student assignment the original objective of our experimental survey was not to invalidate or undermine aristotelian two-valued logic, but to evaluate which of the methods of erd to class diagram (modular transformation) rules (already used in the published literature of the international researchers as given in [ ] and [ ]), would be used by our students for the relevant modular transformation. the conclusion which we actually wanted to arrive at was that one of the modular transformation methods (out of the two suggested by the authors of [ ] and by the authors of [ ]), was correct, in order to use it as a benchmark or a basis in order to program a software which could convert a drawn erd diagram into its compatible class diagram. since we originally theorized that the least logical method (eg. [ ]) would be discarded by the students, who would choose the other method (eg. [ ]), we first classified the [ ] method as true, and the [ ] method as falls, as a sort of hypothesis (not for a research degree, but for the more limited purpose of publishing in a conference or in a / journal). what happened during this experiment was actually a significant deviation from our original objective(s), and we appreciate it if our peer reviewers could evaluate this paper in that context. . introduction this particular experiment is part of a main research which focuses on modular transformations as well as on ontologies [ ], [ ], [ ], [ ], [www ]. modular transformations have today become an interesting research topic. one reason for this is that there are many legacy systems designed using legacy system development models, and in order to make them compatible with modern object oriented models to enable effective maintenance and modification of those systems, it has become a requirement to transform those existing legacy design models into object oriented models. two main output diagrams obtained from a discovery of the relevance of eastern four-valued (catuskoti) logic to define modular transformations when there are multiple ways of representing the same modular transformation the same modular transformation were the output diagrams obtained by two groups of international researchers (eg. tran et.al. [ ] and fries [ ], [ ]) by transforming entity relationship diagrams (erds) to class diagrams. to see which of these two output diagrams would be more logical, we assigned a group of computer science students whom we considered prospective software engineers, the task of doing the same modular transformation (erd to class diagram) which was done by both tran et.al. (whose was assumed as the “false” output) and fries (whose was assumed as the “true” output) above. to our surprise we found that most of the students who had done the modular transformation had obtained an entirely new output diagram which we also found was logically correct. to explain this surprising result, we found that it is more appropriate to use eastern four-valued logic or catuskoti [www ], [www ]. . background scientific reasoning in the western cultural context has been for a long time based on two valued logic, which is supposed to have arisen in ancient greece due to philosophers such as aristotle. this is also known as boolean logic, in relation to the two states “ ” and “ ” (comparable to the two values “true” and “false”) with respect to boolean algebra. according to [www ] the laws of this aristotelian logic are as follows; . the law of identity - a is a everything is the same as itself; or a statement cannot not remain the same and change its truth value. . the law of non-contradiction - not (a and not a) – or nothing can both be a and not-a nothing can both exist and not exist at the same time and in the same respect; or no statement is both true and false. . the law of excluded middle - either (a or not a) something either exists or does not exist; or every statement is either true or false. two of the aristotelian laws of thought - the “the law of the excluded middle” which simplistically means than everything must either be a or not a; and the law of non-contradiction which means that nothing can both be a and not-a [ ] (which really signifies the boolean two valued logic) – were (and in many respects, still are) fundamental in formal logic which is also the study of inferences and so are important in both modern science and modern mathematics [www ]. . are there drawbacks in applying aristotelian logic to experimental results? however, is boolean / aristotelian two valued logic really able to help in the analysis of all the different kinds of objects, phenomena and their interrelationships which exist in the world? the purpose of researching quotes of academic oriented personalities on the drawbacks / limits of aristotelian logic, was for us to better understand why the majority of our students arrived at an object diagram which was previously unforeseen by us, the researchers who planned and carried out this experiment, initially expecting quite a different result from that which we got. to give hypothetical example (about the limitations of aristotelian logic) from our own personal understanding; in the beginning of electronic computers, the two states of the vacuum tubes which represented data units inside computers were perfectly and adequately represented by boolean logic. because of this, logic gates could be designed and the used to develop computer programming logic further. however, with the development of the transistor, there was the introduction of three electronic states (active mode, saturation mode and cut-off mode). out of these, only the saturation mode and cut-off mode are used for digital computers while the saturation mode seems to have been totally ignored (it seems to be totally outside the boolean “universe” of computer machine code.). while this is our own personal understanding, it seems that the fact that there are limits to the capabilities of boolean logic has (to varying extents) been validated by the observations of other philosophers, authors and researchers. as further examples to illustrate this point, a reading of the quotes of just three of these eminent personalities (who found it difficult to agree to a universal application of aristotelian logic) would serve to highlight this point with respect to the weaknesses of aristotelian / boolean two-valued logic in particular, and of aristotelian logic in general. those relevant quotes are as given below; quote : as the author l. zadeh stated in a publication in that, “in particular, treating truth as a linguistic variable with values such as true, very true, completely true, not very true, untrue, etc., leads to what is called fuzzy logic. by providing a basis for approximate reasoning, that is, a mode of reasoning which is not exact nor very inexact, such logic may offer a more realistic framework for human reasoning than the traditional two-valued logic. [www ]” the limits of boolean logic became especially evident in this case where another type of logic called fuzzy logic was needed [www ] in knowledge areas from control theory to artificial intelligence. fuzzy logic was mainly needed due to the weaknesses of the law of the excluded middle, but in this particular case, it was nothing but re-including the (formerly excluded) “middle”. there was no thinking out of the box, thinking of other “dimensions” of values, or going computer science and information technology ( ): - , perpendicular to the “middle” between the two extremes. it could be compared to allowing yourself to look at the total range of the box including its middle as well, whereas, in boolean / aristotelian logic (law of the excluded middle) you were firmly stuck to the sides of the box unable to either go in or out. quote : according to robert maynard pirsig an american writer and philosopher, and the author of philosophical novels, also outlines a certain different failure of the simple two-values logic in his novel “zen and the art of motorcycle maintenance”. the authors of this paper view this quote as an attempt to try to look beyond the box, (although no specific experiment was mentioned in this webpage resource from where this quote was extracted from). he highlights it in this way with respect to a “state” referred to as “mu” in the following way : “for example, it's stated over and over again that computer circuits exhibit only two states, a voltage for "one" and a voltage for "zero." that's silly! any computer-electronics technician knows otherwise. try to find a voltage representing one or zero when the power is off! the circuits are in a mu state.” [www ] quote : dr. nalin de silva, (who is a retired professor and was the former dean of the faculty of science kelaniya / vidyalankara university of sri lanka) who has a phd in theoretical cosmology (university of sussex, uk.) had this to say regarding the failure of aristotelian / boolean logic: “however, the logic that is abstracted from aristotelian-newtonian experiences is not capable of dealing with change in general and motion in particular. it is demonstrated by the famous zeno’s paradox that deals with an arrow in motion. the aristotelian logic is faced with contradictions when it is employed to describe motion and one would end up by showing that motion is impossible! the calculus of newton and leibniz, though their approaches were not the same, tried to get over this difficulty using infinitesimals intuitively without formally defining them. however, infinitesimals were not liked by the western mathematicians and philosophers and there were objections to these "ghosts" by people such as berkeley. euler, one of the greatest western mathematicians with an intuition that surpassed most of the others freely used infinitesimals in his formulation of mathematical analysis. however, as the western mathematicians did not like these infinitesimals that according to berkeley were neither finite nor not finite, later mathematicians dedekind, cantor and cauchy "exorcised" infinitesimals from mathematical analysis and introduced what is known as the epsilon - delta definition of limit, which is based on aristotelian logic. the calculus that tried to deviate from aristotelian logic at the beginning was brought back to an "arithmetical" definition based on that logic in the nineteenth century. it is interesting to note that something similar is happening in quantum physics. bohr (and heisenberg) who tried to deviate from the classical physics world view in the thirties created what is known as the copenhagen interpretation of quantum physics. copenhagen interpretation was obviously not in agreement with aristotelian logic and the tendency at present is to formulate a new interpretation based on aristotelian logic and doing away with heisenberg’s uncertainty principle.” [www ] . methodology originally, the purpose of our research was not to find any incompatibilities with aristotelian / boolean logic, but to see whether the method used by tran et.al. or the method used by fries would be more feasible or more realistic for doing modular transformations from erds to class diagrams (or in other to see words which of the two considered methods would be more appropriate to transform an erd diagram to a class diagram). to achieve the above, what we decided to find out was, whether a group of prospective future software engineers (eg.: computer science degree students) would use tran’s method or fries’ method to do the modular transformation from erd to class diagram and to see whether all students would follow the same method or else, if some students would select tran’s method and if others would select fries’ method to do their assignment, to reach a conclusion based on which method would be followed by the majority of the students. the following was the input diagram or the source diagram which was the legacy erd diagram that was given to the students to convert into a class diagram (figure ). this source for the case study was obtained by us from the relevant research paper by fries [ ], while fries had apparently got it from a literature attributed to yourdon known as “modern structured analysis” [ ]. a discovery of the relevance of eastern four-valued (catuskoti) logic to define modular transformations when there are multiple ways of representing the same modular transformation figure . the input / source erd diagram given to the students to be transformed into a class diagram [ ], [ ] the relevant experiment was conducted as an assignment where a batch of students of the science faculty of the university of colombo were required to transform the given erd diagram (figure ) encountered in structured systems analysis and design (ssadm) which is a legacy design methodology, into a class diagram found in unified modelling language (uml). please note that the following output diagrams are all from the published research papers of tran and fries [ ], [ ]. this is the reason that figure appears to be different from figures and , since figure is from tran [ ], figure is from fries [ ], and figure is just a modified version of figure , which was done to give a generalized form of the common output diagram produced by the majority (or ) of our students. there are two major output class diagrams when transforming an erd diagram to a class diagram using the erd of the hoosier-burger case study as the input diagram, that of fries (figure ) and that of tran et. al (figure ). our expectation was that the students would transform the input diagram into the output diagram produced by fries, which we considered as the true / yes boolean value – since fries had already evaluated trans’ paper which can before and fries seemed to us to have “corrected” some of the rules used previously by tran [ ], such as disposing of the transformation of relationships between erd entities into methods in the compatible class diagram (which was one of the rules followed by tran), while the output diagram produced by tran we considered as the false / no boolean value since fries seemed to have already studied trans’ paper, an declared some of the transformation rules used by tran to be “invalid”. the assignment was given as part of their subject module called visual programming technologies subject as % of the marks for the project (which maps to % of the marks for the entire visual programming technologies subject module). they were not taught about either the tran’s or the fries’ transformation methods in order to prevent experimental bias. please be kind enough to note that although we the researchers have been researching publications dealing with modular transformations from erds to class diagrams since the beginning of the year (for well over four years!) we could still only find the two modular transformation methods (from erds to class diagrams) computer science and information technology ( ): - , used by tran and by fries, in the international research arena. so, to date, we have found only the algorithmic methods used by tran and by fries, in the international research arena, for this particular modular transformation. if there are three or more other methods to do the modular transformations from erds to class diagrams, then we obviously have not found them, which is very unfortunate (as well as highly improbable) from the point of view of this research paper. the output class diagram generated using the fries’ method is shown below (figure ). figure . the output class diagram generated by the method followed by fries, t. p. given here as fries’ method [ ] the output class diagram generated using the tran et.al.s’ method is shown below (figure ). figure . the output class diagram generated by the method followed by tran, t.n., khan, k.m., lan, y.c. given here as tran’s method or tran et. al.’s method [ ] an example of the output achieved by the majority of students is shown below (figure ) which could have been achieved neither by following fries’ method nor by following tran’s method. a discovery of the relevance of eastern four-valued (catuskoti) logic to define modular transformations when there are multiple ways of representing the same modular transformation figure . an example of the output class diagram generated by of the students the actual method, and / or the rules of the transformation which these students used, could not be extracted from the students at this point (since they were not asked or taught to follow any rules or a method in this particular modular transformation), which should be looked at as a future work in this ongoing research project. please note that it is the final output diagram generated by of the students which we considered as most important for this particular research paper rather than the exact rules of the method which the students followed. please also note that the “recipe” class at the bottom of the above diagram (figure ) is actually out of the main diagram structure, and only connects to the main structure by a dashed line (signifying the case of a “weak” relationship between the other six classes). this is certainly not the case in either the output generated by fries’ method (figure ) or that generated by tran’s method (figure ). in other words, the students seemed to have generated a totally unique output which neither fries’ nor by tran’s research groups could even seem to be able to visualize – a totally out of the box solution for a particular case study within this limited domain of erd to class diagram modular transformations. . results we got the following results from this experiment with regard to the number of assignments which showed which method was followed. ) fries' method = assignments ) trans et. al.’s method = assignments ) both methods (hybrid) = assignment ) neither method = assignments ) neither method (but specific rd method) = assignments from the results it could be seen that many students ( of them) got neither output diagram as their result. although of those transformations were wrong since some of the entities in the erd were not represented as classes in the relevant class diagram, making the number of classes less than the number of respective entities, we filtered them out, but still we had assignments that were correct and had received a specific result which neither followed trans’ nor fries’ method. it is also significant that at least one student achieved a “hybrid” result leading one to come to the conclusion that both trans’ and fries’ methods could be merged to give a result on some occasions (specifically depending on the practical system under study). in this particular case the number of classes ( classes) was not violated. the results (once we filtered out the results which we considered wrong), are given in the table (table ) below. table . representation of the final student assignment outputs which the authors considered as correct according to the uml rules method followed by the student(s) to convert an erd diagram into a class diagram number of assignment outputs in each case fries' method trans et. al.’s method both methods (hybrid) neither method (but specific rd method) in the chart given on the next page, the first (left most) column maps to the yes/true boolean value, and the middle left column on this chart maps to the no/false boolean value (but does not have any data since none of the students generated that output diagram). the two columns to the right do not map to any of the boolean values, and seem to be outside of aristotelian / boolean logic with respect to this particular experiment. kindly go through the chart on the next page (page ). computer science and information technology ( ): - , (the columns extreme left “true”, middle left “false”, middle right “both true and false”, extreme right “neither true nor false”) chart . the same results in table are given in a more graphic . results & discussion from this data it can be seen that the large majority of students got a specific unique output that was generated by neither tran nor fries who were the international researchers (research groups), whose work we considered as the basis for this particular experiment. in our domain of transformation of erds to class diagrams, we considered the output diagrams of tran and fries as the only two correct possibilities. since fries was the later researcher who had also criticized trans’ work, we identified fries’ output as true (affirmation) and trans’ output as false (negation). however, the existence of a rd possibility which was compatible with neither the outputs of tran nor the outputs of fries was a totally unexpected and surprising result. generally a proposition in boolean / aristotelian or classical logic or two valued logic (which is of western origin) is that there are only two possible values - affirmation (true or x) or negation (false or x) leading to conclude that only the methods of fries and that of tran were possible in the case of this particular experiment. while analyzing our unique result, we came to the conclusion that the simple true / false logic (where affirmation or negation were the only two possibilities) was not enough to make sense of the total output of this particular experiment. we found it more appropriate to use eastern (catuskoti or tetralemma) logic [www ] where a proposition could have four values as possible answers. those possible values found in tetralemma logic (catuskoti or four valued logic) are as shown below [www ] ; x = affirmation (true) x = negation (false) x x = both affirmation and negation (both true and false) (x x) = neither affirmation nor negation (neither true nor false) the numbers of students who followed the different a discovery of the relevance of eastern four-valued (catuskoti) logic to define modular transformations when there are multiple ways of representing the same modular transformation methods to do the assignment (as given in the results section above) are shown against the appropriate values of tetralemma logic in the table below (table ). table . the results obtained by us are shown against the relevant values of tetralemma logic in this table tetralemma logical value (symbolic) meaning of the logical value relevant practical / experimental situation (the method followed to do the assignment) number of relevant assignment outputs in each case x affirmation fries' method x negation trans et. al.’s method x x both affirmation and negation both methods (hybrid method) (x x) neither affirmation nor negation neither method (but specific rd method) in our situation, of the outputs (specifically the outputs of those which we identified as correct), followed neither the fries’ method nor the trans’ method. therefore it corresponds to the (x x) (neither) affirmation nor negation possibility found in tetralemma logic (refer table , above). as a reason for the fact that none of our students used tran’s method (since none of them failed to get that output) we hypothesize the lack of consideration by the students that the relationships in an erd diagram could be transformed into the methods in a class diagram could have prevented students from even considering that such a method was possible; specially since the methods of erd to class diagram conversion was not taught to the students before the conversion. however, at the current moment, this remains purely hypothetical since the curriculum did not provide enough time schedule to interview the students and find out the reasons. . conclusions and future work according to the results of this experiment, the authors / researchers could conclude that at least in this experiment of this type, the use of aristotelian two-fold logic (true / false) maybe inadequate to evaluate the validity of a hypotheses. if the eastern four-fold logic was applied, a result more compatible to reality could be obtained. in that case, a hypotheses could be true, false, both true and false; or, neither true nor false. as future work, one could also further analyze the results of the following as given in the results section above.  neither method = assignments  neither method (but specific rd method) = assignments although of the assignments in “ )” were identified as incorrect, leaving us with only assignments (since – = ), as shown in “ )”, we could only come to this correct / incorrect assumption due to the data provided to us by the other cited authors as given in [ ] and [ ]. as to whether these assignments are absolutely incorrect has to be verified by re-examining the actual business environment of the “hoosier-burger” case study from which the data for the original erd diagram was extracted. if future authors could find this situation today, they could check the validity of the incorrect assignments. however, the authors wish to state that this anomaly does not have any bearing on the conclusions regarding the requirement of eastern four-fold logic, since our input data did not rely on the actual business environment of the “hoosier-burger” case study, but on testing the feasibility of transforming an already created entity relationship diagram (erd) into a class diagram. another possible future work would be to find out the exact rules and / or logic the students used to convert generate the unique output class diagram that of them generated (as shown in figure of page , above). finally, a comparison of tetralemma logic with fuzzy logic to further highlight the significance of the conclusions arrived at in this research we wish to give the readers a brief comparison of tetralemma with fuzzy logic to help justify the fact that tetralemma logic is more applicable in the case of the results of this experiment than fuzzy logic. fuzzy logic deals more in bridging the gap between the two extremes of true and false, and has more to do with dealing with the restrictions of the law of the excluded middle. however (fuzzy logic) still remains within the confines of the two extremes of logic, or within or limited by) the range. tetralemma logic on the other hand forces one to think beyond the confines of two extremes within the same range of values. tetralemma logic deals wiith the un-thought, unanticipated, unexpected “other-dimensional” type realities than what one could assume would be the possible range of values (for a particular variable), before an experiment is done. tetralemma forces one to think of the possibility of other ranges of values or dimension which the researchers had not thought before doing the experiment. in this case we conclude that tetralemma is more applicable, since the unexpected results we received were definitely not ones which we could place either on one of the extremes, or between the two extremes. references web references [www ] http://en.wikipedia.org/wiki/ontology_(information_scienc e) last accessed - - [www ] http://en.wikipedia.org/wiki/tetralemma last accessed - - [www ] http://en.wikipedia.org/wiki/catuṣkoṭi last computer science and information technology ( ): - , accessed - - [www ] http://oregonstate.edu/instruct/phl /modules/philosophers /aristotle/aristotle_laws_of_thought.html last accessed - - [www ] https://en.wikipedia.org/wiki/mathematical_logic# th_cen tury last accessed - - [www ] http://www.sciencedirect.com/science/article/pii/ last accessed - - [www ] https://en.wikipedia.org/wiki/mu_(negative) last accessed - - [www ] http://www.island.lk/ / / /midwee .html last accessed - - [www ] https://en.wikipedia.org/wiki/logic#the_study_of_logic - - [ ] tilakaratna, p., rajapakse, j. ( ), forward engineering the object oriented analysis and design, - ., proceedings of the th malaysian conference in software engineering. [ ] benevides, a.b. and guizzardi, g.,( ) a model based tool for conceptual modeling and domain ontology engineering in ontouml, (lecture notes in business information processing, vol. ), p. – . isbn - - - - . [ ] tilakaratna p., and rajapakse j., ( ) “ontological framework for object-oriented analysis and design,” - . in american journal of engineering and applied sciences ( ). [ ] li x., and parsons j., ( ) “ontological semantics for the use of uml in conceptual modeling,” in: proceedings of tutorials, posters, panels and industrial contributions at the th international conference on conceptual modeling - er auckland, new zealand. crpit, . grundy, j., hartmann, s., laender, a. h. f., maciaszek, l. and roddick, j. f., eds. acs. - . [ ] tran, t.n., khan, k.m., lan, y.c. ( ), a framework for transforming artifacts from data flow diagrams to uml. in: proceedings of the iasted international conference on software engineering, innsbruck, austria. [ ] fries, t.p. ( ), a framework for transforming structured analysis and design artifacts to uml., proceeding of the th annual conference on design of communication, usa. [ ] jilani a.a.a., usman m., nadeem a., malik z.i. and halim z., ( ), “comparative study on dfd to uml diagrams transformations,” world of computer science and information technology journal (wcsit), vol. , pp. - . [ ] kodish, b.i, “dare to inquire: sanity and survival for the st century and beyond”, pp . [ ] yourdon, e., modern structured analysis. prentice hall (india) pvt. ltd., . short communications the wilson journal of ornithology ( ): - , incidence of nest material kleptoparasitism involving cerulean warblers kelly c. jones,l kirk l. roth,' kamal paul b. hamel and carl g. smith, abstract.-we document observations of in- terspecific stealing of nesting material involving ce- rulean warblers (dendroica cerulea), red-eyed vireos (vireo olivaceus), blue-gray gnatcatchers (polioptila caerulea), northern parulas (parula americana), black-throated green warblers (d. virens), american redstarts (setophaga ruticilla), and orchard orioles (zcterus spurius) that occurred during studies of ce- rulean warbler breeding biology. these incidents in- volved a variety of combinations of nest owner and nest material thief suggesting that each of these species is both a perpetrator and recipient of this behavior in our study areas. kleptoparasitic incidents occurred at all stages of the nesting cycle from nest-building through post-fledging. two possible motivations for this behavior are related to saving time in finding nest materials and collecting this material for nest construc- tion. received november . accepted septem- also been observed gathering nesting materi- als from vireo nests (species unspecified) in new jersey (dater ). cerulean warbler breeding biology studies have focused on nest observation beginning in in tennessee and arkansas, and in in southern indiana. this paper documents interspecific contests for nesting material in- volving cerulean warblers that were observed during stages of the nesting cycle (table ). we present three detailed accounts of nest ma- terial kleptoparasitism involving cerulean warblers as both victim and perpetrator. own- er defense usually started at the point at which the owner discovered the robber at its nest. ber . observations the cerulean warbler (dendroica cerulea) is a canopy-nesting bird of eastern deciduous forests. nests of this species are composed mainly of bark fiber, fine grass stems, weed stalks, hairs, spider webs, grapevine bark, li- chen, and moss (e-g., bent , ehrlich et al. , oliarnyk and robertson , harnel ). nests are typically on horizontal branches and are concealed from above by nest-tree and/or vine foliage (bent , ha- me , roth ). few published studies have documented interspecific nest kleptopar- asitism involving cerulean warblers. hamel ( ) noted in the mississippi alluvial val- ley that cerulean warblers and american redstarts (setophaga ruticilla), and cerulean warblers and blue-gray gnatcatchers (poliop- tila caerulea) interacted aggressively over nesting materials. cerulean warblers have ' department of biology, ball state university, muncie, in , usa. usda, forest service, center for bottomland hardwoods research, i? . box , stoneville, ms , usa. the following two accounts document nest material kleptoparasitism with the cerulean warbler as victim. the first incident occurred on may between and hrs cst when a red-eyed vireo (vireo olivaceus) was observed taking material from a cerulean warbler nest in a grove of black walnut (jug- lans nigra) at big oaks national wildlife ref- uge ( " ' n, " ' w) near madison, in- diana. the vireo landed on the nest branch within a meter of the nest, then flew to the nest and removed a piece of the outer cup when the cerulean warbler was not present. the nest material was sufficiently large to be seen in the vireo's beak as it flew to a more densely wooded area. no bird visited the nest for a period of several minutes until a vireo again landed on the nest branch. the vireo hopped toward the nest when the male ceru- lean warbler chased it into the heavily wood- ed area. the female cerulean warbler then flew to the nest and sat in it. the female ce- rulean warbler was first observed building the nest on may and incubating on may. she was last observed incubating on may and corresponding author; e-mail: kislam@bsu.edu the nest had failed on may. the walnut the wilson journal of ornithology vol. , no. , june table . nest material kleptoparasitism involving cerulean warblers recorded during cerulean warbler breeding biology studies in tennessee, arkansas, new york, and indiana. study area date nest owner stage kleptoparasite meeman shelby forest, tn meeman shelby forest, tn meeman shelby forest, tn meeman shelby forest, tn meeman shelby forest, tn chickasaw nwr, tn chickasaw nwr, tn chickasaw nwr, tn chickasaw nwr, tn desha, ar desha, ar desha, ar meeman shelby forest, tn yellowwood state forest, in meeman shelby forest, tn desha, ar desha, ar big oaks nwr, in ulster county, ny (smith ) hoosier national forest, in meeman shelby forest, tn may may may may may jun may may jun may may may jun may may may jul may may may may cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler cerulean warbler red-eyed vireo red-eyed vireo blue-gray gnat- catcher building building abandoned building building building incubation immediately after depreda- tion abandoned abandoned dismantling and cerulean warbler reconstructing dismantling and cerulean warbler reconstructing incubation building building incubation unknown incubation abandoned building immediately after depreda- tion american red- start american red- start american red- start american red- start american red- start blue-gray gnat- catcher blue-gray gnat- catcher blue-gray gnat- catcher blue-gray gnat- catcher blue-gray gnat- catcher blue-gray gnat- catcher blue-gray gnat- catcher blue-gray gnat- catcher black-throated green warbler northern parula orchard oriole orchard oriole red-eyed vireo cerulean warbler cerulean warbler cerulean warbler grove had an open canopy, permitting detailed observations of the birds' behavior. the second incident occurred on may between and hrs when a fe- male black-throated green warbler (dendroi- ca virens) was observed taking material from a cerulean warbler nest in yellowwood state forest ( " ' n, " ' w) near blooming- ton, indiana. it landed on the nest branch within m of the nest, approached it in a quick hopping manner, and stole nest material. the female black-throated green warbler re- peated this behavior three times and each time she was chased from the nest by the female cerulean warbler. in one of the three inci- dents, a male cerulean warbler was observed chasing with the female cerulean warbler. eventually, both male and female cerulean warblers exhibited aggressive behavior (mak- ing repeated harsh call notes and chasing) to- " short communications ward the female black-throated green war- bler any time she perched within m of the nest. the male black-throated green warbler was not observed participating in chasing or nest robbing. the cerulean warbler nest was in a red elm (ulmus rubra) near a planting of shortleaf (pinus echinata) and eastern white pine (p. strobus). this may have increased the chances of proximity to a black-throated green warbler territory, as that species is of- ten associated with coniferous forests (morse ). the following account describes nest ma- terial kleptoparasitism with cerulean warbler as the perpetrator. on may between and hrs a female cerulean warbler was observed taking material from a red-eyed vireo nest in the pleasant run unit of the hoosier national forest ( " ' n, " ' w) near bloomington, indiana. the cerulean warbler made three separate trips from its own nest to the vireo nest approximately m to the east, each time successfully acquiring material. on the first trip, the vireo presum- ably did not detect the cerulean warbler, as no interaction occurred; however, the vireo gave chase on the warbler's second and third trips. the stolen material was sufficiently large to be seen in the cerulean warbler's beak, and was incorporated into its nest. both the cerulean warbler and the red-eyed vireo nests appeared to be mostly completed at the time of the raid. construction of the cerulean warbler nest was first observed on may at hrs; on may the nest was confirmed to have failed. discussion the studies in which these ob'servations oc- curred were specifically focused on cerulean warbler nests and it is not surprising that most incidents ( of observations, table ) in- volved cerulean warblers as victims rather than perpetrators. most observations of female cerulean warblers returning to their nests with nesting material did not include observ- ing them collect the material. therefore, ce- rulean warblers may be robbing other nests more often than we are aware. red-eyed vireos and black- throated green warblers are not unlikely participants for nest material thievery interactions with cerulean warblers. red-eyed vireos are one of the most abundant bird species in our study areas (kcj and klr, pers. obs.), and both red-eyed vir- eos and black-throated green warblers forage at the same heights where cerulean warbler nests occur (morse , cimprich et al. ). additionally, and perhaps most impor- tantly, cerulean warblers, red-eyed vireos, and black-throated green warblers may have similar requirements for nest composition and compete for materials. red-eyed vireos have been documented to use all of the same ma- terials as cerulean warblers, with the excep- tion of moss, in their nests (harrison , ehrlich et al. , cimprich et al. ). black-throated green warblers have been documented to use all materials except grape- vine bark and lichen (ehrlich et al. , morse ). prolonged completion of nest building may indicate that nest robbing has taken place, as was likely the case in the second nest robbing account. in this case, the cerulean warbler nest was probably not the pair's first attempt of the season based on the late date of nest construction. they began building their sec- ond nest on may , four days before nest robbing was observed. immediately prior to observing nest material kleptoparasitism, the female cerulean warbler was seen bring- ing nesting material to her nest. as replace- ment nests are typically constructed more quickly than first nests, a -day spread of nest building was unexpected (kcj and klr, pers. obs.). despite the interference, the nest suc- cessfully fledged at least two offspring. why would individuals risk physical con- flict with neighboring birds to steal nesting material? it is possible that ( ) some materials may take a great amount of time to locate, and ( ) some materials may not be difficult to lo- cate, but may be difficult to remove and col- lect in quantity in an appropriate size or shape to be incorporated into a nest (yezerinac ). nest material may be in limited supply, as a result of the large demand for it by a variety of users, or the phenology of the source is limited in time. nest construction is a time- consuming process, as it requires the adult bird to locate materials, to gather the materials into the nest site, and to form the actual nest. nest construction appears to be a costly pro- cess in terms of energy expenditure. this pro- the wilson journal of ornithology vol. , no. , june cess involves a multitude' of flights to gather material. bent ( ) indicated a single fe- male american redstart might make trips in the construction of a single nest. the fe- male pulls material from sources, such as dried cambium of broken tree branches and grapevines, even from inner portions of stems of herbaceous materials. the female must identity the most useful adhesive materials for attaching nests to supports and for holding surfaces together. "spider webs" or silk from cocoons of emerged moths are often listed as the adhesive material. a variety of types of spider silk exist; variation among species may indicate that some may be more useful than others and the possibility of discriminate se- lection by avian users (gosline et al. , gurovec and sehnal ). early nesting be- havior may in part be favored by the ability to find the most effective nest materials. com- petition for materials that are limited in time may occur between users of the favored ma- terial, irrespective of cost considerations. benefits of stealing nest materials are sub- stantial reductions in ( ) distance the female must travel from her nest, ( ) time spent away f r o m her nest, and/or ( ) amount of effort spent while away from the nest. birds may be more likely to resort to time-saving nest klep- toparasitism because the benefits outweigh the risks. the red-eyed vireo and the black- throated green warbler may have robbed ce- rulean warbler nests because of the pressure of time during what was probably (based on the time of season) their second nesting at- tempt. another potential benefit of engaging in nest kleptoparasitism is decreased predation risk. nest building birds may experience in- creased predation risk when gathering so much material in such a short period of time, often from relatively few locations which are repeatedly visited. canopy-nesting species other species, including but not limited to birds, mammals, snakes, and parasitic insects, may also observe locations frequented by nest building birds. banks and martin ( ) noted that when visitations by nest owner decreased in frequency, rates of brown-headed cowbird (molothrus ater) nest parasitism increased. therefore, less time spent foraging for nesting materials would allow the female more time for activities such as egg-laying and nest guarding. all of these factors, acting individually or in concert, indicate that time of nest construc- tion is a critical part of the life cycle of a bird, when the most important reproductive activi- ties, including mate selection and egg produc- tion, occur. study of the process is difficult, particularly for small species that nest in tree canopies in forest habitats. it is difficult to ob- serve canopy-nesting species gathering nest- ing materials. consequently, few data exist on predation risks associated with gathering nest material and the propensity to avoid certain locations as sites for gathering material. it is not clear exactly how advantageous or disadvantageous kleptoparasitism of nest ma- terial is to individuals which participate in it. the perpetrators in all three detailed accounts presented, consistently returned to the victim- ized nests, despite the threat of being chased, indicating the rewards of kleptoparasitism out- weighed the risks. one of the two cerulean warbler nests that was raided was successful. this indicates that victimization did not ne- cessitate failure. the cerulean warbler pair that stole from the vireo nest was unsuccessful in raising young despite any benefits obtained from nest robbing. further studies on nest rob- bing behavior, as related to the phenology, availability, and selection of nest materials will contribute to our understanding of when a nest-building female engages in nest mate- rial kleptoparasitism. must often resort to gathering nesting material near or on the ground. these sites are not typ- acknowledgments ical for the species; the birds may have less we thank the u.s. fish and wildlife service, u.s. experience with the potential escape routes forest service, sigma xi, indiana academy of sci- from predators available in such situations. ence, garden club of america, and ball state univer- the large number of to and from a nest sity office of academic research and sponsored pro- grams for funding support. work in tennessee and ar- site increases the likelihood that a nest para- kansas was supported by the u.s. forest service and site may observe the location, follow the pro- initiated by p. smith and r. g . cooper. we ac- gress of the construction, and time their par- knowledge the dedication to accurate field observation asitic laying event. it is equally possible that displayed by roger allan, david arbour, bob ford, short communications jenna hestir, dawn hinebaugh, cary leiper, matt marshall, curt mccasland, darren pierce, melinda schaefbauer, larry wood, chris woodson, and rich young, whose observations form the substance of this report. our manuscript was greatly improved by the comments and suggestions of carol ann islam, sarah jones, three anonymous reviewers, and the editor. literature cited banks, a. j. and t e. martin. . host activity and the risk of nest parasitism by brown-headed cowbirds. behavioral ecology : - . bent, a. c. . life histories of north american wood warblers. part . u.s. national museum bulletin . cimprich, d. a., e r. moore, and m. p. guilfoyle. . red-eyed vireo (vireo olivaceus). the birds of north america. number . dater, e. e. . first successful nesting of the ce- rulean warbler in new jersey. wilson bulletin : - . ehrlich, p r., d. s. dobkin, and d. wheye. . the birder's handbook: a field guide to the natural history of north american birds. simon and schuster, new york, usa. gosline, j. m., p a. guerette, c. s. ortlepp, and k. n. savage. . the mechanical design of spider silks: from fibroin sequence to mechanical function. journal of experimental biology : - . hamel, p b. . cerulean warbler (dendroica cer- ulea). the birds of north america. number . harrison, h. . a field guide to the birds' nests. houghton mifflin, boston, massachusetts, usa. morse, d. h. . black-throated green warbler (dendroica virens). the birds of north america. number . oliarnyk, c. j. and r. j. robertson. . breeding behavior and reproductive success of cerulean warblers in southeastern ontario. wilson bulletin : - . roth, is. l. . cerulean warbler breeding biology. thesis. ball state university, muncie, indiana, usa. smith, g. c. . . breeding cerulean warblers (dendroica cerulea) in esopus and west park, u - ster county, new york. thesis. state university of new york at new paltz, new paltz, usa. yezerinac, s. m. . american redstarts using yel- low warblers' nests. wilson bulletin : - . zurovec, m. and e sehnal. . unique molecular architecture of silk fibroin in the waxmoth, gal- leria mellonella. journal of biological chemistry : - . the wilson journal of ornithology ( ): - , home range and dispersal of juvenile florida burrowing owls robert j. mryka ,l.~ melissa m. grigione,' and ronald j. sarno abstract-we present the first use of necklace telemetry capabilities. aerial telemetry assisted in lo- radio transmitters to document the home range and dis- cating one juvenile burrowing owl using scrub oak persal of juvenile burrowing owls (athene cunicularia (quercus spp.) habitat approximately . krn southeast jeoridana) during the breeding and post-breeding period of its main and satellite burrows. received februaly in rural florida. juvenile burrowing owls (n = ) were . accepted october . detected close to main and satellite burrows during day-time relocations. home range estimates ( % ker- nel) for juvenile owls varied from to m . juvenile burrowing owls were not detected near main and sat- ellite burrows during three evening relocations. dis- persal of juvenile owls coincided with flooding of bur- rows during the rainy season. juvenile owls upon fledg- ing used an extensive patch of saw palmetto (serenoa repens) before dispersing beyond the range of ground department of environmental science and policy, university of south florida, e. fowler ave., tampa, fl , usa. department of biology, university of south florida, bsf , e. fowler ave., tampa, fl , usa. corresponding author; e-mail: rnrykalo @hotmail.com early observations of florida burrowing owls (athene cunicularia jloridana) describe their propensity to excavate burrows in short grass habitat (hoxie , rhoads , scott , palmer ). typically, a breeding pair of owls excavate one breeding burrow and one or more satellite burrows (scott , neil , wesemann , mealey ). bur- rows, which can be - m in length, contain an enlarged nest chamber at their terminus (rhoads , scott , nicholson , sprunt ). male and female florida bur- rowing owls can breed at l year of age (haug society of american archivists the society of american archivists will hold its thirty-sixth annual meeting from october to november at t h e sheraton- columbus motor hotel in columbus, ohio. this is t h e f i r s t four day meeting ever held by the society and the f i r s t in the midwest f o r several years. program chairman is herman j. viola, who received a doctorate from indiana university and is a former assistant editor of the zndiana magazine of history. the sessions will include papers, workshops, and panels on archives and library administra- tion, publications, oral history, special collections, manuscripts, and many other subject matter fields of research. one of the featured speakers will be professor thomas d. clark of indiana university. the latest archival equipment, literature, and supplies will be on display along with professional exhibits from historical and archival institutions which will demonstrate new techniques, projects, or topically relevant studies. for those interested in employment as an archivist, representatives from t h e archival, historical, and library organizations attending the meeting will be available f o r questions concerning career opportunities. for further information o r copies of the program write to david r. larson, saa local arrangements chairman, c/o the ohio historical society, columbus, ohio . society of indiana archivists a group of hoosier archivists have recently formed t h e society of indiana archivists. membership is open to all hoosiers and organ- izations interested in collecting and preserving manuscripts and permanent archives needed to provide adequate documentation for indiana history. help and infor- mation for indiana record depositories-including religious, busi- ness, university, and county archives-which seek counsel concerning a proper program of archival management, aid to members through discussions and technical programs, publications f o r loan through a technical library, and analysis of unique indiana archival problems through the publication of a newsletter and a technical leaflet series. the f i r s t meeting of sia will be at the depauw university library, greencastle, on saturday, september , . an organiza- tional meeting will begin at a. m. followed by lunch and an informal period to meet archivists and curators specializing in various areas, to evaluate archival supplies, and to learn of archival pro- grams such as microphotography. the program will feature a speaker and will end by : p.m. for further information write to john j. newman, society of indiana archivists, indiana state library, north senate, indianapolis, indiana . the organizers plan several areas of action: horizons the journal of the college theology society published at villanova university volume fall, number articles joseph a. bracken, s.j., intentionality analysis and intersubjectivity michele saracino, hybridity and trespass: with jesus at the borders of identity brian m. doyle, social doctrine of the trinity and communion ecclesiology in leonardo boff and gisbert greshake cara anthony, josiah royce and catholic religious experience editorial essays robert masson, rahner’s primordial words and bernstein’s metaphorical leaps: the affinity of art with religion and theology norbert j. rigali, s.j., the ecclesial responsibilities of theologians, forty years after vatican ii editorial symposium avery cardinal dulles, s.j., catholicism : challenges to a theological education six perspectives: william l. portier, lawrence s. cunningham, anne m. clifford, john c. cavadini, mary ann hinsdale, christopher ruddy author’s response: avery cardinal dulles, s.j. report of the college theology society review symposium bernard p. prusak, the church unfinished: ecclesiology through the centuries four perspectives: dennis m. doyle, susan k. wood, paul lakeland, francine cardman author’s response: bernard p. prusak plus reviews of books h o r iz o n s v o lu m e n u m b e r f a ll a quarterly journal devoted to the american catholic experience one year subscription: $ (u.s.) $ (outside u.s.) � payment enclosed, payable to : american catholic historical society subscription services • po box • hanover pa - � bill me for credit card orders call - - : subscription services my subscription (and/or gift from): name ___________________________________________ address __________________________________________ city/state/zip ______________________________________ e-mail ___________________________________________ gift to: name ___________________________________________ address __________________________________________ city/state/zip ______________________________________ e-mail ___________________________________________ (your e-mail address will not be sold or disclosed to other organizations) � yes, please notify the recipient of my gift. � detach here give american catholic studies to a friend, relative, pastor, etc. the am e r ic a n ca th ol ic h isto r ic a l s o ciety• in corp. dec. , american catholic studies journal of the american catholic historical society www.amchs.org � www.publications.villanova.edu/acs/ all subscribers are also automatically members of the american catholic historical society, the oldest catholic historical society in the united states, founded july , . american catholic studies became the new title of records of the achs in . the journal has thus been in continuous publication for well over one hundred years. it is published at villanova university. subscribers / members of the society include universities, libraries and individuals of various denominations. the achs is committed to preserving and documenting the part played by catholics in the development of the united states. terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core articles intentionality analysis and intersubjectivity .... joseph a. bracken, s.j. hybridity and trespass: with jesus at the borders of identity ..................................................... michele saracino social doctrine of the trinity and communion ecclesiology in leonardo boff and gisbert greshake ........................................................... brian m. doyle josiah royce and catholic religious experience ............. cara anthony editorial essays rahner’s primordial words and bernstein’s metaphorical leaps: the affinity of art with religion and theology ................................................... robert masson the ecclesial responsibilities of theologians, forty years after vatican ii ................................. norbert j. rigali, s.j. editorial symposium avery cardinal dulles, s.j., catholicism : challenges to a theological education six perspectives ............................................................. william l. portier lawrence s. cunningham anne m. clifford john c. cavadini mary ann hinsdale christopher ruddy author’s response .......................................... avery cardinal dulles, s.j. report of the college theology society ................................ review symposium bernard p. prusak, the church unfinished: ecclesiology through the centuries four perspectives ............................................................ dennis m. doyle susan k. wood paul lakeland francine cardman author’s response ........................................................ bernard p. prusak issn— — horizons the journal of the college theology society published at villanova university volume fall, number terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core book reviews j. albert harrill, slaves in the new testament: literary, social, and moral dimensions ............... regina a. boisclair richard lennan, risking the church: the challenges of catholic faith .......................................... nathan kollar theodore walker, jr., mothership connections: a black atlantic synthesis of neoclassical metaphysics and black theology ....................... bryan n. massingale terrence w. tilley, history, theology, and faith: dissolving the modern problematic ................................. ann r. riggs andré trocmé, jesus and the nonviolent revolution ............................................................ j. milburn thompson john r. connolly, john henry newman: a view of catholic faith for the new millennium .................. walter e. conn douglas john hall, the cross in our context: jesus and the suffering world ............................................. stephen j. duffy david ray griffin, two great truths: a new synthesis of scientific naturalism and christian faith .................................................. joseph a. bracken, s.j. paul j. philibert, o.p., the priesthood of the faithful: key to a living church ............................... dennis m. doyle kevin l. hughes, constructing antichrist: biblical commentary and the development of doctrine in the early middle ages ....... michael w. blastic, o.f.m. joseph a. komonchak, ed., history of vatican ii. vol. . church as communion, third period and intersession september –september ............. gregory baum mary doak, reclaiming narrative for public theology ... loretta devoy michael g. lawler, what is and what ought to be: the dialectic of experience, theology, and church ...................................................................... donna teevan erik borgman, edward schillebeeckx: a theologian in his history. vol. : a catholic theology of culture ( – ) ....................... daniel speed thompson marie thérèse archambault, mark g. thiel and christopher vecsey, eds., the crossing of two roads: being catholic and native in the united states ............. michael f. steltenkamp, s.j. marshall d. johnson, the evolution of christianity: twelve crises that shaped the church .............................................................. herbert j. ryan, s.j. dianna ortiz, the blindfold’s eyes: my journey from torture to truth ................................................. tom powers, s.j. daniel c. maguire, a moral creed for all christians ........ gerald s. vigna john wall, moral creativity: paul ricoeur and the poetics of possibility ................................................... john b. lounibos horizons terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core albino barrera, o.p., god and the evil of scarcity: moral foundations of economic agency ................. joseph w. devlin susan rakoczy, great mystics and social justice: walking on the two feet of love .................................. ellen stratton john t. noonan, jr., a church that can and cannot change: the development of catholic moral teaching ........................................................... gerard s. sloyan judith w. kay, murdering myths: the story behind the death penalty ........................................... stephen j. casey kevin f. burke, s.j., pedro arrupe: essential writings; philip endian, karl rahner: spiritual writings; charles e. moore, ed., sadhu sundar singh: essential writings .................................... felicidad oberholzer michael o’laughlin, henri nouwen: his life and vision ......................................................... suzanne mayer, i.h.m. pamela cooper-white, shared wisdom: use of the self in pastoral care and counseling ................... wallace n. fletcher stephanie kaza, ed., hooked! buddhist writings on greed, desire, and the urge to consume ................ paul f. knitter judith a. berling, understanding other religious worlds: a guide for interreligious education ...... leonard j. biallias books received ...................................................................................... index ......................................................................................................................... copyright @ college theology society printed in the united states of america terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core horizons the journal of the college theology society horizons, the journal of the college theology society is published bian- nually in spring and fall at villanova university, villanova, pa . yearly subscription: individual $ . , institutional $ . . send remit- tance, payable to college theology society, p.o. box , hanover, pa . changes of address should also be sent to this address. horizons is automati- cally received through membership in the college theology society. manuscripts, editorial and business correspondence, book reviews, ad- vertising copy, and orders for single issues (individual $ . , institutional $ . ) should be addressed to the editor, horizons, st. mary’s hall, villanova university, villanova, pa . telephone: ( ) - . e-mail: horizons@villanova.edu in submitting manuscripts, please send four clear copies. author’s name should appear only on an unpaginated cover sheet, with address and essay’s title. no other identifying reference to author should appear on the manuscript. all references to author’s work in text or notes should be in the third person. essay’s title should appear at top of first page of text. notes should be placed on separate sheets at end. full double spacing should be used throughout, includ- ing indented quotations and notes. top, bottom, and right-hand margins should be at least one inch; left-hand margins at least one-and-a-half. manuscripts should be accompanied by an abstract of one hundred fifty words or less and an author vita of one hundred words or less. authors should state that the manu- script will not be submitted to another journal until horizons’ evaluation is com- pleted. manuscripts will be returned if postage is supplied. articles appearing in horizons are abstracted and indexed in: the catho- lic periodical and literature index; guide to social science and religion in periodical literature; index to religious periodical literature; new testament abstracts; religious and theological abstracts; and social science citation index. articles are indexed in religion index one: periodicals; book reviews indexed in index to book reviews in religion. horizons is available in micro- form from bell & howell, north zeeb road, p.o. box , ann arbor, mi - . it is also available on line in full text to atlas subscribers. college theology society the college theology society is a professional organization of college and university professors of religion in the united states and canada. the purpose of the college theology society is to improve the quality of the teach- ing of religion: by stimulating and sharing scholarly research; by developing programs of theology and religious studies which meet student needs and interests; and by exploring, evaluating, and encouraging effective ways of teaching which are interdisciplinary and ecumenical. annual membership dues in the society are $ . (full professional or associate), $ . (joint professional for husband and wife), and $ . (graduate student). mem- bership in the society includes a subscription to horizons. contact elena procario-foley, religious studies, iona college, north ave., new rochelle, ny . telephone: ( ) - . e-mail: eprocariofoley@iona.edu terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core editors walter e. conn villanova university anthony j. godzieba villanova university associate editors lisa sowle cahill boston college denise lardner carmody santa clara university anne carr university of chicago bernard cooke college of the holy cross (emeritus) lawrence s. cunningham university of notre dame charles e. curran southern methodist university margaret a. farley yale divinity school elisabeth schüssler fiorenza harvard divinity school francis schüssler fiorenza harvard divinity school mary ann hinsdale boston college elizabeth a. johnson fordham university paul f. knitter xavier university (emeritus) pheme perkins boston college sandra m. schneiders, i.h.m. jesuit school of theology at berkeley william m. shea college of the holy cross gerard s. sloyan temple university (emeritus) william thompson-uberuaga duquesne university david tracy university of chicago book review editors joann wolski conn neumann college eileen flanagan neumann college business editor ellen murray, r.s.m. gwynedd-mercy college editorial assistant irene c. noble terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core horizons the journal of the college theology society a journal exploring developments in catholic theology, the total christian tradition, human religious experience, and the concerns of creative teaching from the college and university environment. “what’s in a name?” for horizons, perhaps everything. this issue features names long associated with horizons—founding co-editor bernard prusak, former associate editor norbert rigali, current associate editors lawrence cunningham and mary ann hinsdale, and such frequent past authors as joseph bracken, dennis doyle, robert masson, and william portier. they are joined by many others familiar to us all, including anne clifford, our new president of the college theology society. we especially welcome avery cardinal dulles participating in our editorial symposium on his “catholicism .” well known names are also interestingly paired in some of our titles and essays: karl rahner and leonard bernstein, leonardo boff and gisbert greshake, bernard lonergan and alfred north whitehead. after a distinguished career at indiana university mary jo weaver has retired from teaching and all other aspects of academic life, including, she claims, her zealous support of the indiana hoosiers in basketball and the fight- ing irish in football. for many years she has been a generous and valuable associate editor, and i am grateful for all her astute contributions. she will be missed here. i also want to thank all my other colleagues who have given so much to horizons over the years and have made my tenure as editor so easy and en- joyable. in this instance i will not begin to name names. they include, of course, cts officers and all those listed formally as editors, whether associate, book review, or business. but there are also many more not so listed who have served generously and significantly as manuscript reviewers. and i am espe- cially grateful to the fine authors and book reviewers whose writing has graced our pages. in the end, however, horizons would not function without the dedicated and expert work of the editorial assistant, and we have been blessed with two wonderful ones, who must be named: first teresa byrne, and now irene noble. as anthony godzieba carries us forward, great thanks to you all! —walter e. conn terms of use, available at https://www.cambridge.org/core/terms. https://doi.org/ . /s downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core https://www.cambridge.org/core/terms https://doi.org/ . /s https://www.cambridge.org/core digital health/social media a cardiovascular health and wellness mobile health intervention among church-going african-americans: formative evaluation of the faith! app laprincess brewer , ashok kumbamu , christina smith , sarah jenkins , clarence jones , sharonne hayes , christi a. patten , lisa a. cooper , and lora burke mayo clinic; mayo clinic college of medicine; hue-man partnership; johns hopkins university school of medicine; university of pittsburgh objectives/goals: to evaluate the faith! (fostering african- american improvement in total health) app mhealth lifestyle intervention by using post-intervention feedback obtained from par- ticipants in our intervention pilot study. methods/study population: we used qualitative methods (focus groups) to elicit post-intervention feedback. participants who completed the pilot study were recruited to one of two focus groups. semi-structured focus groups were conducted to explore participants’ views on the app functionality, utility and satisfaction as well as its impact on healthy lifestyle change. sessions were audio-recorded, transcribed verbatim and qualitative data were analyzed by systematic text con- densation thematic analysis. results/anticipated results: nine individuals participated (n = and n = ) in each of the two focus groups. their mean age was . years (sd . ), % were women, and all had at least an education level of some college. six overarching themes emerged from the data: ( ) overall impression, ( ) content usefulness ( ) formatting, ( ) implementation, ( ) impact and ( ) suggestions for improvement. underpinning the themes was a high level of agreement that the intervention facilitated healthy behavioral change through cultural tailoring, multimedia education modules and social networking. among the suggestions for improve- ment were streamlining of app self-monitoring features, personaliza- tion based on individual’s cardiovascular risk and attentiveness to nuanced cultural perspectives. discussion/significance of impact: this formative evaluation found the faith! app mhealth lifestyle intervention had high reported satisfaction and impact on the health-promoting behaviors of african-americans, thereby improving their overall cardiovascular health. the findings provide further support for the acceptability of mhealth interven- tions among african-americans. conflict of interest description: none. all in for health: promoting good health and engaging a health research volunteer community in the hoosier state jessica hall , christine drury, and carmel egan indiana university school of medicine objectives/goals: • to improve and expand health and research literacy throughout indianabysharinghealth-focusedresourcesandresearchoutcomes. • to encourage and increase health research participation throughout indiana by promoting health research opportunities, including clinical studies. methods/study population: • discover and understand community concerns and barriers to good health and clinical research participation by providing a platform for individuals and communities to share their voices. • educate indiana residents on the importance of participating in health research. • engage with the community to meet them where they are (online) and continue to build relationships throughout the state. • promote healthy living for indiana residents by sharing health education and resources from existing state health organizations and initiatives. • develop and maintain the largest statewide database of research volunteers. results/anticipated results: • the anticipated results from this program include engagement of all populations and all communities throughout the state in con- versation and education around good health and health research, aswell as participation inhealth research acrossthe ctsi’s partner organizations. large-scale growth is expected in both the online communityandconsentedvolunteerregistryisexpectedtoinclude and engage racially and ethnically diverse populations, as well as special health populations, such as representatives of rural com- munities,aged,rarediseasesurvivors,andtransgenderindividuals. discussion/significance of impact: • thorough this work, the indiana ctsi has developed a unique program, educating the public about health research and oppor- tunities to participate, while simultaneously supporting research departments with marketing promotion of their efforts, and a ready statewide volunteer community. an app a day: examining clinical evidence for safety and efficacy of diabetes mobile health apps avantika pathak, ms , susan bain , and eunjoo pacifici usc school of pharmacy; usc department of regulatory and quality sciences objectives/goals: mobile health applications are widely used by the public but vary in how they are classified and regulated. this study examines the evidence of the safety and efficacy of mobile medical applications specifically focusing on those that are used to manage dia- betes. methods/study population: to understand the cur- rent regulatory landscape of mobile health applications (mhealth apps) for diabetes, a literature survey was conducted using the pubmed database. top mhealth apps were identified by searching the apple store website using key terms associated with diabetes management applications. a maximum of ten results, when available for each key term, were studied by exploring the fda databases to understand how the products were regulated and if any were subject to recalls. these selected mhealth apps were also searched on clinicaltrials.govtoseeifthere wereongoingorcompletedclinicaltrials and if the trials were designed to include efficacy and safety outcome measures. results/anticipated results: of the mhealth apps for diabetes management that were identified, were regulated. jcts abstract supplement https://www.cambridge.org/core/terms. https://doi.org/ . /cts. . downloaded from https://www.cambridge.org/core. carnegie mellon university, on apr at : : , subject to the cambridge core terms of use, available at http://clinicaltrials.gov https://www.cambridge.org/core/terms https://doi.org/ . /cts. . https://www.cambridge.org/core [pdf] locally advanced non–small cell lung cancer: what is the optimal concurrent chemoradiation regimen? | semantic scholar skip to search formskip to main content> semantic scholar's logo search sign increate free account you are currently offline. some features of the site may not work correctly. 	doi: . /ccjm. .s . corpus id: locally advanced non–small cell lung cancer: what is the optimal concurrent chemoradiation regimen? @article{videtic locallyan, title={locally advanced non–small cell lung cancer: what is the optimal concurrent chemoradiation regimen?}, author={g. videtic}, journal={cleveland clinic journal of medicine}, year={ }, volume={ }, pages={s - s } } 	g. videtic 	published 	medicine 	cleveland clinic journal of medicine the optimal chemoradiation regimen for patients with locally advanced non–small cell lung cancer (nsclc) has yet to be defined. disease and patient heterogeneity prevent a “one size fits all” approach to treatment. concurrent chemoradiation up front is the definitive strategy for patients with unresectable stage iii nsclc; the addition of consolidation chemotherapy following definitive treatment has produced conflicting results with respect to overall survival. biologic therapies have yet to… expand view on cleveland clinic mdedge.com save to library create alert cite launch research feed share this paper citationsbackground citations view all figures and topics from this paper 	 figure 	 figure 	non-small cell lung carcinoma 	chemoradiotherapy 	small cell carcinoma of lung 	lung consolidation 	chemotherapy 	seizures 	unresectable citations citation type citation type 	all types 	cites results 	cites methods 	cites background has pdf publication type author more filters more filters filters sort by relevance sort by most influenced papers sort by citation count sort by recency proton therapy for non-small cell lung cancer: the road ahead. 	e. brooks, m. ning, v. verma, x. zhu, j. chang 	medicine 	translational lung cancer research 	 	 	 	view excerpt, cites background save alert research feed the effect of radiotherapy dose on survival in stage iii non-small-cell lung cancer patients undergoing definitive chemoradiotherapy. 	m. koshy, r. malik, + authors r. weichselbaum 	medicine 	clinical lung cancer 	 	 save alert research feed inhibiting glucosylceramide synthase facilitates the radiosensitizing effects of vinorelbine in lung adenocarcinoma cells. 	w. chiu, h. h. chen, + authors chiou-feng lin 	medicine 	cancer letters 	 	 	view excerpt, cites background save alert research feed lactoferrin-appended solid lipid nanoparticles of paclitaxel for effective management of bronchogenic carcinoma 	v. pandey, k. r. gajbhiye, v. soni 	medicine 	drug delivery 	 	 	 	view excerpt, cites background save alert research feed an analysis of the immune phenotypes of hla dr, cd and cd for the dcs and of cd , cd and cd for the cik 	 save alert research feed immune response, clinical outcome and safety of dendritic cell vaccine in combination with cytokine-induced killer cell therapy in cancer patients 	y. cui, xuejing yang, w. zhu, j. li, xiaojing wu, y. pang 	medicine 	oncology letters 	 	 	pdf save alert research feed egcg enhances the efficacy of cisplatin by downregulating hsa-mir- - p in nsclc a cells 	dong-hu zhou, x. wang, q. feng 	chemistry, medicine 	nutrition and cancer 	 	 	view excerpts, cites background save alert research feed paris saponin i induces apoptosis via increasing the bax/bcl- ratio and caspase- expression in gefitinib-resistant non-small cell lung cancer in vitro and in vivo. 	h. jiang, pengjun zhao, d. su, j. feng, s. ma 	medicine, biology 	molecular medicine reports 	 	 	 save alert research feed combination of low concentration of (−)-epigallocatechin gallate (egcg) and curcumin strongly suppresses the growth of non-small cell lung cancer in vitro and in vivo through causing cell cycle arrest 	dong-hu zhou, x. wang, mingmin yang, x. shi, wenbin huang, q. feng 	medicine 	international journal of molecular sciences 	 	 	pdf save alert research feed large-scale gene-centric meta-analysis across studies identifies multiple lipid loci. 	f. asselbergs, y. guo, + authors f. drenos 	biology, medicine 	american journal of human genetics 	 	 	pdf save alert research feed ... ... references showing - of references sort byrelevance most influenced papers recency management of unresectable stage iii non-small-cell lung cancer with combined-modality therapy: a review of the current literature and recommendations for treatment. 	neil a bayman, f. blackhall, p. jain, l. lee, n. thatcher, c. faivre-finn 	medicine 	clinical lung cancer 	 	 save alert research feed concurrent cisplatin, etoposide, and chest radiotherapy in pathologic stage iiib non-small-cell lung cancer: a southwest oncology group phase ii study, swog . 	k. albain, j. crowley, + authors r. livingston 	medicine 	journal of clinical oncology : official journal of the american society of clinical oncology 	 	 save alert research feed concurrent chemoradiotherapy in non-small cell lung cancer. 	n. o’rourke, marta roqué i figuls, nuria farré bernadó, f. macbeth 	medicine 	the cochrane database of systematic reviews 	 	 save alert research feed combined chemoradiotherapy regimens of paclitaxel and carboplatin for locally advanced non-small-cell lung cancer: a randomized phase ii locally advanced multi-modality protocol. 	c. belani, h. choy, + authors w. curran 	medicine 	journal of clinical oncology : official journal of the american society of clinical oncology 	 	 save alert research feed phase iii study of cisplatin, etoposide, and concurrent chest radiation with or without consolidation docetaxel in patients with inoperable stage iii non-small-cell lung cancer: the hoosier oncology group and u.s. oncology. 	n. hanna, m. neubauer, + authors l. einhorn 	medicine 	journal of clinical oncology : official journal of the american society of clinical oncology 	 	 save alert research feed phase ii study of pemetrexed, carboplatin, and thoracic radiation with or without cetuximab in patients with locally advanced unresectable non-small cell lung cancer: calgb . 	r. govindan, j. bogart, x. wang, l. hodgson, r. kratzke, e. vokes 	medicine 	journal of clinical oncology : official journal of the american society of clinical oncology 	 	 save alert research feed long term survival in stage iiib non-small cell lung cancer (nsclc) treated with consolidation docetaxel following concurrent chemoradiotherapy (swog s ) 	d. gandara, k. chansky, l. gaspar, k. albain, p. lara, j. crowley 	medicine 	 	 save alert research feed final results of phase iii trial in regionally advanced unresectable non-small cell lung cancer: radiation therapy oncology group, eastern cooperative oncology group, and southwest oncology group. 	w. sause, p. kolesar, + authors a. turrisi 	medicine 	chest 	 	 	pdf save alert research feed meta-analysis of concomitant versus sequential radiochemotherapy in locally advanced non-small-cell lung cancer. 	a. aupérin, c. le péchoux, + authors j. pignon 	medicine 	journal of clinical oncology : official journal of the american society of clinical oncology 	 	 , save alert research feed radiation therapy oncology group (rtog) - and eastern cooperative oncology group (ecog) : preliminary results of a phase iii trial in regionally advanced, unresectable non-small-cell lung cancer. 	w. sause, c. scott, + authors a. turrisi 	medicine 	journal of the national cancer institute 	 	 save alert research feed ... ... related papers 	abstract 	figures and topics 	 citations 	 references 	related papers stay connected with semantic scholar sign up about semantic scholar semantic scholar is a free, ai-powered research tool for scientific literature, based at the allen institute for ai. learn more → resources datasetssupp.aiapiopen corpus organization about usresearchpublishing partnersdata partners   faqcontact proudly built by ai with the help of our collaborators terms of service•privacy policy the allen institute for ai by clicking accept or continuing to use the site, you agree to the terms outlined in our privacy policy, terms of service, and dataset license accept & continue table assessing individual risk attitudes using field data from lottery games* connel fullenkamp rafael tenorio duke university depaul university and robert battalio university of notre dame * we thank gabriella bucci, larry samuelson, richard sheehan, various workshop participants, and two anonymous referees for numerous comments. part of this project was completed while tenorio visited the management and strategy department at northwestern university. he is thankful for their hospitality. trish and marcos delgado provided superb research assistance. all errors or omissions are the authors'. please address correspondence to: rafael tenorio, depaul university, department of economics, e. jackson blvd. – suite , chicago, il - . e-mail: rtenorio@condor.depaul.edu. assessing individual risk attitudes using field data from lottery games abstract we use information from the television game show with the highest guaranteed average payoff in the u.s., hoosier millionaire, to analyze risk-taking in a high-stakes experiment. we characterize gambling decisions under alternative assumptions about contestant behavior and preferences, and derive testable restrictions on individual risk attitudes based on this characterization. we then use an extensive sample of gambling decisions to estimate distributions of risk aversion parameters consistent with the theoretical restrictions and revealed preferences. we find that although most contestants display risk averse preferences, the extent of the risk aversion implied by our estimates varies substantially with the stakes involved in the different decisions. jel codes: c , d , c i. introduction a series of recent studies (rabin a, b; rabin and thaler ) have pointed out the inadequacy of expected utility theory to provide proper characterization of risk aversion when monetary stakes are small. the basic contention of this criticism is that the concavity of the utility function would have to be so pronounced to explain small stake risk aversion, that this same utility function would imply absurd levels of risk aversion for large stakes. rabin ( b) concludes that "expected-utility theory seems to be a useful and adequate model of risk aversion for many purposes, such as understanding large-stakes insurance…," but "is manifestly not close to the right explanation of risk attitudes over modest stakes…" a number of studies have attempted to characterize individual risk aversion under relatively large stakes using expected utility theory. experiments by binswanger ( ), and kachelmeier and shehata ( ) face subjects with decisions involving small nominal stakes that are large relative to the subjects' wealth. their findings are mixed: while the binswanger subjects exhibit more risk averse behavior as stakes are increased, kachelmeier and shehata conclude that "the effects of monetary payoffs are real, albeit subtle, and are in need of further study." unfortunately, the inherent limitations of rewards in experimental studies make it difficult to get more robust conclusions about risk-taking in large stakes settings. a second group of papers analyzes risk-taking behavior using information from natural experiments involving large stakes. gertner ( ) uses betting decisions from the bonus round of the television game show card sharks, and concludes that individuals exhibit a small degree of risk aversion. metrick ( ), hersch and mcdougall ( ), analyze field data from wagering decisions on the television game shows final jeopardy! and illinois instant riches, respectively, and find that contestants display behavior that is statistically indistinguishable from risk neutrality. in contrast, beetsma and schotman ( ), conclude that contestants in the dutch television show lingo, exhibit substantial risk aversion. in this paper we use information from the television show hoosier millionaire to gain additional insight into the behavior of individuals in high stakes situations. although this show has undergone a number of regime changes, the current regime is representative of the typical decision a contestant faces: she may either (a) take a sure $ , , or (b) play a game in which she gets $ , , $ , , , , , or $ , each with equal probability. if she draws $ , or $ , , she may stop playing and keep the prize, or continue drawing from the remaining alternatives until she either decides to stop, or draws $ or $ , , . prizes are not cumulative. we believe, for various reasons, that the structure of the decision problems in hoosier millionaire provides us with a very appealing natural experiment to study individual decisions under risk. first, unlike previous game studies, the realization of the risky prospect is the outcome of a very simple probability process, where no subjectivity or skills are involved . second, the stakes are larger than those in previous studies, thus allowing a more proper characterization of risk aversion using expected utility theory. finally, our data comprises a time series of games involving several regime changes, which allows us to study the sensitivity of individual behavior across various gambles. our approach to analyzing risk aversion differs significantly from previous game-based studies. exogenous wagers and lack of variation in the gambles prevent us from estimating risk aversion parameters using regression techniques. instead, we use the sample properties of the decisions in the different regimes of hoosier millionaire to estimate the parameters of the distribution of risk aversion coefficients in the contestant population. in addition, unlike previous studies, we explicitly account for various degrees of contestant rationality in our estimation. our results show that most individuals behave in a way consistent with risk aversion, and as noted in recent studies, the implied degree of risk aversion varies largely with the stakes of the decision. in fact, our estimates imply substantial risk aversion for gambles comparable to those in hoosier millionaire, while also implying near risk neutrality for smaller gambles of similar structure. we also show that the impact of different behavioral traits on decisions and risk aversion is only evident under large stakes. ii. a brief history of hoosier millionaire the hoosier lottery’s weekly television show, hoosier millionaire, first aired on october , . since its inception, each show has involved six contestants chosen randomly from a pool of entries submitted by people playing a scratch-off ticket game. all contestants participate in the first of two phases played on each show. in phase i, all six contestants play a series of purely random draw games. the contestant who amasses the most cash wins and proceeds to phase ii, while the remaining five contestants leave the show. all six contestants keep the cash and prizes accumulated in phase i. four regimes have governed phase ii of hoosier millionaire. in regime , the phase i winner has the option of choosing one of four numbers randomly associated with the dollar amounts $ , , $ , , and $ , , , and a consolation prize of $ , . if the contestant draws $ million, she wins that amount, paid-out in equal annual payments, and the game is over. if the contestant draws $ , or $ , , she can keep that amount or give it up and make another choice among the remaining alternatives. she continues until she chooses to stop, wins the $ million, or draws the consolation prize. in october of , the two intermediate amounts offered in the gamble changed from $ , and $ , to $ , and $ , . all other features remained the same. we call this regime . in february of , more changes were made to phase ii. first, the lowest amount attainable in the initial drawing changed from $ , to $ , . second, the consolation prize went down from $ , to , scratch-off lottery tickets with a purchase price of $ each. finally, lottery officials changed the payment of the $ million prize to twenty annual payments of $ , . we call this regime . rules currently governing phase ii of hoosier millionaire were instituted in february of . according to pat traub, the lottery's acting deputy director, these changes offer "...the highest guaranteed prize of any game show in the nation..." and were made to "...enhance the show's entertainment value." two revisions were made. first, a contestant is now automatically endowed with a guaranteed $ , prize. she may then decide to keep that amount and walk away from the show, or give it up and choose one of the four numbers just as in regime . the other change involved extending the period of time over which the $ , , prize is paid out from twenty to twenty-five years . we call this regime . iii. analytical issues in analytical terms, the problem a hoosier millionaire contestant faces is one of sequential decisions with no recall. this means that the outcome of the first decision may give the contestant the option to continue playing, but once a continuation decision is made at any stage, the status quo is no longer available. in a problem like this, the optimal decisions follow from using a dynamic programming (or backward induction) technique. this means starting at the final decision node, assessing the values of continuing and stopping at that stage, and then factoring in these values when making decisions at the previous stage(s). this procedure is equivalent to finding the subgame perfect equilibrium in a game against nature. in analyzing the optimal strategy for each regime, we consider the following behavioral hypotheses: full rationality (fr). a contestant performs backward induction at each decision node, and discounts the $ , , prize based on the annuity system . bounded rationality (br ). same as fr but no discounting is applied to the $ , , prize. bounded rationality (br ). contestants do not perform backward induction at every decision node, i.e., they do not take into account the value of the option to continue playing the game, and evaluate each lottery as a simple rather than a sequentially compounded lottery. annuities are discounted. bounded rationality (br ). same as br but annuities are not discounted. our bounded rationality scenarios are motivated by biases observed in decision settings comparable to our problem. empirical and experimental evidence suggests that low or even negative discounting is common in decisions involving evaluation of income streams over time (br , br ). loewenstein and prelec ( ), loewenstein and thaler ( ), and gigliotti and sopher ( ), note that most subjects prefer a more constant and spread out stream of income over a strictly decreasing pattern of payments , . there is also extensive evidence that individuals fail to perform backward induction in sequential decision environments (br , br ). carbone and hey ( ), report a series of decision-making experiments where individuals do not apply bellman’s principle of optimality. similarly, camer et. al. ( ) show that subjects do not reason backwards in simple alternating-offers bargaining experiments. instead, most subjects base their decision on current round payoffs . figures and show the game trees associated with regimes and of hoosier millionaire. we subsequently analyze optimal decisions under each of the behavioral hypotheses outlined above. (i) risk neutrality. if decisions depend solely on expected values at each decision node, simple calculations show that: (a) boundedly rational contestants who do not discount annuities (br and br ) always choose the gamble over the sure prospect. here, the $ , , prize makes the gambles’ expected values very large across the board, even when contestants do not backward induct. (b) if contestants discount annuities (fr and br ), gambling may become unattractive as discounting increases. table shows the discount rates that make a contestant indifferent between gambling and keeping the sure prospect at the various decision nodes. as shown, for sufficiently low discount rates (r< . %), it is optimal to gamble in any regime at any decision node for both the fr and br contestant types. as discounting increases, some of the gambles' expected values fall below the sure prospects. in regimes and , discount rates must be very high (above %) to discourage individuals from gambling. this is because the $ million annuity payments are only spread across years and the consolation prize is substantial ($ , ). in regimes and , where annuities are paid over - years and the consolation prize is only $ , , lower discount rates may induce individuals to take on various gambles. (ii) risk aversion . to make our results comparable to those of previous studies we assume that hoosier millionaire contestants display one of the following two types of preferences: constant absolute risk aversion (cara). these preferences, which imply increasing relative risk aversion, are often assumed in studies of individual decision-making. their generic utility representation is: awewu −−=)( ( ), where w is the individual's wealth, a is the coefficient of absolute risk aversion, and aw is the coefficient of relative risk aversion. this formulation is convenient because it allows one to calculate absolute risk aversion coefficients without any wealth information. constant relative risk aversion (crra). these preferences, which imply decreasing absolute risk aversion, are commonly used in macro and asset-pricing studies. their generic utility representation is: b w wu b − = − )( ) ( ( ), where w is the individual's wealth, b is the coefficient of relative risk aversion, and b/w is the coefficient of absolute risk aversion. since this formulation implies that risk aversion depends on wealth, we must know something about this variable to make meaningful inferences. given lack of comprehensive wealth information, we make alternative assumptions about the reference wealth level that individuals use to make decisions. we first use a restrictive definition of wealth, w , encompassing only the winnings accumulated in hoosier millionaire. this partial asset integration in decision-making is to some extent consistent with kahneman and tversky's prospect theory. as a test of robustness we will also use a broader wealth definition (w ), which includes the median household income of the individual’s census tract in . table shows the values for the cara and crra (w ) parameters that make contestants indifferent between gambling and keeping the sure prospect on every initial subgame in every regime of hoosier millionaire. that is, for any risk aversion parameter larger than a given parameter in this table, the contestant will prefer the sure prospect to the gamble. as expected, as discounting increases, the value of the $ , , annuity payments becomes small relative to the lump-sum payoffs. thus, at some discount rate the expected value of a gamble becomes smaller than the sure prospect, and the contestant would have to be a risk lover to gamble (see selected entries in table when r= % and r= %). also, for a given discount rate, the risk aversion coefficient that dissuades a contestant from gambling decreases with the amount of the sure prospect, regardless of the degree of contestant rationality. this follows from certainty-equivalence. iv. data we obtained the hoosier millionaire data from the hoosier lottery office in indianapolis, and from press releases in the south bend tribune. with a few exceptions, our data set contains the winnings and census tract characteristics of all the participants in both phases of hoosier millionaire since its original inception. table presents a summary of the contestants' endogenous initial gambling decisions . we also report mean phase winnings and census tract median income figures for our sample. as seen in table , although % of the individuals in our sample take the initial gamble, this fraction varies across regimes. in fact, the rough comparative statics of decisions across regimes indicate that contestants correctly respond to changes in the award structure. for instance, in regime the sizable consolation prize ($ , ) substantially reduces the downside of gambling. as a result, all of the contestants in this regime choose to gamble. this contrasts with regime , where the small consolation prize ($ , ) makes the downside of gambling rather onerous, thus inducing of contestants to keep the sure prospect. the fourth column of table shows phase winnings for each contestant in our data set. as seen, there is not much variation in these winnings. we obtained the addresses of all phase hoosier millionaire contestants, and found matching data from their census tracts. table also shows the average median family and per-capita income for our sample. although both of these income measures are below state averages, they are not significantly different from them . however, there is no systematic difference between the incomes of contestants that took the gambles and contestants that chose the certain prospects . v. estimation (i) methodology previous studies such as gertner ( ), metrick ( ), and hersch and mcdougall ( ) estimated the coefficient of absolute risk aversion using a non-linear probit approach. unlike those studies, where individuals face gambles involving different stakes, each contestant within a regime of hoosier millionaire faces a gamble with identical stakes. thus the probit technique is inappropriate because a key source of variation --that between the size of each gamble-- is absent. instead, we use a probabilistic approach and estimate the distribution of the risk aversion parameter. we assume that contestants are endowed with risk aversion parameters that are independent draws from a normal population distribution . if a contestant's realization of the risk aversion parameter is less than the value that yields indifference, or equality of expected utilities from gambling and not gambling, she chooses the gamble. the probability that the contestant's risk aversion parameter lies below the indifference value is given by the value of the cumulative normal distribution evaluated at the indifference value. as we know, the normal distribution depends on its mean and standard deviation. to estimate these parameters, we use the indifference values in table and the binomial probabilities implied by the empirical frequencies in table . we infer the mean and standard deviation as follows: each subgame in the history of hoosier millionaire is associated with a cara or crra indifference parameter, which we denote ρ, and an observed frequency of taking the gamble, denoted p. we choose pairs of subgames, and solve for the (unique) values of the mean µ and standard deviation σ that satisfy the following system: )( ) ( pdxe x =∫ ∞−         −−−ρ σ µ σπ ( a) )( ) ( pdxe x =∫ ∞−         −−−ρ σ µ σπ ( b) we then used monte carlo simulations to place confidence intervals around the estimated mean and standard deviation. for each monte carlo trial, we generated two data sets consisting of indicator random variables, and used the implied p’s from this pseudo data to solve for µ and σ. the data generating process for each set of indicator variables was a binomial distribution with parameter (probability of success on each binomial trial) equal to the observed p from the respective subgame. the number of observations in each set of pseudo data was set equal to the number of actual observations of the respective subgame. we repeated this experiment , times for each regime pair and behavioral assumption, ranked the estimated means and standard deviations, and identified % confidence intervals. although we performed this experiment using several pairs of subgames, we focus on the pairing of regimes and . . . table reports estimated means and variances of the cara and crra coefficients for each behavioral assumption, along with the monte carlo generated % confidence intervals. (ii) results a feature of our results is their robustness across utility specifications. the parameter estimates for the cara and crra(wo) functions yield similar distributions and exhibit comparable patterns. the following discussion makes use of this similarity by grouping the results from both specifications whenever possible. first, our results consistently support risk aversion. estimated mean risk aversion coefficients are always positive and the confidence intervals around these means never include zero. mean risk aversion parameter estimates under cara range from . e- to . e- , while those under crra utility range from . to . . moreover, the minimum means implied by the monte carlo simulations are always positive. this evidence supports the hypothesis that on average, the individuals on our sample are risk averse. however, the estimated standard deviations of the risk aversion parameters indicate that some individuals may display risk neutral or risk loving preferences. this is particularly the case for crra utility, where a significant fraction of estimated risk aversion parameters are non-positive. a two-standard deviation interval around the mean risk aversion parameter generally includes only positive values for the cara utility function but encompasses zero and negative values in the crra case. our second main result is that different behavioral assumptions affect the estimates of mean risk aversion. for fixed discounting, confidence intervals for these means under full rationality (fr) and non- backward induction (br and br ) are mutually exclusive. in addition, when comparing no discounting and % discounting, holding behavior fixed, confidence intervals are mutually exclusive in three of four cases. we discuss the direction and significance of these differences in the next section. as a test of robustness of our crra results, we estimated the distribution parameters using a more inclusive definition of wealth (w ). in this specification, in addition to phase i winnings, the initial wealth also included the average median income for the contestant’s census tract in . as we see in table , the higher initial wealth uniformly increases the means and standard deviations of the crra parameter distributions. however, most of these increases are of very small order (between . and . ), and as we show in the next section, do not affect the basic interpretation of the results. the similarity between the two sets of crra estimates is due to the fact that that both wealth definitions are a relatively small fraction of the stakes of the gambles in hoosier millionaire . vi. discussion and conclusions the results from the previous section uniformly indicate that most individuals display some degree of risk aversion. to gain insight into the economic significance of these results, we first compare them to those obtained in related work. our estimated means of the cara parameter distributions are lower than the estimated cara parameters of other game show studies. gertner ( ), using two alternative methods, estimates statistically significant lower bounds on this parameter of . or . . metrick ( ), and hersch and mcdougall ( ) report estimates ranging from . to . , but these estimates are not statistically different from zero, leading to the conclusion of risk neutrality. in contrast, our largest estimated mean cara parameter is . , and the largest upper bound on a confidence interval is . , both of which are nearly an order of magnitude smaller than those reported elsewhere. our crra parameter estimates are also generally lower than those reported elsewhere. friend and blume ( ), using data from individual portfolio holdings find estimates of this parameter ranging from to , while chou et. al. ( ), in an asset-pricing context, estimate this parameter to be around . beetsma and schotman ( ), using game data from lingo, estimate this parameter to be around . in our case the mean crra parameters range from . to . , once again below most estimates from other studies. to gain some intuition into the meaning of our results and how they compare to those of other studies, we calculate certainty equivalents implied by our mean estimates for a variety of gambles (see table ). for instance, for a gamble offering a - chance of winning nothing and $ , , the certainty equivalent ranges from $ . (gertner ) to $ . (hersch and mcdougall ), with most estimates implying a certainty equivalent above $ . in our experiments, certainty equivalents implied by point estimates of the mean of the parameter distribution range from $ . to $ . . thus for gambles of this magnitude, not only are our results similar to those in other studies, but they imply behavior close to risk neutrality. moreover, the tightness of the range of these certainty equivalents indicates that the differences among risk aversion parameters across behavioral assumptions may not have much economic significance when stakes like these are involved. in fact, as seen in table , near risk neutrality across all behaviors obtains even as stakes rise to several thousand dollars. this point, also emphasized on table of rabin and thaler ( ), suggests that gambles of this size unavoidably lead to parameter estimates that imply near risk neutrality within the expected utility paradigm. thus, given that most previous studies based their estimates on gambles with maximum stakes around $ , , it is not surprising at all that the results point towards risk neutrality or mild risk aversion . however, we estimated our risk aversion parameters from decisions involving unusually high stakes. in fact, stakes in hoosier millionaire are several orders of magnitude ( or more in most cases) higher than those in any of the games analyzed in other studies. thus, we must base our inferences on stakes of this magnitude. for instance, for a - gamble between winning nothing and $ million, the certainty equivalents that our mean estimates imply range from $ , to $ , , indicating substantial risk aversion. moreover, there is wide variation of certainty equivalents depending on rationality assumptions. more precisely, fully rational contestants display uniformly higher certainty equivalents than all boundedly rational contestants, and contestants who do not backward-induct and fail to discount their payoffs display the lowest certainty equivalents. these facts suggest that the willingness to take on large gambles is inversely related to the extent of rationality of the contestant. because boundedly rational contestants fail to account for the option value of continuing to make decisions, their perceived expected utility of gambling is below that of a fully rational contestant. table also reports the ce’s associated with two intermediate gambles, which shows how risk aversion starts to surface more dramatically as the stakes grow larger. what can we learn from these results? first, one should be very careful in drawing general conclusions about preferences from context-specific estimates. in our case, when taken to the context of previous studies, our estimates point towards very mild risk aversion. however, once we account for the actual context of our estimation, the results indicate substantial risk aversion. thus, the recent claims in rabin ( a, b) and rabin and thaler ( ) about the limited applicability of expected utility theory when stakes are small, should be taken very seriously. second, not only does the size of the stakes affect risk aversion inferences, but it is also very important in helping us distinguish across behaviors in decision-making problems. as we showed, estimates based on small stakes may yield decision rules that are economically indistinguishable for fully rational and boundedly rational decision-makers. only at high stakes are these behavioral differences clearly observable. finally, a common criticism of field studies data pertains the selection of participating subjects. it is possible that individuals participating in these games may be in some sense different from the rest of the population. the evidence presented here and also in hersch and mcdougall ( ) suggests that this is not the case in terms of observable characteristics. however, our contestant sample is more likely to be selected from individuals who are heavy lottery ticket buyers. thus our results can be interpreted as providing a quasi-lower bound on the mean risk aversion parameter for the population at large. references beetsma, roel, and peter schotman ( ), "measuring risk attitudes in a natural experiment: data from the television game show lingo," mimeo, university of maastricht. binswanger, hans ( ), "attitudes towards risk: theoretical implications of an experiment in rural india," economic journal , - . camerer, colin, eric johnson, talia rymon, and sankar sen ( ), "cognition and framing in sequential bargaining for gains and losses," in frontiers of game theory, - , edited by ken binmore, alan kirman, and piero tani, mit press, cambridge, massachusetts carbone, enrica, and john hey ( ), "a test of the principle of optimality," mimeo, university of york. chou, ray, robert f. engle, and alex kane ( ), "measuring risk aversion from excess returns on a stock index," journal of econometrics , - . friend, i., and m. blume ( ), "the demand for risky assets," american economic review , - . gertner, robert ( ), "game shows and economic behavior: risk-taking on 'card sharks'," quarterly journal of economics , - . gigliotti, gary, and barry sopher ( ), "violations of present value maximization in income choice," theory and decision , - . hersch, philip, and gerald mcdougall ( ), "decision making under uncertainty when the stakes are high: evidence from a lottery game show," southern economic journal , - . kachelmeier, steven, and mohammed shehata ( ), "examining risk-preferences using high monetary incentives: evidence from the people's republic of china," american economic review , - . kahneman, daniel, and amos tversky ( ), "prospect theory: an analysis of decision under risk," econometrica , - . loewenstein, george, and drazen prelec ( ), "negative time preference," american economic review , - . _______________, and richard thaler ( ), "anomalies. intertemporal choice," journal of economic perspectives , , . metrick, andrew ( ), "a natural experiment in jeopardy!," american economic review , - . rabin, mathew ( a), "risk aversion and expected utility theory: a calibration theorem," econometrica , - . _______________ ( b), "diminishing marginal utility of wealth cannot explain risk aversion," forthcoming in choices, values, and frames, edited by daniel kahneman and amos tversky, cambridge university press, new york. _______________, and richard thaler ( ), "anomailes. risk aversion," journal of economic perspectives , - . table discount rates that make a risk neutral contestant indifferent between taking and not taking the gamble regime fully rational contestants in subgame boundedly rational- contestants in subgame fully rational contestants in subgame fully rational contestants in subgame . > . > . . n.a. . . . > . n.a. . . > . . n.a. . . . . n.a. . . . . n.a. . . . . n.a. . . . n.a. n.a. . n.a. n.a. . . . n.a. n.a. . . notes: each regime is indexed by the value of the contestant’s first draw. a . (. ) extension denotes an initial draw equal ot the higher (lower) of the two intermediate dollar prizes. fully rational contestants perform backward induction at each decision node of the tree, whereas boundedly rational- contestants do not perform backward-induction (i.e., they ignore the option value of continuing to play the game.). both fully rational and boundedly rational- contestants fully discount annuity payments. n.a. = non- applicable. table risk aversion parameters that yield equal expected utilities from taking and not taking the gamble panel a: constant absolute risk aversion (cara) utility function. regime fr r = % fr r = % fr r = % br r = % br r = % br r = % br r = % br r = % . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . - . . . . . - . e- . . - . e- - . . . . - . - . . . - . e- - . . . . . - . . . . e- - . e- . panel b: constant relative risk aversion (crra) utility function. regime initial wealth fr r = % fr r = % fr r = % br r = % br r = % br r = % br r = % br r = % . $ , . . . . . . . . . $ , . . . . . . . . . $ , . . . . . . . . . $ , . . . . . . - . . . $ , . - . - . . . . - . . . $ , . - . < - . . - . < - . . $ , . . - . . . . . . notes: each regime is indexed by the value of the contestant’s first draw. a . (. ) extension denotes an initial draw equal to the higher (lower) of the two intermediate dollar prizes. fully rational contestants (fr) perform backward induction at each node of the tree, and discount annuity payments. each contestant may use a different discount rate. we use discount rates of , , and percent. thus, fr, r = % refers to a fully rational contestant that discounts future cash flows using a % rate. boundedly rational- contestants (br ) are the same as the fully rational contestants with the exception that they use a discount rate of zero percent (i.e., they do not discount future payments). while bounded rational- contestants (br ) discount future cash flows, they act myopically in that they ignore the option value of continuing to play the game. bounded rational- contestants (br ) ignore the option value of continuing to play the game and they use a discount rate of zero percent. the cara utility function is specified as follows: u(w)=-e-aw, where w is the individual’s initial wealth and a is the risk aversion parameter. the crra utility function is specified as follows: u(w)=w( -b)/( -b), where w is the individual’s initial wealth and b is the risk aversion parameter. initial wealth refers to the median amount of winnings in phase of the hoosier millionaire. table descriptive statistics regime take gamble? sample size phase winnings median income per-capita income no n.a. n.a. n.a. yes $ , $ , $ , . total $ , $ , $ , no n.a. n.a. n.a. yes $ , $ , $ , . total $ , $ , $ , no $ , missing obs. missing obs. yes $ , $ , $ , . total $ , $ , $ , no $ , $ , $ , yes $ , $ , $ , . total $ , $ , $ , no $ , $ , $ , yes $ , missing obs. missing obs. . total $ , $ , $ , no $ , $ , $ , yes $ , missing obs. missing obs. . total $ , $ , $ , no $ , $ , $ , yes $ , $ , $ , . total $ , $ , $ , total $ , $ , $ , indiana $ , $ , notes: each regime is indexed by the value of the contestant’s first draw. a . (. ) extension denotes an initial draw equal to the higher (lower) of the two intermediate dollar prizes. phase refers to the preliminary phase of the hoosier millionaire, during which contestants randomly draw cash prizes. the contestant amassing the most cash in phase moves onto phase , where she is confronted with the gambles analyzed in this paper. income figures are from the census. median and per-capita incomes are not available for all contestants due to a lack of demographic information. this affects seven contestants in regime , two in regime , and two in regime . “missing obs.” denotes a cell in which demographic information is missing for at least one contestant. n.a. = non- applicable. table estimates of risk aversion coefficients rationality assumption utility function specification fr r = % br r = % br r = % br r = % mean . . . . ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) std. dev. . . . . cara ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) mean . . . . ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) std. dev. . . . . crra w = $ , ( % c.i.) ( . , . ) ( . , . ) ( . , . ) ( . , . ) mean . . . . crra w = $ , std. dev . . . . notes: estimates of risk aversion coefficients are obtained by using maximum likelihood and observations from regimes . and . to estimate the system of equations described on equations ( a)-( b) of the text. each regime is indexed by the value of the contestant’s first draw. a . extension denotes an initial draw equal to the lower of the two intermediate dollar prizes. fully rational contestants (fr) perform backward induction at each node of the tree, and discount annuity payments. each contestant may use a different discount rate. thus, fr, r = % refers to a fully rational contestant that discounts future cash flows using a % rate. boundedly rational- contestants (br ) are the same as the fully rational contestants with the exception that they use a discount rate of zero percent (i.e., they do not discount future payments). while bounded rational- contestants (br ) discount future cash flows, they ignore the option value of continuing to play the game. bounded rational- contestants (br ) ignore the option value of continuing to play the game and they use a discount rate of zero percent. the cara utility function is specified as follows: u(w)=-e-aw, where w is the individual’s initial wealth and a is the risk aversion parameter. the crra utility function is specified as follows: u(w)=w( -b)/( -b), where w is the individual’s initial wealth and b is the risk aversion parameter. two estimations are done for the crra utility function. the first uses an initial wealth (w ) of $ , and the second uses an initial wealth (w ) of $ , . standard deviation is abbreviated std. dev. and confidence interval is abbreviated c.i. table certainty equivalents for various gambles implied by mean estimates of risk aversion coefficients certainty equivalents for a - gamble with payoffs of and x utility function specification rationality assumption x = $ , x = $ , x = $ , x = $ , , fr (r = %) $ . $ , . $ , . $ , . br (r = %) $ . $ , . $ , . $ , . br (r = %) $ . $ , . $ , . $ , . cara br (r = %) $ . $ , . $ , . $ , . fr (r = %) $ . $ , . $ , . $ , . br (r = %) $ . $ , . $ , . $ , . br (r = %) $ . $ , . $ , . $ , . crra w = $ , br (r = %) $ . $ , . $ , . $ , . notes: estimates of risk aversion coefficients are obtained by using maximum likelihood and observations from regimes . and . to estimate the system of equations described on equations ( a)-( b) of the text. each regime is indexed by the value of the contestant’s first draw. a . extension denotes an initial draw equal to the lower of the two intermediate dollar prizes. fully rational contestants (fr) perform backward induction at each node of the tree, and discount annuity payments. each contestant may use a different discount rate. fr (r = %) refers to a fully rational contestant that discounts future cash flows using a % rate. boundedly rational- contestants (br ) are the same as the fully rational contestants with the exception that they use a discount rate of zero percent (i.e., they do not discount future payments). while bounded rational- contestants (br ) discount future cash flows, they ignore the option value of continuing to play the game. bounded rational- contestants (br ) ignore the option value of continuing to play the game and they use a discount rate of zero percent. the cara utility function is defined as follows: u(w)=-e-aw, where w is the individual’s initial wealth and a is the risk aversion parameter. the crra utility function is specified as follows: u(w)=w( -b)/( -b), where w is the individual’s initial wealth and b is the risk aversion parameter. endnotes both card sharks and illinois instant riches involve non-trivial probability calculations, while final jeopardy and lingo involve subjective assessments of one's (or other players') ability to solve a puzzle or answer a question. this change, not widely publicized by the lottery commission, became a heated point of contention in the indiana state elections. all earnings are taxable. given the amounts at stake, as long as marginal tax rates are constant over the annuity payments horizon, decisions should be tax-neutral. we also assume a constant inflation rate over the annuity horizon. thus our discount rates may be interpreted as real discount rates. a further possibility is that contestants are unaware of the annuity system, or that their decision frames are affected by the fact that $ million winners are presented with a large symbolic check for that amount. a variety of other decision problems involving time-delayed payoffs, but not streams of payoffs, actually show that individuals may over-discount future payoffs. however, financial companies openly advertise their readiness to convert lottery annuity payments into lump-sum payoffs at discount rates in the - % range. thus, over-discounting appears unlikely in our problem. the authors conducted an experiment to gain insight on this issue. the hoosier millionaire regime decision problem was given as a final exam question in two different courses: first year mba microeconomics and senior level risk management and insurance. whereas the mba students had been exposed to the concept of backward induction, seniors were mostly unfamiliar with it. our results show that of ( . %) mba students and of ( . %) seniors used backward induction to solve this problem. it is well known that expected utility theory fails to account for some empirical regularities in decision- making under uncertainty. the two most important violations relate to the decision makers’ (a) asymmetric evaluations of gains and losses, and (b) use of decision weights instead of probabilities. the decisions we analyze are such that (a) is absent because there are no losses, and the impact of (b) should be minimal due to the simple probability structure of the gambles. in addition, the recent papers by rabin ( a, b), and rabin and thaler ( ) conclude that risk aversion results based on expected utility theory are most accurate when they pertain to decisions involving large stakes, such as those in our games. as such we feel comfortable staying within the expected utility framework. we concentrate on the initial gambling decisions due to the very limited number of individuals choosing to gamble more than once. in regimes - contestants face a compulsory (exogenous) initial gambling decision. as such we consider only on the (endogenous) decisions after that compulsory stage. in contrast, all initial gambling decisions in regime are endogenous. hersch and mcdougall ( ) note that the median income of illinois lottery players is nearly identical to the state-wide figure. although no figures are presented to back this claim, its accuracy is subject to the same problem regarding the coarseness of variables within census tracts. we also collected other census tract data for our contestant sample, like schooling, household size, and age. we do not show this information due to its coarseness and insignificant inter-regime variation. although we choose a normal distribution for our analysis, the results are qualitatively similar if we use a one-parameter symmetric distribution, such as the logistic distribution. we do not know ex ante if risk aversion parameters are symmetrically distributed among individuals, but we are unaware of hard evidence showing otherwise. strictly speaking, the distribution parameters are over-identified, as there are several regime pairs that could be used to estimate them. however, the choice of regime pairs does not introduce major qualitative variations in the estimates. the similar results we obtained in the logistic case, which is not subject to the pairing problem, reinforces this point. the income statistics are from the census ( figures), whereas the gambling figures are from - . although post- contestant census tract figures are not available, the income variation within this period is not significant enough to meaningfully affect our crra estimates or its interpretation. for instance, median household income in indiana was $ , in , indicating only a modest nominal increase of . % relative to . the high risk aversion parameters estimated by beetsma and schotman ( ) are an exception. given the relatively small average stakes of their gambles (around $ ), it is possible that the subjective elements involved in their decisions may be influencing their estimates. 	between taking and not taking the gamble 	risk aversion parameters that yield equal expected utilities 	r = % 	r = % cucurbit breeding – department of horticultural science 	skip to content 	 	 		 			 				cucurbit breeding 				department of horticultural science 							 		 		 			 				 	 		 			toggle navigation 			 			 			 		 	 	 		 			 					home 	cucumber 	gourd 	melon 	watermelon 	stevia 	publications 	personnel 	wehner 				 								 			 		 	 			 		 	 	 		 							 					 	 		 			 				 					cucurbit breeding at nc state 				 				the cucurbit breeding project at north carolina state university, raleigh is under the direction of todd wehner. project personnel are involved with research that includes the breeding and genetics of cucumber, watermelon, specialty melon, and other cucurbits such as luffa sponge gourd. we also work on stevia breeding. our research focuses on the following areas: collection and evaluation of germplasm, qualitative genetics of important traits, quantitative genetics, selection methods, yield improvement, induced mutation, and resistance to cold, insects, and disease. our work is done primarily in laboratories, greenhouses, drying barns, disease chamber, propagation bench, growth rooms, and experiment stations (clinton, goldsboro, oxford, and whiteville) made available to us by the university. these web pages will provide you with an overview of our work, and will let you access information, articles, and software we have produced. you can also visit the pages for the courses i teach (hs , hs and hs ), and for the plant breeding program at nc state (see the link for related groups). general information 	our short address is https://cucurbitbreeding.wordpress.ncsu.edu/ 	courses 	publications 	cucurbit genetics cooperative 	cucurbit genetics cooperative meetings 	project personnel 	environmental sustainability 	intercrossability of cucurbit species the cucurbits (cucurbitaceae) 	one of top crop families of the angiosperms (besides poaceae, fabaceae, brassicaceae, solanaceae) 	cucumber and its relatives 	cucumis sativus, cucumber 	cucumis sativus var. hardwickii, wild cucumber 	cucumis hystrix (and cucumis hytivus), xishuangbana gourd 	melon and its relatives 	cucumis melo, melon (cantaloupe) 	cucumis melo var. flexuosus, snake melon 	cucumis anguria, west indian gherkin 	squash and pumpkin 	cucurbita pepo, summer squash and pumpkin 	cucurbita maxima, winter squash and pumpkin 	cucurbita moschata, tropical pumpkin 	watermelon, citron, egusi melon, and colocynth 	citrullus lanatus, sweet watermelon 	citrullus amarus, citron 	citrullus mucosospermus, egusi melon 	citrullus colocynthis, colocynth 	gourds and other cucurbits 	benincasa, wax gourd 	luffa, luffa gourd and ridge gourd 	momordica, bitter gourd 	cucurbitaceae taxonomy dendrogram (kocyan, zhang, schaefer & renner, ) 	citrullus taxonomy dendrogram 	synteny of watermelon, melon and cucumber 			 		 	 					 				 		 	 		 			 				 				 					 						 							cucurbit breeding dr. todd wehner department of horticultural science north carolina state university raleigh, nc - phone: . . email: tcwehner@gmail.com cucurbitbreeding.wordpress.ncsu.edu 						 						 													 						 													 						 													 					 					 						 														cucurbit breeding 								copyright © 	· nc state university 	· accessibility 	· privacy 	· university policies 	· log in 						 					 				 			 			 				 					 						nc state university 						 							north carolina state university 							raleigh, nc 							 . . 						 					 				 			 		 		 	 wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ an integrated approach to discretized d modeling of geomechanical properties for unconventional mature field appraisal in the western canadian sedimentary basin original paper - exploration geology an integrated approach to discretized d modeling of geomechanical properties for unconventional mature field appraisal in the western canadian sedimentary basin kalyan saikia • clara e. ikuku • bhabesh c. sarkar received: september / accepted: november / published online: november � the author(s) . this article is an open access publication abstract in mature field appraisal and development, dis- cretized geomechanical property models play a vital role in providing information on in situ stress regime as a guide for placement of directional wells. laboratory methods of measuring these properties, in most cases, take only small samples from consolidated rocks. these isolated samples may not be representative of the entire elastic regime existing in the reservoir owing to sample size. in general, geomechanical studies are performed on a well-by-well basis and then these measurements are used as calibration points to convert d seismic data (if available) to geome- chanical models. however, elastic properties measured this way are restricted to the well location and interpolation across the reservoir may not be always appropriate. to overcome these challenges, this paper describes an inte- grated approach for deriving d geomechanical models of the reservoir by combining results of d geocellular models and basin models. the basin model reconstructs the geologic history (i.e., burial history) of the reservoir by back-stripping it to the original depositional thickness. through this reconstruction, the mechanical compaction, pore pressures, effective stress, and porosity versus depth relationships are established. next, these mechanical properties are discretized into d geocellular grid using empirical formulas via lithofacies model even if no d seismic data are available for the reservoir. the dis- cretization of elastic properties into d grids results in a better understanding of the prevailing stress regimes and helping in design of hydraulic fracturing operations with minimal risks and costs. this approach provides an inno- vative way of determining effective horizontal stress for the entire reservoir through d distribution of elastic properties. keywords reservoir � geomechanics � basin model � unconventional � geostatistics introduction in unconventional tight reservoirs, where hydraulic frac- turing is the key technique for enhanced oil production, a thorough knowledge of distribution of geomechanical properties helps to maximize return on investment. hori- zontal wells with multistage hydraulic stimulation are the primary production strategies during development of the unconventional tight reservoirs. it is therefore important to understand in situ stress conditions such as effective min- imum horizontal stress as it guides the stimulation design which will in turn control the fracture propagation. accu- racy of effective minimum horizontal stress model is pri- marily dependent on knowledge of the spatial distribution of reservoir’s elastic property. in general, for upstream reservoir modeling processes, reservoir geomechanical properties are determined either by ( ) analysis of seismic data (s-wave and p-wave velocities) or ( ) by laboratory measurement of core plugs or ( ) a combination of both techniques. these are con- ventional laboratory methods which may not be repre- sentative of the entire elastic regime existing in the & kalyan saikia kalyansaikia@gmail.com halliburton, piazza drive, richmond, tx , usa clenik petrotech, tremblant way sw, calgary, ab t h a , canada department of applied geology, indian institute of technology (indian school of mines), dhanbad , india j petrol explor prod technol ( ) : – https://doi.org/ . /s - - - http://orcid.org/ - - - http://crossmark.crossref.org/dialog/?doi= . /s - - - &amp;domain=pdf http://crossmark.crossref.org/dialog/?doi= . /s - - - &amp;domain=pdf https://doi.org/ . /s - - - reservoir as in most cases these small samples are taken from consolidated rocks and used to measure geome- chanical properties. furthermore, these measurements are performed on a well-by-well basis on core samples. measurements taken at the core samples are often used in geomechanical analysis as calibration points to convert the d seismic data (if available) to d volumes of geome- chanical properties. previous studies on the gross effect of mechanical properties on productivity of the unconventional reservoir clearly elaborated that the use of only seismic velocities and elastic properties measured on samples are not suffi- cient enough to describe spatial distribution pattern of elastic properties for entire reservoir. havens ( ) car- ried out similar approach for bakken petroleum system to establish the spatial behavior elastic properties and con- cluded that conventional approach of consideration of small core samples plugs in formulating d reservoir geomechanical model is not a practical solution. during fracture modeling of unconventional tight gas reservoirs, gonzalez et al. ( ) and deng et al. ( ) demonstrated that incorporation of d geomechanical model gives a more realistic representation of the orientation and geom- etry of hydraulic fractures compared to traditional way of semi-analytical calculations of fracture geomechanical properties. other published studies on similar topics in the western canadian sedimentary basin reveal that spatial distribution of rock mechanical properties is highly heterogeneous in nature and is result of complex basin- forming (depositional) process (ferdous et al. ; lavoie and séjourné ; tong et al. ). to overcome these challenges, this paper presents a practical approach for deriving and discretizing geome- chanical and other elastic properties in unconventional tight reservoir by integrating results of d geocellular and basin models. this technique provides a relationship between petrophysical properties and elastic rock proper- ties of the reservoir rock, explaining their response to mechanical stresses at reservoir condition. such relation- ship is very important for defining trends in the reservoir description. the current approach also incorporates wire- line logs representing the entire sediment sequence which can be used as input for effective minimum horizontal stress modeling for the entire reservoir. the case study demonstrated here assumes that elastic properties measured conventionally are restricted to the well locations and cannot be interpolated across the reservoir if no d seismic data are available. integrated modeling approach the integrated approach of deriving discretized geome- chanical model of the reservoir can be divided into three main steps. while the first two steps are highly dependent on the input data and selection of appropriate modeling algorithms, the third step is an integration of results of the first two steps. step this step involves creating a high-resolution d geocel- lular reservoir model of the petrophysical properties and depositional environment of the reservoir. initially existing geological, geophysical, and well data are interpreted and incorporated to create a structural framework and empty geocellular grid. next, facies curves derived from the well logs are used to generate cell-wise distribution of lithofa- cies within the reservoir layers employing advanced geo- statistical algorithms. lastly, spatial distribution model of petrophysical property in terms of porosity, permeability, saturation etc., is generated and constrained to the litho- facies model. this helps in capturing geological uncer- tainty while creating d petrophysical property distribution model for the reservoir. while carrying out d geocellular modeling, mainly two geostatistical simulation algorithms, viz. sequential gaussian simulation (sgs) and plurigaussian simulation (pgs) were employed. sequential gaussian simulation (sgs) (deutsch and journel ) is one of the popular geostatistical algorithms used for interpolation of continu- ous variable in reservoir modeling. it was originally introduced as a solution to the smoothing problem of interpolation by kriging. sequential gaussian simulation algorithm reproduces a globally structured two-point statistics using variogram model, whereas kriging provides a best estimate at each location with minimum error vari- ance and ignores estimates made at other locations. thus, with regards to modeling of hydrocarbon reservoirs, sgs algorithms provide more relevant d reservoir models honoring global spatial variations in sample points. the implementation of sequential simulation consists of reproducing the desired spatial properties through the sequential use of conditional distributions (arpat ; dimitrakopoulos and luo ). consider a set of n ran- dom variables z(ua), a = ,…, n defined at n locations ua. the aim is to generate l joint realizations {z(l)(ua),_ = , …, n} with l = , …, l of the n, conditional to n available data and reproducing the properties of a given multivariate distribution. to achieve this goal, the n-point multivariate distribution is decomposed into a set of n univariate con- ditional distributions: j petrol explor prod technol ( ) : – f(u , …, un; z , …, zn|(n)) = f (un; zn|(n ? n - )) f (un - ; zn - |(n ? n - )) ��� f(u ; z |(n ? )) f(u ; z |(n)), where f (un; zn|(n ? n - )) = prob {z(un) b zn|(n ? n - )} is the conditional cumulative distribution function of z(un) given the set of n original data values and the (n - ) realizations z(l)(ua), a = , …, n - of the previously simulated values. this decomposition allows generating a realization by sequentially visiting each node on the simulation grid. multivariate gaussian distribution model is the only model for which the above-mentioned decomposition is analyti- cally available (deutsch ). the conditional probability distribution in a gaussian model is determined using sim- ple kriging of each unknown value at any node uj given the original data and previously simulated values (n ? j - ) (deutsch and journel ). this concept is very conve- niently used in sgs making it one of the popular algo- rithms in reservoir modeling. the whole process of sgs used in current study can be summarized in five steps: . generate a random path through the grid nodes. . visit the first node in that path and use kriging to estimate a mean and standard deviation at that node based on surrounding data (i.e., local conditional probability distribution). . select at random a value from the distribution and set the node value to that number. . include the newly simulated value as part of the conditioning data. . repeat steps – until all grid nodes have a simulated value. truncated gaussian simulation (tgs) (matheron et al. ) and plurigaussian simulation (pgs) (galli et al. ) technique are two popular methods of reservoir modeling workflow for simulating lithotype and facies of sedimentary rocks. tgs technique is best suitable for reservoirs where the lithotypes occur in a sequential order and it helps in defining the geometry and internal archi- tecture of the reservoir rocks (armstrong et al. ). pgs is a natural extension of tgs and capable of handling complex geological situations when the lithotypes occur in complicated relationship rather than simple sequence. the underlying principle of tgs and pgs simulations is to set up one or more simulations of gaussian random fields at every grid point and then use some rules to convert these gaussian values into lithotype indicators (allard et al. ). in case of pgs simulation, generally two gaussian random fields are used to define the lithofacies structure. armstrong et al. ( ) distinguishes four steps in a pgs approach as: . choosing the rock-type rules depending on the types of relations among lithotypes. . estimating the parameter values. two key factors control pgs including the thresholds at which the different gaussians are truncated and the variogram model of the underlying gaussian variable. the proportion of each facies, the ‘‘rock-type’’ rule, and the correlation between the two underlying gaussian random functions determines the thresholds. once the thresholds have been worked out, the variograms and cross-variograms of the indicators are calculated experimentally. . generating gaussian values at the conditioning points. . simulating gaussian values at each grid node with a usual gaussian simulation algorithm. step in this step, a basin modeling study is initiated covering the reservoir area. the basin model reconstructs the geologic history (i.e., burial history) by back-stripping the reservoir to its original depositional condition. through this recon- struction, the mechanical compaction, pore pressures, effective stress, and porosity versus depth relationships are established for the entire reservoir. well data in the form of log curves constitute the main input for basin modeling process in association with other basin-related information such as regional temperature and pressure profile. basin modeling is a technique which allows to under- stand effect of physical and chemical processes in a sedi- mentary basin to generate hydrocarbon. basin modeling techniques help to reconstruct the burial and temperature history of the basin through time and to understand source rock maturation and subsequently hydrocarbon expulsion and migration (hanstschel and kauerauf ; peters ). the basin modeling process dynamically models changes in rock properties through geologic time by numerically solving coupled partial differential equations with moving boundaries on discretized temporal and spa- tial grids. this computation results in many outputs of different rock properties including properties such as vit- rinite reflectance, temperature, effective stress, and poros- ity. in the final stage, the basin models need to be calibrated with existing data such as vitrinite reflectance, temperature, and porosity (alkawai ) to make it more realistic. overall implementation of basin modeling process contains a wide range of mathematical algorithms and methods each of them appropriate for each ‘‘sub-model,’’ and detailed discussion of these goes beyond the scope of this paper. additional information and overview of these numerical methods are provided by press et al. ( ) and j petrol explor prod technol ( ) : – hantschel and kauerauf ( ). in basin modeling, numerical solution of differential equations is fundamental requirement and most challenging, complex and costly process to accomplish. in most of the scenarios, tempera- ture and pressure are modeled with parabolic diffusion equations. multi-dimensional differential equations with complicated geometries are usually solved with the finite element method, whereas finite differences are often used in special cases of approximately one-dimensional prob- lems, such as simplified crustal layer models (hantschel and kauerauf ). step it is the integration step where results from basin modeling study (i.e., distribution of mechanical properties of reser- voir rocks over geological time) are correlated with each cell of geocellular model using empirical relationships (eberhert-phillips et al. ; ingram and urai ; horsrud ; yarus and carruthers ). use of these empirical relationships provided a practical solution in discretizing geomechanical properties in the current study area. other published evidences of use of these relation- ships in basin modeling process include alkawai ( ), szydlik et al. ( ), and schneider et al. ( ). the output of this process is a discretized geomechanical model of the reservoir along with other petrophysical properties. thus, the methodology adopted in this study is a true integration of results and parameters of both geocellular and basin models. diagrammatic representation of the integrated approach is provided in fig. . a step-by-step description of the implementation of the entire modeling process for appraisal of unconventional mature field in western canadian sedimentary basin is explained in the subsequent section of the paper. geological background of the study area the study area is located at the western boundary of sas- katchewan, canada, close to the alberta border in hoosier field (fig. ). previous studies by white ( ) identified two oil pools in the field. the study area falls in the north hoosier oil pool at the west central saskatchewan. in this field, the bakken shale formation is the important forma- tion due to its high hydrocarbon potential. the bakken formation was deposited during the geological age of late devonian and early mississippian. it lies unconformably over the big valley formation (upper devonian) and is conformably overlain by the madison group as shown in fig. (zhang et al. ). deposition of bakken formation occurred through a series of onlap–offlap cycles during the tamaroa sequence (wheeler ). time-equivalent shale units of bakken shale include the exshaw/banff in the alberta basin, the woodford shale in the anadarko basin, the chattanooga in the southern appalachian basin, and the antrim in the michigan basin (meissner ). the madison group which overlies the bakken formation is a thick carbonate sequence of mississippian age. the rocks of madison were deposited in a generally shallow marine setting as indicated by richly fossiliferous rocks and favor for accumulation of petroleum resources (porter ). it has been divided into three formations, viz. lodgepole, mission canyon, and charles. the oldest lodgepole con- sisting limestone and dolomites conformably overlies the bakken formation is a major hydrocarbon-producing horizon (heck ). abnormally high pore pressures in the bakken shales have been attributed to hydrocarbon generation associated with the thermal anomaly in some parts of the basin (price et al. ). hydrocarbon pro- duction in the bakken depends on accurate placement of horizontal wells with multistage fracture stimulations. the effective minimum horizontal stress is a primary control- ling factor for fracture growth. thus, knowledge of distri- bution of elastic properties of the bakken shales is very important to accurately determine the effective minimum horizontal stress (havens ). in terms of wireline log data interpretation, bakken shale is easily recognizable because of the strong contrast in lithology (havens ). the upper and lower shales have unusually high gamma ray readings and high resistivity, while the middle member has a signature similar to clastic and carbonate rocks. data collection and conditioning for the current study, geologic data on hoosier field were sourced and collected from public domain sources (al- berta geological survey; government of saskatchewan). the data gathered for the current study mainly comprised of well location information, interpreted formation tops demarcating geological boundaries, wireline logs, geo- logical maps, etc. no seismic data are available covering the study area to enable understanding of lateral conti- nuity of lithological properties within reservoir layers. thus, facies logs interpreted and derived from available wireline log data were heavily relied upon for establishing spatial continuity of depositional model of the reservoir. three main geological markers (formation tops) repre- senting three most important geological units, i.e., man- nville, detrital, and bakken, were identified from the wireline log interpretations of well data. figure shows a stratigraphic column with the formations of interest. j petrol explor prod technol ( ) : – fig. diagrammatic representation of geocellular model and basin model integration fig. location map of study area showing the hoosier field. modified after o’connell et al. ( ) j petrol explor prod technol ( ) : – d geocellular modeling the interpreted formation tops namely mannville, detrital, and bakken were first used to generate depth structure grids (i.e., maps). these depth structure grids (fig. a) serve as input to create the structural framework of the reservoir with top and base of the framework as mannville and bakken formations. internal stratigraphic unit thick- nesses and their relationship to each other helped in establishing the layering schemes in the structural frame- work. the structural framework constitutes the basic grid skeleton for discretization of reservoir properties. the d structural framework is then divided into blocks of size m m with vertical layer thickness of . m (fig. b, d). a detailed multi-well lithostratigraphic correlation was carried out to establish the continuity of subsurface lithology. multivariate statistical analysis technique is employed to group the log curves into electrofacies groups. each of the electrofacies classes has been assigned a rock class based on the sedimentological interpretation of the wells. lithological classes thus assigned are correlated with each of the wells covering the entire study area. the assigned lithology classes formed the percentage of rock values (fig. c) that would be used to fill in the grid during facies modeling. the well-wise rock interpretation is next combined with petrophysical information to establish its effect on porosity, permeability, and saturation distribution controlling dynamic fluid behavior. log measurement data from wells in terms of petrophysical properties are converted to point sets and then blocked to the grid. a point set is collections of generic points that has d spatial location information, i.e., easting (x), northing (y), and depth (z) and rock property measurement values such a porosity and permeability corresponding to each point locations. while converting a log curve to point set, first measured depth (md) (z) values and corresponding rock property measurements from log curves (such as porosity) are captured in columns in a data file. next, easting (x) and northing (y) information is cap- tured from well head and added to each data points as two separate columns. thus, the resultant point set data file has series of columns starting with x, y, z and followed by rock property values. generally, each of these data points is taken at the scale of actual log measurement, i.e., . feet (approximately cm). after creation of the point set, the next step is to block the point set to the grid resolution (shepherd ). usually, the vertical dimensions of geo- cellular grids are larger than the vertical sampling interval of log curves (or the point set). as a result, each geocells passing through well locations carry multiple points. in order to use the point information for modeling and inter- polation, the multiple points within each cell need to convert to one point per grid cells using an averaging technique such as arithmetic, geometric, and harmonic for continuous properties. in case of discrete properties (such as facies or lithology), the averaging approach is ‘‘most of,’’ i.e., most commonly occurring (mode) lithology within the points occurring in a single geocells (cannon ). this process is referred as blocking of point sets to the geocellular grid. lithotype assignment and vertical proportion curves in terms of three main lithotypes, i.e., dolomitic sandstone, siltstone, and shale are generated. these lithotype proportion curves resulted into lithotype proportion maps (fig. ) which are then incorporated in generation of facies distribution model for the reservoir. variogram models have been constructed to establish data stationarity and spatial correlation of input data points over the study area. establishing stationarity is expedient for an implicit assumption of a geostatistical calculation in the current study. the absence of a nugget effect in the variograms showed that there is no randomness in the sampled well population. the spatial position of well locations produced a large spatial correlation distance for each of the variables. this correlation allows to progress fig. stratigraphic column of study area j petrol explor prod technol ( ) : – with a geostatistical approach to extrapolate the measured data for the entire grid cells. plurigaussian simulation (pgs) algorithm (armstrong et al. ) was employed with two variogram models for creating d lithofacies distribution model. in pgs simu- lation, the controlling component is lithotype rule, which defines the depositional environment (armstrong et al. ). in addition to lithotype rules, the lithotype proportion maps resulted from lithotype curves provided a background guideline for cell-wise facies distribution. pgs is one of the best techniques in geologically complex depositional environment modeling. next, sequential gaussian simulation (sgs) (deutsch and journel ; deutsch ) was used for creating d spatial distribution model of petrophysical properties (such as porosity and permeability) employing variograms created for each of fig. a depth structural map from tops, b d grid layers, c location of wells in the study area and d d block configuration fig. vertical proportion curve and lithotype proportion map showing vertical and areal distribution of rock types j petrol explor prod technol ( ) : – these properties. while creating the porosity model, pre- viously created facies model was used to constrain or bias reducing geological uncertainty of modeling of petro- physical properties. distribution of facies and associated porosity in the reservoir is shown in fig. . building d basin model the concept of basin modeling in general refers to inte- gration of geological, physical, and geochemical processes in a sedimentary basin to simulate sediment burial, com- paction, heat transfer, hydrocarbon generation, and migration and entrapment processes within it. basin mod- eling reconstructs the pressure regime of the basin during sediment deposition and predicts zone of overpressure. this process also delineates hydrocarbon migration path- ways and identifies structural and stratigraphic traps. basin modeling can be performed in d (one-dimensional), d (two-dimensional) or d (three-dimensional) to simulate the burial and thermal histories of the basin with increasing complexities. if the area is relatively small, such that thermal, pressure, and other effects are uniform over the study area, then a d basin model might be sufficient (yarus and carruthers ). if the basin development processes vary significantly over the area, a d study is warranted. in the current study, the necessity of d/ d basin model is not felt due to relatively uniform thermal and pressure effect on the basin-forming process in the study area. comparison d basin model results from few places within the study area also revealed that important parameters controlling the basin-forming process do not vary much. thus, the current study was restricted to con- struction of d basin model and uses the results for derivation of geomechanical properties. the d basin model uses rock and fluid information to evaluate source rocks and thermal histories of basin along a single direction, i.e., vertical. in this process, layers in the d model are sequentially back-striped to its original depositional thickness resulting into effective stress– porosity relationships for all lithologic layers. the lithologies are generally assigned to the model layers by zones using petrophysical logs and other geological infor- mation. these lithologies provide thermal conductivity and heat capacity for each layer. in the current study, the input parameters used in developing d basin model of the reservoir include: • geologic framework in terms of layer thicknesses and ages • rock properties (i.e., lithology) for each framework layers • well data information such as temperature and pressure • identification of source layers with assignment of kerogen types • regional geothermal gradient of the basin figure shows the output of basin modeling process in terms of variation of elastic properties and pressure regim while depositing the sediments over geological periods. modeling of the parameters is governed by the rates of pressure generation (e.g., compaction disequilibrium) and pressure dissipation (e.g., as controlled by permeability) during basin-forming process. another important output of d basin modeling process is burial history or geohistory plot of the reservoir as shown in fig. . geohistory plots indicate how sediments were deposited over geological time in the basin providing thickness variation information of layers. in other words, it shows the time versus thickness relationship for all the layers including the layers that have been eroded or periods of non-deposition (hiatus). based on the lithology information of the d model inputs, depth–porosity relationship is also calculated while estab- lishing burial history. the depth–porosity relationship is then converted to calculate the effective stress regime and temperature profile of each layer through the geological time. the effective stress thus derived for each layer of the basin provides crucial information for calculation of geomechanical properties of the reservoir rock. fig. distribution of facies (rock types) and porosity (facies-constrained) j petrol explor prod technol ( ) : – fig. output of d basin modeling process in terms of reservoir elastic properties and pressure regim fig. geohistory plot (burial history)—output from d basin modeling process j petrol explor prod technol ( ) : – model integration the final stage of the current integrated approach is to combine the result of classical d basin model (burial history) with geostatistical d geocellular model to gen- erate discretized geomechanical model of the study area. lithology identifier of each cell from d facies model provides the link of this integration. each element of burial history model has also been assigned a lithological iden- tifier to define ( ) mechanical compaction, i.e., porosity versus depth relationships, ( ) pore pressure relationships, i.e., porosity versus effective stress, ( ) fluid transport properties, i.e., porosity versus permeability and capillary threshold pressure, and ( ) matrix properties such as ther- mal conductivity. the lithology identifiers available in both the models enable populating layer-wise distribution of effective stress for each cell in the d geocellular grid (fig. ). the effective stress estimation is obtained during the process of burial history modeling of the sedimentary basin. the burial (or geo) history model provides infor- mation on rate of sedimentation including depth and age of deposition. while calculating effective stress regime of the basin, we assume that sedimentation and subsidence is the primary cause of pressure and stress; stresses generally increase with depth and rock stress generally interacts negatively with porosity and compaction (hantschel and kauerauf ). in d basin modeling approach, terza- ghi’s ( ) effective stress concept is widely used as the fundamental relationship for effective stress modeling (hubert and rubey ). effective stress ðr Þ ¼ total stress ðrÞ�pore pressure pð Þ: ð Þ in this relationship, lithostatic pressure (equaling the sediment weight) is considered as total stress and additional stresses due to compaction or extensional forces are neglected (hantschel and kauerauf ). biot ( ) established the theory of poro-elasticity for rocks and introduced biot’s coefficient in the definition of the effective stress (schneider et al. ). r ¼ r � a p; where a is biot’s coefficient: ð Þ in these models, pore pressure of formation is related to incomplete mechanical compaction and a fixed relation between porosity reduction and sediment compaction is assumed (hantschel and kauerauf ). studies by various other researchers established a number of empirical relationships of porosity versus depth versus compaction. these relationships show that changes in porosity with increasing depth is a function of lithology of the sediments. thus, in basin modeling process these lithology-wise empirical relationships are used to satisfy the depth–porosity function and pore pressure calculation. the complete mathematical explanations of these relationships and numerical solutions to complex equations can be obtained from hantschel and kauerauf ( ) and schneider et al. ( ). once the effective stress distribution model is generated for the entire reservoir geocellular grid, rock mechanical properties such as vp, vs, ucs, bri, ocr, young’s modulus, poison’s ratio, and bulk modulus are calculated at each cell locations of geocellular grid using empirical formula as provided below (eberhert-phillips et al. ; ingram and urai ; horsrud ; yarus and car- ruthers ): vp ¼ : � : / � : p c þ : pe � e� : pe � � ; ð Þ vs ¼ : � : / � : p c þ : pe � e� : pe � � ; ð Þ log ucs ¼ � : þ : log : vp � ð Þ; ð Þ ocr ¼ max pe=pe; ð Þ brit ¼ ucs=ucsnc; ð Þ e ¼ qv s v p � v s � � = v p � v s � �h i ; ð Þ m ¼ : v p � v s � � = v p � v s � �h i ; ð Þ k ¼ q v p � = v s � � ; ð Þ where vp = p-wave velocity, vs = s-wave velocity, / = porosity, c = clay content, pe = effective stress, ucs = unconfined compressive strength, ocr = over consolidation ratio, brit = brittleness, nc = uncon- fined, q = density, e = young’s modulus, m = poisson’s ratio, k = bulk modulus. for each of the calculations lithology, porosity and effective stress serve as main input parameter. figure displays a representative view of the geocellular grid with discretized geomechanical property derived from the inte- gration of result of basin and geocellular modeling. results and discussion well coverage in the study area is not adequate as revealed in fig. c. the wells are mostly clustered in the central and north-eastern part. furthermore, in the study area conven- tional cores were collected in only five wells out of wells made available. thus, physical measurement and interpolation of geomechanical properties of these core samples were not considered as reliable source of infor- mation for future drilling and investment decision. addi- tionally, due to non-availability of d seismic data, j petrol explor prod technol ( ) : – interpolation of well-based geomechanical properties beyond well control could not be validated through seismic attribute correlation and integration as secondary input. as a result, attempts to use the conventional approach of using well core sample analysis-based geomechanical property interpolation in the study area provided inappropriate information for full field appraisal and development plan- ning. however, use of current geostatistics-based modeling approach explained in this study helped to nullify these restrictions and provided a full field geomechanical prop- erty distribution model with the integration of parameters from basin modeling approach. such integrated models are highly flexible and appropriate for full field development with reliable distribution of geomechanical property in d space of unconventional reservoir. the model developed in the current study clearly indi- cates that elastic moduli decrease with increasing porosity; therefore, such models would aid in the understanding of the distribution of brittle versus ductile zones in the reservoir. statistical analysis in terms of correlation coefficient of young’s modulus and porosity reveals a value of - . . figure displays a plot of young’s modulus and porosity at k-level of d reservoir grid. fig. distribution of geomechanical properties in the final integrated reservoir grid fig. plot showing correlation of young’s modulus and porosity for k-level of d reservoir grid j petrol explor prod technol ( ) : – based on the spatial distribution pattern of geomechanical properties, the lower part of the bakken formation in the study area has been identified as zone of prime importance with favorable values of geomechanical properties to maximize hydrocarbon production. in the absence of d seismic coverage in the study area, such information aids operators tremendously for effective design of depletion strategies with minimal geological risks and associated costs. furthermore, for a variety of other subsurface operations, including flow simulation, history matching, fluid typing, and hydrocarbon production, these models serve as main input. understanding the burial history of the reservoir is a step forward in modeling unconventional resources (yarus and carruthers ). as part of full field development project, a realistic picture of the depositional history of the reser- voir through geological time provides a more solid foun- dation to build a knowledge base for economic forecasting and maximum return of investment. in the unconventional upstream oil and gas industry, spatial distribution modeling of rock mechanical properties of matured reservoir has always been a major challenge. however, d spatial dis- tribution of model of geomechanical properties created in the current study provides an easy way of knowing pre- vailing rock elastic properties and stress regimes of the reservoir rocks. conclusions field development planning, in the current study area, involves the use of horizontal wells with multistage hydraulic stimulation in shale sweet spots as the primary production strategies. knowledge of in situ stress condi- tions and facies is therefore very important for these operations as it guides the stimulation design which will in turn control the fracture propagation. earlier studies on bakken formation clearly elaborated the use of velocities and strains measured on samples from the core which are not sufficient enough to optimize field development strat- egy using hydraulic fracturing. this is because small sample plugs or slabs may not be always sufficient to describe elastic properties for the entire reservoir. for instance, vp measured from one core sample may not be very diagnostic of lithology as it cannot be used to map shales and sandstones across a reservoir with accuracy. the current integrated approach incorporates wireline logs in the form of hard data covering entire sediment sequence. this provides a true representation of the sub- surface geological condition of the reservoir even if there are no d seismic data available. in this approach, use of geostatistical techniques results into highly accurate d facies model resembling real geological condition of deposition of sediments. the geostatistical techniques via spatial covariance models of sample values helped in quantifying and mapping of geological heterogeneity in sediment deposition in unconventional reservoirs. addi- tionally, conditional simulation algorithms used in geo- statistical modeling approach honor the sample values at location of measurements and produce minimum biasness of interpolation of rock properties from well data. thus, results generated from final integrated model are not only reliable, but also capable of providing accurate local error of estimation (i.e., heterogeneity) within them. similar studies made elsewhere (yarus et al. ; yarus and yarus ; yarus and carruthers ) indicated how combining geostatistics with basin modeling techniques accurately identifies sweet spots in unconventional reser- voirs. verification and comparison of results obtained from such integrated models with actual and blind wells pro- vided high accuracy and reliability score with minimum deviation of results. such convincing results are also true validation of consistency of result of current integrated technique over conventional approach. traditional modeling methods for identifying good and poor reservoir quality in shales have not been successful as reported by many authors. it is mainly due to lack of geospatial understanding of shales properties and difficulty of quantifying their heterogeneity. however, the current integrated geostatistics-based geomechanical models with higher degree of confidence show how the velocity and elastic moduli are decreasing with increasing porosity throughout the reservoir. as a result, in the current study area the concerned field team was able to identify target locations or sweet spots with high degree of accuracy within the shale for fracturing and completion. further- more, these models can be used alongside other reservoir characterization methods in shale play workflows to aid in the planning of unconventional well bores for enhanced production. open access this article is distributed under the terms of the creative commons attribution . international license (http:// creativecommons.org/licenses/by/ . /), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the creative commons license, and indicate if changes were made. references alberta geological survey. http://ags.aer.ca/. accessed july alkawai w ( ) integrated basin modeling with seismic technol- ogy and rock physics. m.sc. thesis, stanford university, stanford allard d, d’or d, biver p, froidevaux r ( ) non-parametric diagrams for plurigaussian simulations of lithofacies. in: th international geostatistical congress, oslo, norway, pp – j petrol explor prod technol ( ) : – http://creativecommons.org/licenses/by/ . / http://creativecommons.org/licenses/by/ . / http://ags.aer.ca/ armstrong a, galli ag, loch gl, geoffroy f, eschard r ( ) plurigaussian simulations in geosciences. springer, berlin, p arpat gb ( ) sequential simulation with patterns. phd thesis. stanford university, stanford biot ma ( ) general theory of three-dimensional consolidation. j appl phys : – cannon s ( ) petrophysics: a practical guide. wiley, london, pp – deng h, aguilera r, settari t ( ) an integrated workflow for reservoir modeling and flow simulation of the nikanassin tight gas reservoir in the western canada sedimentary basin. spe annual technical conference and exhibition, denver, colorado, usa. spe # -ms. doi: . / -ms deutsch cv ( ) geostatistical reservoir modeling. oxford university press, oxford deutsch cv, journel ag ( ) gslib: geostatistical software library and user’s guide, nd edn. oxford university press, oxford dimitrakopoulos r, luo x ( ) generalized sequential gaussian simulation on group size m and screen-effect approximations for large field simulations. math geol ( ): – eberhert-phillips d, han d-h, zoback md ( ) empirical relationships among seismic velocity, effective pressure, poros- ity, and clay content in sandstone. j geophys ( ): – ferdous h, bergman k, qing h ( ) viking lowstand deposits in west central saskatchewan: depositional model for the reservoir units in dodsland-hoosier area, saskatchewan, canada. aapg search and discovery article # galli a, beucher h, le loc’h g, doligez b ( ) the pros and cons of the truncated gaussian method. in: armstrong m, dowd pa (eds) geostatistical simulations. kluwer, dordrecht, pp – gonzalez l, nasreldin g, rivero j, welsh p, aguilera r ( ) d modeling of multistage hydraulic fractures and two-way-cou- pling geomechanics/fluid-flow simulation of a horizontal well in the nikanassin tight gas formation, western canada sedimentary basin. spe reserv eval eng ( ): – government of saskatchewan. http://www.ir.gov.sk.ca/ accessed july hantschel t, kauerauf ai ( ) fundamentals of basin and petroleum systems modeling. springer, berlin havens j ( ) mechanical properties of the bakken formation. m.s. thesis, colorado school of mines, golden heck tj ( ) depositional environments and diagenesis of the mississippian bottineau interval (lodgepole) in north dakota: university of north dakota unpublished m.s. thesis horsrud p ( ) estimating mechanical properties of shale from empirical correlations. spe drilling and completion conference, pp – (spe # ) hubert mk, rubey wm ( ) role of fluid pressure in mechanics of overthrust faulting. bull geol soc am : – ingram gm, urai jl ( ) top-seal leakage through faults and fractures: the role of mudrock properties. geol soc lond spec publ : – lavoie d, séjourné s ( ) geomechanical properties of the upper ordovician macasty shale and its caprock, anticosti island: a regional evaluation for a promising tight oil play. aapg search and discovery article # matheron g, beucher h, de fouquet c, galli a, guerulot d, ravenne c ( ) conditional simulation of the geometry of fluvio- deltaic reservoirs. spe annual technical conference and exhibi- tion, dallas, texas. spe# meissner ff ( ) petroleum geology of the bakken formation, williston basin, north dakota and montana. in: proceedings of williston basin symposium, montana geological society, pp – o’connell sc, wood jm, robinson sp ( ) an ancient tidal sand ridge complex in the bakken formation (fammenian-tour- naisian) of southwestern saskatchewan. geocanada core workshop, calgary, alberta. http://webapp .dlib.indiana.edu/ virtual_disk_library/index.cgi/ /fid /pdf/ .pdf peters ke ( ) getting started in basin and petroleum system modeling. aapg cd-rom # , aapg datapages porter jw ( ) madison complex of southeastern saskatchewan and southwestern manitoba. j alberta soc petrol geol ( ): – press wh, teukolsky sa, vetterling wt, flannery bp ( ) numerical recipes in c??, nd edn. cambridge university press, cambridge price lc, ging t, daws t, love a, pawlewicz m, anders d ( ) organic metamorphism in the mississippian-devonian bakken shale, north dakota portion of the williston basin. in: wood- ward j, meisner ff, clayton jl (eds) hydrocarbon source rocks of the greater rocky mountain region. rocky mountains association-geologists, denver, pp – schneider f, potdevin jl, wolf s, faille i ( ) mechanical and chemical compaction model for sedimentary basin simulators. tectonophysics : – shepherd m ( ) oil field production geology: aapg memoir , pp – szydlik t, helgesen hk, brevik i, prisco gd, clark sa, leirfall ok, duffaut k, stadtler c, cogan m ( ) geophysical basin modeling: methodology and application in deepwater gulf of mexico. interpretation ( ):sz –sz terzaghi k ( ) die berechnung der durchlssigkeitsziffer des tones aus dem verlauf des hydrodynamishen spannungserscheinungen. szgber akad wiss vieu math naturwiss klasse ila ( – ): – tong a, chi g, pedersen pk ( ) sequence stratigraphy and preliminary diagenetic study of the lower cretaceous viking formation, hoosier area, west-central saskatchewan. summ investig sask geol surv : wheeler he ( ) post-sauk and pre-absaroka paleozoic strati- graphic patterns in north america. aapg bull : – white wi ( ) geology and petroleum accumulations of the north hoosier area, west central saskatchewan. department of mineral resources, geological science branch, sedimentary geology division, report no. yarus jm, carruthers d ( ) modeling unconventional resources with geostatistics and basin modeling techniques: characterizing sweet-spots using petrophysical, mechanical, and static fluid properties. aapg search and discovery article # yarus jm, yarus jm ( ) modeling with geostatistics, basin modeling techniques, e & p magazine. http://www.epmag.com/ modeling-geostatistics-basin-modeling-techniques- #p= full yarus jm, mohsenian e, yarus jm ( ) combing geostatistical reservoir modeling with classical basin modeling techniques to identify sweet-spots in unconventional reservoirs. aapg data- pages/search and discovery article # zhang l, buatois la, knaust d ( ) sedimentology, ichnology and sequence stratigraphy of the upper devonian-lower missis- sippian bakken formation in eastern saskatchewan. bull can pet geol ( ): – publisher’s note springer nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. j petrol explor prod technol ( ) : – https://doi.org/ . / -ms http://www.ir.gov.sk.ca/ http://webapp .dlib.indiana.edu/virtual_disk_library/index.cgi/ /fid /pdf/ .pdf http://webapp .dlib.indiana.edu/virtual_disk_library/index.cgi/ /fid /pdf/ .pdf http://www.epmag.com/modeling-geostatistics-basin-modeling-techniques- % p% dfull http://www.epmag.com/modeling-geostatistics-basin-modeling-techniques- % p% dfull http://www.epmag.com/modeling-geostatistics-basin-modeling-techniques- % p% dfull 	an integrated approach to discretized d modeling of geomechanical properties for unconventional mature field appraisal in the western canadian sedimentary basin 	abstract 	introduction 	integrated modeling approach 	step 	step 	step 	geological background of the study area 	data collection and conditioning 	 d geocellular modeling 	building d basin model 	model integration 	results and discussion 	conclusions 	open access 	references wp-p m- .ebi.ac.uk params is empty sys_ exception wp-p m- .ebi.ac.uk no params is empty exception params is empty / / - : : if (typeof jquery === "undefined") document.write('[script type="text/javascript" src="/corehtml/pmc/jig/ . . /js/jig.min.js"][/script]'.replace(/\[/g,string.fromcharcode( )).replace(/\]/g,string.fromcharcode( ))); // // // window.name="mainwindow"; .pmc-wm {background:transparent repeat-y top left;background-image:url(/corehtml/pmc/pmcgifs/wm-nobrand.png);background-size: auto, contain} .print-view{display:block} page not available reason: the web page address (url) that you used may be incorrect. message id: (wp-p m- .ebi.ac.uk) time: / / : : if you need further help, please send an email to pmc. include the information from the box above in your message. otherwise, click on one of the following links to continue using pmc: search the complete pmc archive. browse the contents of a specific journal in pmc. find a specific article by its citation (journal, date, volume, first page, author or article title). http://europepmc.org/abstract/med/ cucurbit breeding – department of horticultural science 	skip to content 	 	 		 			 				cucurbit breeding 				department of horticultural science 							 		 		 			 				 	 		 			toggle navigation 			 			 			 		 	 	 		 			 					home 	cucumber 	gourd 	melon 	watermelon 	stevia 	publications 	personnel 	wehner 				 								 			 		 	 			 		 	 	 		 							 					 	 		 			 				 					cucurbit breeding at nc state 				 				the cucurbit breeding project at north carolina state university, raleigh is under the direction of todd wehner. project personnel are involved with research that includes the breeding and genetics of cucumber, watermelon, specialty melon, and other cucurbits such as luffa sponge gourd. we also work on stevia breeding. our research focuses on the following areas: collection and evaluation of germplasm, qualitative genetics of important traits, quantitative genetics, selection methods, yield improvement, induced mutation, and resistance to cold, insects, and disease. our work is done primarily in laboratories, greenhouses, drying barns, disease chamber, propagation bench, growth rooms, and experiment stations (clinton, goldsboro, oxford, and whiteville) made available to us by the university. these web pages will provide you with an overview of our work, and will let you access information, articles, and software we have produced. you can also visit the pages for the courses i teach (hs , hs and hs ), and for the plant breeding program at nc state (see the link for related groups). general information 	our short address is https://cucurbitbreeding.wordpress.ncsu.edu/ 	courses 	publications 	cucurbit genetics cooperative 	cucurbit genetics cooperative meetings 	project personnel 	environmental sustainability 	intercrossability of cucurbit species the cucurbits (cucurbitaceae) 	one of top crop families of the angiosperms (besides poaceae, fabaceae, brassicaceae, solanaceae) 	cucumber and its relatives 	cucumis sativus, cucumber 	cucumis sativus var. hardwickii, wild cucumber 	cucumis hystrix (and cucumis hytivus), xishuangbana gourd 	melon and its relatives 	cucumis melo, melon (cantaloupe) 	cucumis melo var. flexuosus, snake melon 	cucumis anguria, west indian gherkin 	squash and pumpkin 	cucurbita pepo, summer squash and pumpkin 	cucurbita maxima, winter squash and pumpkin 	cucurbita moschata, tropical pumpkin 	watermelon, citron, egusi melon, and colocynth 	citrullus lanatus, sweet watermelon 	citrullus amarus, citron 	citrullus mucosospermus, egusi melon 	citrullus colocynthis, colocynth 	gourds and other cucurbits 	benincasa, wax gourd 	luffa, luffa gourd and ridge gourd 	momordica, bitter gourd 	cucurbitaceae taxonomy dendrogram (kocyan, zhang, schaefer & renner, ) 	citrullus taxonomy dendrogram 	synteny of watermelon, melon and cucumber 			 		 	 					 				 		 	 		 			 				 				 					 						 							cucurbit breeding dr. todd wehner department of horticultural science north carolina state university raleigh, nc - phone: . . email: tcwehner@gmail.com cucurbitbreeding.wordpress.ncsu.edu 						 						 													 						 													 						 													 					 					 						 														cucurbit breeding 								copyright © 	· nc state university 	· accessibility 	· privacy 	· university policies 	· log in 						 					 				 			 			 				 					 						nc state university 						 							north carolina state university 							raleigh, nc 							 . .