BIO-AGRICULTURAL LIBRARY UNIVERSITY OF CALIFORNIA RIVERSIDE, CALIFORNIA 92502 RESEARCHES NORTH AMERICAN ACRIDIID/E ALBERT PITTS MORSE Research Assistant, Carnegie Institution of Washington PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON 1904 FIG. i Eupatorium thicket amid scattered deciduous trees on a ridge " bald " traversed by a cattle-path, at east end of Roan Mountain, North Carolina. FIG. 2 "Laurel bald" on the slope of Grandfather Mountain, North Carolina, forming the thicket habitat of Podisma glacialis variegata and Melanoplus sylvestris. Berry-pickers camped beside the Yonahlossee road. RESEARCHES NORTH AMERICAN ACRIDIID/E ALBERT PITTS MORSE Research Assistant, Carnegie Institution of Washington Library PUBLISHED BY THE CARNEGIE INSTITUTION OF WASHINGTON 1904 CARNEGIE INSTITUTION OF WASHINGTON PUBLICATION No. 18 PRESS OF W. F. ROBERTS COMPANY WASHINGTON CONTENTS. Page Introduction 7 Biological importance of the Southeastern United States 8 Localities at which collecting was done, with dates of visit 9 Field work 10 Climatal conditions affecting results n Life zones of region visited n Number of broods of Acridiidae 12 Life histories with reference to season 12 Zonal distribution of genera of Acridiidse of eastern North America... 12 Locust societies and habitats 14 Comparison of campestral and sylvan locusts 19 Macropterous and brachypterous locusts 20 Brachypterism in other Orthoptera 21 "Burning-over" as a factor in distribution 22 Hybrids, varieties, and other topics 22 Species of economic importance. 23 List of species and localities from which collected 24-55 3 ILLUSTRATIONS. PLATES. Facing Plate i. Fig. I. Ridge "bald" at east end of Roan Mountain I Fig. 2. "I/aurel bald" on slope of Grandfather Mountain, habitat of Podisma glacialis variegata and Melanoplus sylvestris i Plate 2. Fig. i. Habitat of Trimerotropis saxatilis, Stone Mountain, Ga 56 Fig. 2. Habitat of Trimerotropis saxatilis, Sand Mountain Plateau, near Trenton, Ga 56 Plate 3. Fig. i. Habitat of Trimerotropis dtrina and T. maritima, on sea- beach at Cape Henry, Va 56 Fig. 2. Habitat of Trimerotropis citrina, on sandy river-wash in Piedmont region, Morganton, N. C 56 Plate 4. Fig. i. The Florida strand, habitat of Scirtetica picta, etc 56 Fig. 2. The Florida strand, habitat of Orphulella olivacea, etc 56 Plate 5. Fig. i. The Florida strand, habitat of Hesperotettix pratensis 56 Fig. 2. Pine barrens, habitat of Gymnoscirtetes pusillus, etc 56 Plate 6. Fig. i. Type locality of Melanoplus strumosus 56 Fig. 2. Type locality of Melanoplus symmetricus 56 Plate 7. Fig. i. Normal sylvan habitat of Melanoplus amplectens 56 Fig. 2. Extended habitat of Melanoplus amplectens 56 Plate 8. Fig. i. Summit "balds "east of Roan Mountain, North Carolina... 56 Fig. 2. Protective coloration 56 TEXT FIGURES. Page Fig. i. Map showing the route traveled 10 2-10. Cerci of male Melanoplus: 2, M. sylvestris; 3, M. celatus; 4, M. carnegiei; 5, M. divergens; 6, M, similis; 7, M* tribulus; 8, M. devius; 9, M. deceptus; 10, M. symmetricus 46 11. M. decoratus: Lateral view of end of abdomen of male 51 12. M. strumosus: Lateral view of end of abdomen of male 51 13. M. strumosus: Dorsal view of end of abdomen of male 51 RESEARCHES ON NORTH AMERICAN ACRIDIID^ By ALBERT P. MORSE, Research Assistant, Carnegie Institution of Washington. INTRODUCTION. The work on which report is here made was done under a grant from the Carnegie Institution of Washington and was directed toward the acquisition of facts of faunal, biological, and morpho- logical value concerning the Acridian fauna of the southeastern United States, a portion of the country hitherto largely unexplored with reference to this group of insects. A study of the general topography, based, when possible, on the maps of the United States Geological Survey, led to a plan of operations in accordance with variations in the life history of differ- ent species. This plan involved an examination of the different life zones of the region at as early and as late dates as practicable, in order to secure earlier and later maturing forms, since it is impos- sible, with our present knowledge, to distinguish many of the species with certainty save in the adult state. About 60 widely separated localities were examined in 68 days during a trip of 3,000 miles through the region under inspection. Approximately 6,000 specimens were secured in this time, repre- senting 90 species of Acridiidse (of which one-sixth were new to science), together with notes of great biological interest. Many important data relating to the biology and distribution of the group were secured ; but in some cases, owing to the character of the trip a rapid reconnaissance over a wide extent of territory for the purpose of securing general information the results should be regarded as tentative until further detailed studies can be made. This report includes a general account of the investigation and a statement of the results secured; the data on which the general- ized statements are based, in the form of an annotated list of the species taken, with locality, habitat, and date of capture ; brief notes on the geographical and zonal distribution of the genera of North American Acridiidae ; and contributions to locust biology in 8 NORTH AMERICAN ACRIDIID^E. the shape of a classification of locust habitats and societies and an explanation of brachypterism in Orthoptera. Acknowledgments are due, primarily, to the Carnegie Institu- tion for the means wherewith to pursue this investigation ; secondly, to Mr. Samuel Henshaw, of the Museum of Comparative Zoology, for the opportunity to study critically the Scudder collection, with- out access to which certainty of determination would in many cases have been impossible; and finally, to numerous fellow-scientists, travelers, and observers, for their cordial interest and assistance in various ways. Among these I desire epecially to mention Mr. J. H. Emerton, of Boston; Messrs. Sherman and Brimley Bros., of Raleigh, and Boynton, of Biltmore, North Carolina, for information regarding conditions and localities ; and Messrs. Fernald and Greenman, of Cambridge, Massachusetts, for determination of plant specimens. BIOLOGICAL IMPORTANCE OF THE SOUTHEASTERN UNITED STATES. The chief biological importance of the Southeastern United States, comprising Virginia, North and South Carolina, Georgia, Florida, Alabama, eastern Tennessee, and West Virginia, is con- nected with two facts : First, this region served during the Glacial Epoch as a refuge for boreal forms of life which had been pushed southward by the climatal conditions of the Ice Age, and at the close of that period it became the center of dispersal whence these forms were able to restock the opening country at the north. Second, during this later period its lowland plains served, and probably continue to serve, as a highway of dispersal for austral forms entering the country from the south and southwest, many of which have penetrated far into the heavily glaciated region of the Northern States. Notwithstanding its great biological importance, this region has not received the attention scientifically which it deserves, save with reference to certain groups and in certain sections, such, for instance, as peninsular Florida, which, owing to its southern posi- tion and consequent attractiveness, both biologically and as a winter resort, has received considerable attention and study. As a matter of fact, the Acridian fauna of no area in the United States of equal size and biological importance was so little known at the beginning of the year. For this reason it was decided to spend the summer in a general reconaissance trip through the region, securing as many facts relative to systematic facies, geographical, zonal, and seasonal distribution, habitats and environmental conditions, life histories, etc., as possible. NORTH AMERICAN ACRIDIID^E. 9 LOCALITIES AT WHICH COLLECTING WAS DONE. A list of the points at which collecting was done, with their elevation, if known, especially in the mountainous regions, and the dates of visit, is subjoined. Virginia : Appomattox, Appomattox Co., Sept. 6; 800 feet. Cape Henry, Princess Anne Co., July 2, 4, Sept. 7. Hickory, Norfolk Co. , July 3. Norfolk, Norfolk Co., Sept. 8. Roanoke, Roanoke Co., Sept. 6; 1000 to 1600 feet. Virginia Beach, Princess Anne Co., July 2, 4, Sept. 7. Wytheville, Wythe Co., Sept. 4, 5; 2200 to 3500 feet. North Carolina: Asheville, Buncombe Co., July 21, 22 ; 2000 to 3000 feet. Balsam, Jackson Co., July 23, 24; Aug. 19, 20 ; 3000 to 6200 feet. Blowing Rock, Watauga Co., July 19; 3800 feet. Cranberry, Mitchell Co., July 14, Aug. 28; 3200 feet. Eure, Gates Co., July 5, 6. Governor Island, Swain Co., Aug. 20; 1800 feet. Grandfather Mt., Mitchell and Cald- well Cos., Aug. 29; 4800 to 5500 feet. Greensboro, Guilford Co., July 10. Lenoir, Caldwell Co., July 19; about 1500 feet. Linville, Mitchell Co., July 17, 18; Aug. 30; 3800 feet. Loverings, Burke Co. (between Morganton and Pineola), July 13; 3500 feet. Morganton, Burke Co., July 12, 20; 1200 feet. Murphy, Cherokee Co., July 25, Aug. 22; 1600 to 1800 feet. Pineola, Mitchell Co., July 13, 14, Aug. 28 ; 3800 feet. Raleigh, Wake Co., July 8, 9. Roan Mt., Mitchell Co. (and Carter Co. , Tenn.), July 15, 16, Aug. 31, Sept. 1,2; 4000 to 6300 feet. Roan Valley, Mitchell Co., July 16, Sept. i; 4000 feet. Salisbury, Rowan Co., July u. Saluda, Polk Co., Aug. 17; 2200 to 2400 feet. Selma, Johnston Co , July 7. Tarboro, Edgecombe Co., July 6, 7. Topton, Cherokee Co., Aug. 21; 2700 to 4400 feet. Tunis, Hertford Co., July 5. South Carolina : Columbia, Richland Co., Aug. 16. Denmark, Bamberg Co., Aug. 14,15. Spartanburg, Spartanburg Co., Aug. 16, 17. Georgia ; Blue Ridge, Fannin Co., July 25; 1700 feet. Bolton, Cobb Co., July 29; 800 feet. Jasper, Pickens Co., July 25, 26; 1500 to 2600 feet. Marietta, Cobb Co., July 27; 1000 to 1 200 feet. Sand Mountain, Dade Co. (near Trenton), Aug. 25; 1500 feet. Savannah, Chatham Co., Aug. 13, 14. Stone Mountain, DeKalb Co., July 28; 1000 to 1685 feet. Trenton, Dade Co., Aug. 25; 500 feet. Tybee Island, Chatham Co., Aug. 12, 13. Waycross, Ware Co., Aug. n. West Point, Troup Co. , July 30. Florida ; Carrabelle, Franklin Co., Aug. 9. De Funiak Springs, Walton Co., Aug. 5. Fort Barrancas, Escambia Co., Aug. 3. Live Oak, Suwande Co., Aug. 10. Marianna, Jackson Co. , Aug. 6, 7. Tallahassee, Leon Co., Aug. 8. Warrington, Escambia Co., Aug. 4. Alabama : Flomaton, Escambia Co., Aug. I, 2. Greenville, Butler Co., July 31. Tennessee: Burbank, Carter Co.. July 17, 3000 to 3300 feet. Chattanooga, Hamilton Co., Aug. 24; 700 to 800 feet. Johnson City, Washington Co., Aug. 27; 1700 feet. Lookout Mountain, Hamilton Co., Aug. 23; 2000 feet. Morristown, Hamblen Co., Aug. 27; 1400 feet. Roan Mountain Station, Carter Co., July 17, Sept. 3; 2600 feet. 10 NORTH AMERICAN ACRIDIID^E. While making special efforts to secure Acridiidae, the other species of Orthoptera coming to hand were also taken, broadening somewhat the scope of the inquiry, without interfering with its main object. A folding camera was carried and numerous photo- graphs were secured illustrative of habitats and other features of locust biology, some of which appear with this report. j KENTUCKY / \ ; N E S S I N A O UTH C A .GEORGIA! r/V Aug. 12 FIG. i. Map showing the route traveled. FIELD WORK. Leaving Boston by steamer on June 30, Norfolk, Va., was reached and collecting begun in its vicinity on July 2 ; thence southwest to Tarboro and Selma, N. C., and westward through North Carolina, stopping at frequent intervals, to Morganton, from which point a detour was made by team and rail to Pineola and Roan Mountain, returning via Linville and Blowing Rock ; west- ward again to Asheville, the Balsam Mountains, and Murphy ; southward to Atlanta ; southwest to the Gulf coast at Pensacola ; east through northern Florida to Tallahassee, whence the Gulf NORTH AMERICAN ACRIDIIDjE. 11 coast was reached a second time at Carrabelle ; eastward to Live Oak; northeast via Waycross to Savannah and the Atlantic coast at Tybee Island; north and west through South Carolina to Ashe- ville; a second trip through the valley, including stops at the Balsam and Valley mountains, and on to Chattanooga; northeast to Johnson City, from which point Roan Mountain was visited a second time, and also Linville and Grandfather Mountain; northeast and east through Virginia to Roanoke and Norfolk, returning to Boston by steamer on September 10, after ten weeks in the field. The entire distance traveled was about 4,000 miles, of which 3,000 was through the territory under observation. The map on page 10 will render the route clear at a glance. CLIMATAL CONDITIONS AFFECTING RESULTS. The weather, on which so much depends, was as favorable as could be expected, though in parts of the trip collecting was much interfered with at times by showers, several being not unusual in one day. The heat of the southern summer is also a factor to be reckoned with. The writer was informed at nearly every locality during the first half of the trip, that the season was two to three weeks late. This should be taken into consideration when calculating the time of appearance of species, and attention is called to it in the proper place. LIFE ZONES OF THE REGION VISITED. In the region examined four life zones are represented: The Ivower Austral, including the major part of the Gulf strip of the Austro-riparian division ; the Upper Austral, the Transition, and the Canadian. The Austral zones cover all of the country except the higher mountains; the Transition and Canadian cover the mountains and valleys above an altitude of about 2,500 feet. No attempt was made at an accurate delimitation of these zones, owing to the rapid character of the trip, but many notes on the zonal distribution of the various species will be found in the accompany- ing list (p. 24). Representative species of each zone occur, but some boreal species which it was hoped to find were not observed. The Canadian zone is restricted to limited areas on the higher summits, but austral species, owing to proximity of the Austral zone, are often found within its borders. Its characteristic fauna, while numerically insignificant, is of extreme interest, being represented 12 NORTH AMERICAN ACRIDIID^J. by a wingless species, Podisma gladalis variegata, whose presence is believed to indicate the southward extent of continuous sub- alpine conditions in the Glacial Epoch. NUMBER OF BROODS. Notwithstanding the great length of the season of active life in the I^ower Austral zone, there is, so far as I have been able to learn, but one brood per annum of any species of locust inhabiting the Eastern States ; at least, I have failed to secure definite proof of the occurrence of two broods in any case, though it seems not im- probable that Chortophaga viridifasciata and possibly other species may have two broods in the extreme south. This is a point which might readily be determined by resident observers in Florida or along the Gulf coast. Contrary to the opinion sometimes expressed, the presence of very young or of adult insects in widely separated months of the year does not prove that there are two broods per annum of that species, unless, possibly, in the extreme south, where there is no inactive winter resting season. It needs to be shown that adults appear in numbers twice per annum, alternating with two similar appearances of newly hatched young, in order to establish beyond a doubt the occurrence of two broods annually in the life history of a species. LIFE HISTORIES. There is much variation in individuals of the same species in the time of hatching ; and in different species in their life history with reference to the seasons of the year, some passing the winter as adults, some as nymphs, but by far the most in the egg stage. Schistocercaamericana,Leptysma marginicollis, and several Tettiginse are examples of the first ; in the Northern States Chortophaga viridifasciata, Arphia sulphur ea, and Hippiscus luberculatus are examples of the second. It seems not impossible that some long- lived species may occur as adults in every month of the year in some of the warmer zones. In reference to this point observations are needed during the entire year. ZONAL DISTRIBUTION. Owing to the lack, in many cases, of definite data, the follow- ing statements concerning the zonal origin and distribution of the genera of Acridiidse of eastern North America must be regarded as purely tentative, to be corrected and revised as opportunity permits. NORTH AMERICAN ACRIDHD^E. 13 They are here put forth as a suggestion and stimulus to further research. Genera of boreal origin and distribution. Of the genera of Acridiidse occurring in eastern North America six are boreal: Chlocaltis, Stenobothrus, Mecostethus, Camnula, Circotettix, Podisma. Of these Mecostethus alone is confined to the territory east of the Rocky Mountains, so far as known. Stenobothrus, Mecostethus, and Podisma are Old World genera, and, with Gomphocerus, found among the Rocky Mountains, constitute a reminder of circumpolar land communication. If to these we add Tettix, which is cosmo- politan, we shall have considered all the genera common to Europe and North America, with the exception of Schistoccrca, austral in distribution, one species of which is believed to have crossed from South America to the Old World. Genera of austral origin and distribution. In this group are found six which are peculiar to the Gulf strip of the Lower Austral zone or its immediate vicinity: Paxilla, Rhadinotatum, Macneillia, Gymnoscirtetes, Eotettix, Aptenopedes. Besides these, Dictyophorus, Arnilia, and Leptysma extend further westward in the Lower Austral zone. Others of austral origin and distribution are : Neotettix, Para- tettix, Apotettix, Tryxalis, Mermiria, Syrbula, Eritettix, Ambly- tropidia, Dichro morpha , Clinocephalus t and Paroxya. Of these all but Paratettix, Apotettix, Tryxalis, Mermiria, and Syrbula are confined to the east. Psinidia and Scirtetica, likewise confined to the east but probably of Sonoran derivation, are doubtless of austral origin. Dissosteira, Hippiscus, Spharagemon, Trimerotropis, Arphia, and Hesperotettix, of extended distribution both latitudinally and longitudinally, are probably of austral Sonoran origin, though some species have become adapted to boreal conditions. Chorto- phaga and Encoptolophus , possibly Arphia also, there is reason to think, may have originated in the east. Ageneotcttix, Phcetaliotes, and Mestobregma have apparently but recently entered our territory from the west and as yet occupy but a small portion of it. The place of origin of Pseudopomala is questionable. Schistocerca and Orphulella are undoubtedly of austral origin, though containing species which reach a high altitude and latitude. The same is true of Tettigidea, and possibly of Nomotettix^ both of which seem to be typically eastern in distribution. Melanoplus, the dominant North American genus, covers the continent from Labrador to Mexico, from Alaska to Florida, and contains boreal and austral, eastern and western, and humid-land and arid-land species. 14 NORTH AMERICAN ACRIDIID^E. LOCUST SOCIETIES AND HABITATS. Independently of their zonal distribution, the Acridiidae may be arranged, according to their habits, in local groups which we call, for lack of a better name, societies or associations of species. These groups will vary more or less according to the importance attached to various environmental factors by the person arranging them. We may follow the botanists and recognize hygrophile, xerophile, and other groups, based upon distribution with reference to soil humidity. We may divide them into campestral and sylvan series, frequenting respectively open fields and savannas or forestal environments. Or we may arrange them in still other ways. We find, it is true, species which are characteristic of these various surroundings, and others whose habitats are not so readily classified, or which, widely distributed zonally and geographically, show no especial predilection for any definite kind of environment. A complete and satisfactory classification of locust societies has yet to be made. The hurried character of the trip forbade critical study of this topic, and mention here is made of but a few of the more noticeable features connected with the subject in the territory under consideration. The following table shows the classification in outline of the societies which I have found easily recognizable in the Eastern States, but it needs to be amplified by further study. Fuller details of distribution will be found in the list of species (p. 24). Locust societies of eastern North America. Geophilous division. Phytophilous division. Campestrian group. Campestrian group. Xerophile societies : Xerophile societies : Saxicolous species. Hygrophile " Arenicolous " Sylvan group. Humicolous " Thamnophile societies. Hygrophile societies : Dendrophile " Humicolous species. Paludicolous " Limicolous ' ' Sylvan group. Geophilous locusts are those which freely come into direct contact with the soil and whose local distribution is largely con- trolled by its character. Phytophilous locusts, on the other hand, have much closer relations with the vegetal covering of the soil, be it grass or tree. This primary difference in habits is accompanied by a structural difference in the relative development of the tarsal pulvilli large in the plant-perching species, diminutive, obsoles- NORTH AMERICAN ACRIDIID^E. 15 cent or aborted in the soil-inhabiting species. In eastern North America, with few exceptions, the Tettiginae and Oedipodinae are geophilous; the Tryxalinas and Acridiinae are phytophilous. In the West, owing not only to adaptation in habits to an arid environ- ment, but also in part to the close systematic relationship between the Tryxalinae and Oedipodinse, the distinctness of these two groups is less obvious. GEOPHILOUS DIVISION. The geophilous species of eastern North America are with one partial exception campestral in distribution, as would naturally be expected. The exception, Spharagemon bolli, is an inhabitant of xerophytic forests as well as of open fields, and in the Southern States is found quite as often in the forest as on the open plain. Campestral Geophiles. The campestral geophilous species may be separated into two groups, showing xerophile and hygro- phile tendencies, and represented respectively by the Oedipodinae and the Tettiginae. The former group contains rock-inhabiting, sand-inhabiting, and loam-inhabiting species ; the latter is made up of moisture-loving species, frequenting damp fields, wet meadows, and the shores of streams and ponds. Xerophilous Geophiles. Of saxicolous or rock-frequenting species, Circotettix verruculatus and Spharagemon saxatile of the Northern States are well-known examples. These species are represented in certain parts of the South by Trimerotropis saxatilis, whose habitat is likewise restricted to bare ledges and rocky hills. A colony of this species is located on Stone Mountain, Georgia, * a granite mass which rises 600 feet above the surrounding plain and is almost entirely denuded of soil, whose sun-baked and torrent- washed slopes still provide a scanty existence for a few of these in- sects. (PI. 2, Fig. i.) I found it also on the summit of Sand Mountain plateau, near Trenton, Ga., frequenting the bare rock surfaces of the ' 'glades' ' openings in the forest caused by exposure or proximity to the surface of the underlying rock. (PI. 2, Fig. 2.) Of arenicolous or sand-dwelling species there are several. On the drifting sands of the beach at Cape Henry, Va. , between the shore and the dunes, may be found Trimerotropis maritima, the maritime or sea-side locust, occurring coastwise from southwestern Maine at least as far as North Carolina, and also along the Great Lakes. This species is unknown from inland localities, save as noted. Its congener, T. citrina, however, is found throughout * See Senate Doc. No. 84, 57th Congress, pi. xix. 16 NORTH AMERICAN ACRIDIID^J. the larger part of the austral zones of the Southeastern States where- ever the physical condition of the soil presents a suitable environ- ment, being equally as much at home on dusty roadsides, sun-beaten waste lands, and the sandy river washes of the interior as under the nodding sea oats {Uniola paniculata) of the Virginia cape, the palms of Tybee, or on the snow white strand at Fort Barrancas. Plate 3, Fig. 2 shows one of its inland habitats, a sandy river-wash at Morganton, N. C. (See also PI. 3, Fig. i, and PI. 4, Fig. i). On and near the coast Psinidia fenestralis and Scirtetica picta are frequently associated with it. Hydrophilous Geophiles. This group contains the Tettiginse, whose local distribution is apparently controlled by the presence of conditions favoring the growth of algae and other delicate organisms such as grow upon damp, humus-enriched soil, conditions favored by a considerable degree of moisture and a moderate exposure to light. Even the humus itself is devoured (Hancock). Among them we find species such as Paratettix cucullatus, frequenting the shores of ponds and streams, be they sandy or muddy ; others, like Tettigidea lateralis (and Tettix granulatus of the Northern States) prefer wet meadows and swales; and some, like Nomotettix cristatus and some species of Tcttix, inhabit drier soils of arenaceous loam. PHYTOPHII,OUS DIVISION. i The phytophilous locusts fall very naturally into campestral and sylvan groups. Campestral Phytophiles. The campestral species, like those of the geophilous series, may be arranged in hygrophile and xerophile groups, though perhaps not quite so successfully. Xerophilous Phytophiles. A species belonging to this group and of much interest in the Southern States is Hesperotettix pra- tensis. This locust has a very wide distribution in the western part of the country, from Texas to Washington and from California to Indiana, and was secured last summer in the Southeastern States amid conditions much resembling those of the arid West. A glance at PI. 5, Fig. i will show the desert-like surroundings in which it makes its home along the Gulf shore of Florida, where it is not uncommon among the xerophytic strand vegetation represented by Conradina canescens, Quercus myrtifolia. Ilex vomitoria, Baptisia villosa, and Ceratiola ericoides* the pungent aroma of the latter vividly recalling that of the creosote-bush {Larrea) of the Sonoran deserts. * Plant species kindly identified by Mr. J. M. Greenman, of Harvard University. NORTH AMERICAN ACRIDIID^E. 17 Hygrophilous Phytophiles. Of the hygrophilous group, men- tion should be made of Orphulella olivacea, a species of wide range along the sea-board, recorded from Darien and the Bermudas, and known in the United States from Connecticut, New Jersey, Maryland, Georgia, and western Florida. It is the only halophilous locust of the Eastern States, being restricted to salt-marshes and the vegetation bordering brackish waters. Near Fort Barrancas, Fla., it frequents the " spear-grass " {Juncus) fringing the inlets, and is often accom- panied by Paroxya atlantica and P. ftoridiana. (See PI. 4. Fig. 2.) The two species of Paroxya just mentioned are also typical hygrophiles, the former not uncommon in grassy swamps of the Gulf region, the latter widely distributed in swampy stations in both Upper and Lower Austral zones of the Eastern States, even far in- land, and often accompanied by Tryxalis brevicornis. Leptysma marginicollis, also, is a common and widespread species which fre- quents the erect growth of rushes and sedges which so commonly margin the shores of fresh-water ponds and swampy pools. In meadows and moist fields of the Transition and Canadian zones Stenobothrus curtipennis finds a congenial home, often occurring in countless numbers. In the damper portions of Coastal Plain local- ities, where all topographic features are in low relief, and especially where campestral conditions are mixed with sylvan in the shape of tangled thickets of undershrubs and interlacing herbage (PI. 5, Fig. 2), Gymnostirtetes pusillus, Aptenopedes sphenarioides, Eotettix pusillus, E. palustris, and others are to be found, playing at hide and seek, as it were, among the rushes, white-tufted cotton-grass, yellow- flowered Xyris, pink Sabbatia, and tall Sarracenias. These species may with equal or even greater propriety be placed in the sylvan group, owing to the labyrinthine character of this environment and their adaptation thereto, as we shall see later. Sylvan Phytophiles. Of this group numerous representatives occur in the Southeastern States, as would naturally be expected from the wide extent of the forest and thicket covering character- istic of the humid climate. These will be considered at some length under the next topic and in a general comparison of the campestral and sylvan groups. MOUNTAIN HABITATS. Among the mountains of the southeastern United States cam- pestral environments are largely replaced by sylvan in consequence of the humidity of the climate ; they are, however, represented by 18 NORTH AMERICAN ACRIDIID^E. clearings in the valleys along the streams and by the summit and ridge " balds." The mountain slopes are generally clothed with a forest cover, either of timber or of a dense, shrubby chaparral. The valley clearings lie in the Upper Austral or the Transition zone, according to altitude, and possess a characteristic campestrian locust fauna represented by such phytophilous species as Melan- oplus femur-rubrum and M. atlanis, Stenobothrus curtipennis, Orphulella speciosa, and such geophilous species as Dissosteira Carolina, En- coptolophus sordidus, Chortophaga viridifasciata, Tettix ornatus, T. hancocki, T. arenosus, Paratettix cucullatus, Neotettix bolivari, etc. The "balds" are open spaces of greater or less area situated on the summits or extending along the ridges connecting the summits and are used as pastures for stock. On Roan Mountain and adjacent summits these balds cover hundreds of acres and when not too closely grazed form sedgy lawns of surpassing beauty framed in an unrivaled setting of pink-flowered rhododendrons and dark balsam firs. The locust fauna of these balds is for the most part the same as that of the valley clearings campestral species of wide distribution or of boreal character, but with a distinct austral element (Schistocerca americana, Neotettix bolivari} due to the proximity of that zone. Below, and on the ridges, the balds give place suddenly or gradually to deciduous forests. (PI. i, Figs, i, 2; PI. 8, Fig. i.) The forested areas present two distinct kinds of habitat trees and undergrowth inhabited respectively by dendrophile and by thamnophile (tree-loving and thicket-loving) species. The term thamnophilous I shall apply to those dwelling amid tangled, inter- lacing, vegetal undergrowth, be it of woody plants or herbs, since the biological conditions presented by the two are in certain import- ant respects essentially the same, as will be seen later. No strictly arboreal species (such, for instance, as Melanoplus punctulatus) were observed, though very likely occurring. THAMNOPHILOUS SPECIES. All of the woodland species secured were inhabitants of under- growth, whether found among the more open timber or that forming the sole forest-cover of the mountain-sides. The slopes of Grand- father Mountain (see PI. i, Fig. 2) are largely clothed with a dense, shrubby chaparral or "laurel bald" made up of laurel, rhododen- dron, Menziesia, Leiophyllum, etc., amid which the apterous Podisma glacialis vatiegata finds a home. Here also lives Melanoplus NORTH AMERICAN ACRIDIID^. 19 sylvestris, whose congener celatus of the Virginia mountains bears an even closer resemblance to their northern relative, M. islandicus. On the borders of the clearings and balds and in the more open forests a rank growth of " ol' rich weed " (Eupatorium) is fre- quently found, sometimes covering wide areas. Here Mdanoplus amplectens {=blatchleyi) is a characteristic species, and at the head of Roan Valley has extended its habitat into the higher grassy clearings, thriving in myriads among the timothy and velvet-grass {Phleum and Holcus). (See PI. 7, Figs, i, 2) Many other thicket- dwelling species were secured at various elevations. A COMPARISON OF CAMPESTRAL AND SYLVAN LOCUSTS. Campestrian Species. This group as a whole includes locusts of the open country, be it wet or dry, marsh or mountain, strand or crag. Here belong our commonest and best-known locusts or "grasshoppers," of which typical examples are Melanoplus femur- rubrum and Dissosteira Carolina. Considered carefully, we find that all of the Oedipodinae of the Eastern States fall into this group; of the Tryxalinae the following genera : Mermiria, Tryxalis, Syrbula, Orphulella, Eritettix, Stenobothrus , and Mecostethus ; of the Acridiinae Leptysma, Arnilia, Schistocerca, Paroxya, and about fifteen species of Melanoplus (notably angustipennis ', atlanis, bivittatus, differentiate , extreimis, femoratris, minor, propinquus, symmetricus) ; and a number of the Tettiginae. Sylvan Species. To this group belong primarily those species which inhabit woodlands and thickets or their borders, such as Chloealtis conspersa, Podisma glacialis variegata, Melanoplus amplectens ,. baconi, fasciatus, huroni, islandicus, luridus, morsei, obovatipennis , scudderi, sylvaticus, viridipes, and many of the new species secured during my trip celatus, sylvestris, carnegiei, decoratus, deceptus, devius, divergens, similis, strumosus, and tribulus. To these may properly be added, as already stated, several others whose haunts are amid tangled herbaceous growths wherever found such, for instance, as Dichromorpha viridis, Macneillia obscura, Gymnoscirtetes pusillus, the species of Eotettix and Aptenopedes, and several Tettigince. When we compare the component species of these two groups campestral and sylvan we are at once struck with the fact that a very large majority of the former are long-winged, and of the latter are short-winged or apterous, a fact of much biological interest. 20 NORTH AMERICAN ACRIDIID^E. MACROPTEROUS AND BRACHYPTEROUS SPECIES OF LOCUSTS. The presence of apterous or brachypterous, that is to say, flight- less, species of insects in a relatively large proportion is said to be characteristic of islands and alpine mountain summits; and the current explanation, that it is due to natural selection through the agency of the wind, is widely known. While not questioning the efficiency of this agent in a considerable degree with reference to islands of small area, the evidence from North American locusts requires a different explanation. According to this theory species would be likely to show a tendency toward brachypterism along the shores of continents. It would be interesting to make a quantitative study of variation with reference to this point, but there is no evi- dence at hand to indicate that such is the case. Furthermore, flying species appear to be equally as common on alpine mountain summits as flightless species, other conditions being the same. In North America the predominant group of Acridians is the Melanopli. Of these many are brachypterous or apterous, in short, flightless. The genera Bradynotes, Asemoplus, Podisma, and Para- dichroplus are typically alpine or sub-alpine in distribution and are flightless. The genera Gymnoscirtetes , Eotettix, Aptenopedes, Phoetaliotes, Paratylotropidia, and others are of lowland distribution and are likewise flightless. There are many flightless species of Melanoplus, the dominant genus of the group, which are likewise of lowland distribution, and others which are found at high elevations in mountainous regions, some above and some below timber line. In other groups there are numerous lowland genera containing only brachypterous species (Rhadinotatum, Macneillia, Mesochloa, Dichromorpha, Clinocephalus ', Boopedon, Brachystola, Dictyophorus , Pseudopomala, etc.). Flightless genera and species are distributed impartially as to numbers over lowland and highland areas, in Austral and Boreal life zones. Some factor other than the wind must consequently be involved. The solution of this problem lies in the arrangement of the Acridiidse in the two groups, campestral and sylvan, and the factor involved is, adaptation of structure to habits brought about by a sylvan environment. This has caused a change in structure through disuse of the organs of flight. In short, Brachypterism in locusts is a more complete adaptation to a leaping mode of progression brought about by life in situations where flight is difficult or impracticable, and consequently disadvantageous. That this is the true explanation is indicated by the habits and haunts of the majority of the flightless species (sylvan surroundings or tangled undergrowth wherever NORTH AMERICAN ACRIDIID^E. 21 found); by their distribution locally, horizontally, and vertically; and by the equally characteristic habits, haunts, and distribution of macropterous species as inhabitants of the open field, desert, or savanna. The advantages of progression by flight dispersal widely and easily effected, often aided by the wind, ease of escape from many enemies, etc., and the superiority of this mode in open lands are evident to all. On the other hand, long wings and locomotion by flight are disadvantageous amid dense underbrush, where a leaping mode of progression has decided advantages. Organs unused or disadvantageous tend to dwindle and disappear; hence the loss of wings. If the members of a group of locusts, be it genus or subfamily, differ widely in habits as regards these two kinds of environment campestral and sylvan we find a corresponding difference in wing length, as witness the genus Melanoplus, the group Melanopli, the subfamily Acridiinse. On the other hand, if uniformity of habits characterizes a group, similarity of structure accompanies it, as witness the Oedipodinae, a subfamily characteristic of open, more or less arid surroundings, inhabiting barren fields and washes, the drifting dunes of the seashore, or bare crags of mountain summits, shy and wary of approach, seeking safety in flight, and with few exceptions equipped with large and powerful wings. BRACHYPTERISM IN OTHER ORTHOPTERA. Among other Orthoptera apterous species are numerous both in saltatorial and non-saltatorial families. On examination we find that genera and species inhabiting trees and shrubs (Scudderia, Microcentrum, Cyrtophyllus , Oecanthus, etc.) and open grassy lands {Conocephalus) , are prevailingly long- winged, while those inhab- iting either undergrowth (Xiphidium in part, Odontoxiphidium) , crevices and caves (Ceuthophili, Decticinse, many Blattidse, and Gryllidse), or burrows, either of other species or of their own make (Cryptocercus, Myrmecophila, mole-crickets in part), are very likely to be apterous or brachypterous, an evident adaptation in structure to habits directly parallel with that of brachypterous locusts (cf. also ants, termites). On close analysis it is found that Orthoptera frequenting habi- tats involving passage over open spaces of considerable extent, such as fields, between trees in forests, and bushes or thickets in deserts, are usually long-winged, flying species ; and others dwelling in an environment of more'or less dense, intricate, interlacing vegetal growth, BIO-AGRICULTUR UNIVERSITY OF ( Rl\/FPQinc: rAi u 22 NORTH AMERICAN ACRIDIID^E. be it sub-alpine or sub-tropical, in forest or swamp or in burrows, crevices, etc. in short, in stations where wings are not needed or are at a disadvantage, are very generally apterous or brachypterous. Brachypterism, therefore, appears to be largely not so much a case of natural selection through the agency of the wind as an adaptation in structure to habits. The fact that the heavier-bodied female is more frequently or completely brachypterous than the male and that the tegmina in the latter sex when used as musical instruments are retained in a less degenerate condition (even when entirely useless in flight), confirms this explanation of brachypterism. "BURNING OVER" AS A FACTOR IN DISTRIBUTION. A factor which must seriously affect the distribution of some species is the widespread custom of " burning over" the mountains to improve the range for stock. This practice results necessarily in the destruction of much of the woody debris lying upon the ground which is used as a nidus for the eggs by Chloealtis conspersa, a short- winged boreal species of wide distribution in the Northern States, where it is plentiful in numbers even on the offshore islands. A very few examples of this species were found at high elevations in the mountains of western North Carolina and Virginia, and it would seem not unlikely that the custom referred to may be largely responsible for its scarcity in that region. Since the burning is usually done during the winter, or at least in the inactive season of locust life, it would affect a species having the habit of oviposition mentioned more injuriously than one ovipositing in the earth. HYBRIDS, VARIETIES, AND OTHER TOPICS. What is apparently a case of hybridization resembling that of Basilarchia artkemis and B, astyanax among butterflies occurs in Virginia between two species of Trimerotropis, T. citrina has been already referred to as being widely distributed in the Southeastern States, frequenting sandy areas. At Cape Henry and Virginia Beach it meets T. maritima, a more northern species whose range extends to North Carolina at least. Where the ranges of the two overlap, typical forms of both species are found, and intergrades of all degrees of resemblance to either occur in numbers. Were not the typical forms present it would appear to be simply intergradation, as appears to be the case in the related genus Scirtetica. In the latter genus S. marmorata, found in Massachusetts and Connecticut, differs markedly in degree of pigmentation of the wings and slightly in structure from 5". picta of Florida ; but North Carolina specimens NORTH AMERICAN ACRIDIID^E. 23 are so exactly intermediate that they can be referred to neither species with certainty. If intergradation prove to be the case, the increased amount of pigmentation in the southern form is noteworthy, paralleled as it is in other genera. Additional material in series, and, if practicable, experimentation, is needed to determine the relation of these forms and the possible effect of climatic conditions. Other topics connected with the biological study of this group of insects, such as coloration and stridulation, were given little attention, owing to the limited time for observation in the field. It is sufficient to say that many of the Oedipodinse collected (espe- cially Trimerotropis , Scirtetica, Hippiscus, Psinidia, and Spharagemon) exhibited in a high degree their customary phases of coloration for concealment when at rest (see PI. 8, Fig. 2) and for display (recog- nition or sexual coloration) when on the wing, in the latter case accompanied by a crepitating flight which in itself attracts attention. SPECIES OP ECONOMIC IMPORTANCE. The only species met with in sufficient numbers to be of economic importance were Schistocerca americana, Stenobothrns curti- pennis, Melanoplus differentialis , M. atlanis, M. ampledens, and possibly Chortophaga viridifasciata. But three localities were found where any of these were abundant enough to cause considerable damage. Schistocerca americana occurs throughout the entire region visited, from the seashore to the tops of the highest mountains, but was found in numbers only near Chattanooga, Tenn., where it was accompanied by M. differentialis. M. atlanis also was observed nearly everywhere and is perhaps the most dangerous, potentially, of any species inhabiting the region. It was excessively plentiful at I^.;?. * C \ , - x^-.- J Tf, :v v- ;V*K*vi FIG. 2 Protective coloration of Ttimerotropis citrina. Male and female locusts on sandy soil marked with raindrop impressions. Tarboro, North Carolina. (The rule to show scale.) Library i.xperimert Station varsity of California DATE DUE A 000 643 696 |QL508 |A2M8 Morse, Albert Pitts Kesearches on North American Acridlidae. A2M8 Morse, Albert Pitts Researches on North American Acridiidae. BIO-AGRICULTURAL LIBRARY UNIVERSITY OF CALIFORNIA RIVERSIDE, CALIFORNIA 92502 II