key: cord-0253580-hh5q8xzd
authors: Sturiale, Samantha L.; Bailey, Nathan W.
title: Within-generation and transgenerational social plasticity interact during rapid adaptive evolution
date: 2021-10-22
journal: bioRxiv
DOI: 10.1101/2021.10.21.464843
sha: de7cf7c9432b632ef79ac7adcd46fda896d4c528
doc_id: 253580
cord_uid: hh5q8xzd

The role of within-generation phenotypic plasticity (WGP) versus transgenerational plasticity (TGP) during evolutionary adaptation are not well understood, particularly for socially-cued TGP. We tested how genetics, WGP, and TGP jointly influence expression of fitness traits facilitating adaptive evolution in the field cricket Teleogryllus oceanicus. A male-silencing mutation (“flatwing”) spread to fixation in ca. 50 generations in a Hawaiian cricket population attacked by acoustically-orienting parasitoids. This rapid loss of song caused the social environment to dramatically change. Juveniles carrying the flatwing (fw) genotype exhibited greater locomotive activity than those carrying the normal-wing (nw) allele, consistent with genetic coupling of increased locomotion with fw. Consistent with adaptive WGP, homozygous fw females developing in the absence of song showed reduced body condition and reproductive investment at adulthood. Adult but not juvenile offspring exhibited TGP in response to maternal social environment for structural size, somatic condition, and reproductive investment, whereas adult locomotion and flight was only influenced by WGP. WGP and TGP interacted to shape multiple traits at adulthood, though effect sizes were modest. Interactions between genetic effects and social plasticity within and across generations are likely to have influenced the evolutionary spread of flatwing crickets. However, interactions among these effects can be complex, and it is notable that TGP manifested most strongly later in development. Our findings stress the importance of evaluating trait plasticity at different developmental stages and across generations when studying phenotypic plasticity’s role in evolution.

Dissecting how phenotypic plasticity affects trait expression within and across generations 27 is necessary to fully understand its role in adaptive evolution. Within-generation plasticity 28 (WGP), where an individual's phenotype shifts as a response to its own environmental 29 conditions, has long been argued to influence evolutionary processes (Robinson and Dukas which is predicted to favor adaptive TGP (Leimar and McNamara 2015) . Third, crickets do 92 not possess a fully-developed auditory system until adulthood, though late juveniles may 93 have limited auditory capabilities (Young and Ball 1974; Yack 2004; Staudacher 2009 ). 94

Young juveniles are therefore not likely capable of accurately assessing their own social 95 environment acoustically, which theory predicts will favor the evolution of TGP (Leimar and 96 McNamara 2015) . 97

We performed three experiments to dissect the potential contributions and 98

interactions of genetic evolutionary responses, WGP, and TGP to rapid adaptation observed 99 in Hawaiian T. oceanicus. Supplementary figure S1 provides an experimental overview. 100

Across these experiments, we focused on four traits relevant to mate competition and the loss 101 of acoustic sexual signaling during the evolutionary spread of flatwing crickets: structural 102 size (in juveniles and adults), body condition (in adults), investment in reproductive tissues 103 (in adults), and locomotive activity (in juveniles and adults). Locomotion was a key trait 104 because there is evidence from other insect species that individuals exposed to crowded 105 conditions increase dispersal themselves (WGP) or produce offspring with increased dispersal 106 In Experiment 1, we measured how juvenile locomotive behavior varied across fw 111 and normal-wing (nw) genotypes, before any maternal social manipulation. This allowed us 112 to determine whether the rapid spread of flatwing males might have been associated with 113 genetic changes in juvenile expression of a trait relevant to the changing social environment. 114

We expected that fw-carriers benefit from dispersing less as juveniles and therefore 115 aggregating at higher densities upon reproductive maturity, increasing the chances of mating 116 in a song-less environment. In Experiment 2, we investigated WGP to the acoustic social 117 environment. We focused on fw-carrying individuals because the source population is now the mutation is associated with increased socially-induced WGP (Pascoal et al. 2018 ). We tested whether, consistent with previous studies on females carrying nw genotypes, fw-121 carrying females (the maternal generation) raised in different acoustic environments 122 increase reproductive investment and alter mating behaviors in a way that increases 123 probability of mating with a silent male. Finally, in Experiment 3, we tested 124 transgenerational consequences of the maternal social environment by measuring size and 125 locomotive activity in acoustically-naïve juveniles, and final size, somatic condition, 126

reproductive investment, and locomotive activity in adult offspring. In this last experiment, 127 adults were exposed to either matched or mis-matched acoustic cues compared to their 128 mothers, permitting a direct test of the interaction between WGP and TGP. 129

Experiment 1: Genotypic differences in juvenile locomotion 131

We compared early juvenile behavior across the two cricket morph genotypes using 6 133 laboratory stock lines -3 pure-breeding for fw and 3 pure-breeding for nw. Lines were 134 established in 2016 from a series of controlled crosses of Kauai-derived individuals to ensure 135 homozygosity (Pascoal et al. 2016) . Stock crickets were kept in 16 litre plastic containers with 136 cardboard egg cartons for shelter. Twice weekly, they were provided ad libitum food (Burgess 137 Supa Rabbit Exel Junior pellets; blended for juveniles) and moistened cotton for water and 138 oviposition. Crickets in isolated-rearing conditions were kept in 100 mL plastic deli pots with 139 shelter, food, and water as above. All subjects were kept in the same growth chamber at 25°C 140 on a photo-reversed 12:12 hour light:dark cycle unless otherwise indicated. To obtain 141 juveniles for this experiment, we collected eggs from each line twice weekly for 4 weeks. After 142 approximately two weeks we monitored egg pads daily (16:00 -18:00) and isolated new 143

An open field test (OFT) was used to track individual crickets' movements in an unobstructed 146 arena and measure their total distance travelled, a useful proxy for measuring behaviors 147 related to dispersal, mate location and foraging (Fraser et al. 2001; Dingemanse et al. 2003 ; 148 tested in an OFT at 15-days and 45 days post-hatching. Juveniles of these ages do not have 150 mature hearing structures (Young and Ball 1974) . All OFTs were performed under red light 151 during the dark portion of the crickets' 12:12 light:dark cycle, between 23-25°C. Subjects were 152 placed in small glass vials within their deli pot to reduce handling disturbance before testing. 153

The vial was gently turned over onto the center of an 11x17 cm clear plastic arena atop white 154 poster paper and the cricket was allowed to acclimatize for two minutes. Upon lifting the vial, 155 we began recording for 5 minutes at 30 frames/second using a camera (Nikon D3300) 156 mounted ca. 40 cm above the arena. The arena was wiped down with 70% ethanol before each 157 trial to minimize residual chemical cues. Two crickets were assayed at once in side-by-side 158 arenas. It is unlikely that they were aware of one another due to their inability to see in red 159 wavelengths of light. After the OFT, each cricket was photographed overtop a micrometer 160 using a Leica DFC295 digital camera affixed to a Leica M60 dissecting microscope. ImageJ 161

(v.1.8.0_112) was used to record pronotum length (a proxy for structural size) from the 162 images. 163

We used DORIS v.0.0.17 (Friard 2019) to extract coordinates of the test subject within each 165 video frame, followed by coordinate path smoothing implemented in R (R Core Team 2020) 166 to increase measurement precision (see Supplemental Methods and Figure S2 ). Using these 167 coordinates, we measured total distance traveled ("distance") during trials. In this and later 168 experiments involving open field tests, we also explored other movement parameters 169

("proportion explored" as a measure of exploratory activity; and "origin time", "middle time", 170 and "edge time" as measures of space usage and thigmotaxis). However, variation in these 171 parameters was largely accounted for by overall differences in distance moved, confirming 172 that distance was the most salient locomotion trait in the experiment. For completeness, we 173 discuss the measurement and analysis of all other movement traits in the Online 174

Supplementary Information. 175

All statistical tests were carried out using R version 4.0.2 (R Core Team 2020). We compared 177 distance between wing morph genotypes in 15-day old and 45-day old offspring using a linear 178 model. Individuals who jumped during their assay (n = 2 in 45-day assay) or whose video was inadvertently deleted before analysis (n = 2 in 45-day assay) were excluded. All data 180 transformations are shown in Table S1 . 181

Morph and sex were modelled as categorical variables, with line nested within morph 182 to account for inter-line variation. Pronotum length, temperature, and time of day were 183 included as covariates. Thirty-nine individuals died before their sex could be identified, so to 184 verify that sex did not qualitatively affect the findings, models including sex as a fixed effect 185 were run on the subset of individuals for which sex could be identified. Sex did not approach 186 significance in this model (all p > 0.2) and the qualitative outcome did not differ. Thus, the 187 model retaining all individuals, and excluding sex as a fixed effect, was retained (Equation 1 188 of Supplementary Table S2 ). Finally, the model was run first with all individuals, then with 189 only those who moved during the assay to confirm that genotypic variation in distance was 190 not due to differences in the likelihood of initiating movement. Excluding crickets that failed 191 to initiate movement did not affect interpretations of genotype differences, so final models 192 included stationary crickets (Supplementary Table S3) . 193

The three pure-breeding fw lines used in Experiment 1 were reciprocally interbred to create 196 an admixed pure-breeding fw stock population. Following previous work, we isolated juvenile 197 females from this stock when sex became apparent to ensure virginity and more easily 198 manipulate their acoustic environment (Bailey and Zuk 2008; Pascoal et al. 2018 ). We also 199 segregated a group of juvenile males into single-sex 16-L box to maintain their virginity. All 200 group rearing conditions were identical to Experiment 1. Isolated females were placed in a 201 separate, temperature-controlled 25˚C incubator on a 12h:12h photo-reversed light cycle, 202 with no male calling. Females were checked daily for adult eclosion, whereupon they were 203 haphazardly assigned one of two acoustic social treatments: Song or No Song. Females do not 204 achieve reproductive maturity until several days after adult eclosion, so our acoustic 205 treatment targets the developmental period when mate assessment is possible but mating is 206 not (Swanger and Zuk 2015) . We also recorded the number of days spent isolated prior to 207 eclosion to account for any differences in growth rate that might be associated with time 208 spent without song prior to adult acoustic treatment. We kept each female in their acoustic 209 treatment for 15 days post-eclosion. 210

In the Song treatment, Kauai male calls reflecting population averages for key song 212 parameters were played at 80-85 dB (measured at the lid of the deli cup which has an acoustic 213 impedance of ca. 10 dB) during the night portion of the crickets' light:dark cycle to best match 214 calling dynamics in the wild (Zuk et al. 1993) . Playbacks used in the Song treatment have incubator. Calls were broadcast from computer speakers (Logitech Z120 2.0) and the calling 219 schedule programmed using the Task Scheduler application on a desktop computer. Twice a 220

week, we switched which incubator housed each acoustic treatment to prevent any incubator-221 related experimental confounds. 222

At 15 days post-eclosion, isolated adult females were weighed and their pronotum width was 224 measured using digital calipers. Each female was placed in a 16 x 18 cm plastic container 225 with cardboard, rabbit chow, and moistened cotton. We haphazardly selected an adult virgin 226 male from the flatwing stock population, weighed it, measured its pronotum width, and 227 placed it in the container with the female. Trials were performed between 20-23°C under red 228 light between 16:00h and 18:00h. They lasted for 20 minutes, and we noted whether the 229 female mounted the male and whether the male transferred a spermatophore. Afterwards, 230 pairs were placed in a separate incubator without male song at the same temperature and 231 light:dark cycle as in Experiment 1. After 24 hours, the male was removed to reduce potential 232 paternal influences on offspring phenotype. After another 24-48 hours, the female was 233 removed, and the egg pad was collected for use in Experiment 3. 234

To compare female body condition, we used pronotum width and total body weight to 236 calculate the scaled mass index (SMI) of each individual (Peig and Green 2009). A subset of 237 females drawn haphazardly from each treatment (total n = 23) were dissected at 15 days 238 post-eclosion rather than mated. We recorded their pronotum width using digital calipers, weighed them, and then determined wet mass of their dissected ovaries. Somatic mass was 240 calculated by subtracting ovary mass from total weight. 241

First, we compared SMI across acoustic treatments using a linear model (Equation 2 of Table  243 S2) with acoustic treatment as a categorical factor, days isolated before treatment as a 244 covariate, and experimental replicate (block one or block two of the experiment). Replicate 245 only had two factor levels so we included it as a fixed effect. Second, we compared mating 246 behavior across acoustic treatments. We first ran a generalized linear model (GLM) with 247 binomial error to examine presence vs. absence of female mounting during trials, including 248 acoustic treatment, female SMI, and male SMI as predictors. Next, we ran a binomial GLM 249 examining presence vs. absence of spermatophore transfer. For this, we only included the 49 250 mating trials (out of 65 total) where mounting had occurred, because spermatophore transfer 251 cannot occur without mounting. Acoustic treatment was included as a categorical factor and 252 female SMI and male SMI were included as covariates. Equation 3 in Table S2 gives the 253 general form of these models. Table S2 ). we noticed that some adults attempted to fly out of the arena during the assay. When that 293 happened, we stopped the recording and placed the subject into an incubator without song 294 for 10 minutes. After re-acclimation, we started the trial again. The number of flight attempts 295 was recorded for each individual. We collected movement coordinates and calculated distance 296 using DORIS (v.0.0.17) as in Experiment 1. 297

Following the OFT, we photographed each juvenile overtop a micrometer using the same 299 camera and dissecting scope as in Experiment 1 and measured pronotum length using 300 ImageJ (v.1.8.0_112). We euthanized adults after OFTs at 8 days post-eclosion, then weighed 301 them and measured pronotum length to the nearest 0.01 mm using digital calipers. We then 302 dissected, blotted excess fluid, and weighed their gonads (male testes and accessory glands, 303 female ovaries). Here we used pronotum length and soma weight to calculate SMI, by 304 subtracting gonad weight from total weight. In this experiment, SMI was thus a measure of 305 somatic body condition, which allowed us to investigate whether differences in maternal or 306 offspring acoustic environments influenced relative investment in somatic tissues while 307 scaling to structural size. SMI was calculated separately for each sex. Gonad weight was later 308 compared directly. 309

First, we tested whether juvenile offspring size differed between maternal acoustic 311 treatments by running a linear mixed model (LMM) using pronotum length as the response. To investigate TGP and WGP and their interaction, we tested the effect of maternal 327 and offspring acoustic treatments on adult pronotum length and somatic condition (SMI). 328

Because there are large sex differences in physiology and the possibility of sex-specific maternal effects, we ran separate models for each sex. Each LMM included maternal and 330 offspring treatments as factors plus their interaction. Non-significant (p > 0.2) interactions 331 were removed. We analyzed the effect of acoustic treatments on adult offspring reproductive 332 investment using sex-specific models with gonad weight as the response. First we compared 333 unscaled reproductive investment, then we added pronotum length as a covariate to examine 334 whether variation in reproductive investment could be explained by structural size. Finally, 335

we added log-transformed somatic mass to examine whether variation in reproductive 336 investment might be explained by somatic weight. Replicate was included as a random effect. 337

These models took the general form shown in Equation 7 of Supplementary Table S2 . 338

We then tested the impact of TGP and WGP on adult distance using separate LMMs 339 for each behavior and sex. All models included maternal treatment, offspring treatment, 340 temperature, time, and somatic SMI, plus replicate as a random effect (Equation 8 in Table  341 S2). Non-significant (p > 0.2) maternal*offspring treatment interactions were removed. 

Fw-carrying nymphs moved further than nw nymphs, both at 15 days and 45 days posthatching (Table 1, Figure 1 ). The effect size of distance differences was considerable. 15-dayold fw nymphs, which had a mean length of 3.81 mm, moved an average of ca. 300 mm further than nw nymphs. This difference in distance moved is ca. 79x their body length in a relatively short period of 5 minutes. For 45 day-old nymphs, the average movement differential was 132.73 mm (ca. 12.9x the mean body length of a 45 day-old nymph). 

The acoustic social environment affected physiology (Table 2, Figure 2 ), but not mating 352 behavior (Supplementary Table S5 ; Supplementary Figure S4 ), in homozygous fw females. 353

Those raised in Song attained higher condition (SMI) (Table 2, Figure 2A ) and had heavier 354 ovaries relative to somatic mass (Table 2, Figure 2B ). Female mounting was not influenced 355 by prior acoustic experience, though females were more likely to mount higher condition 356 males (Supplementary Table S5 and Supplementary Figure S4A) . Similarly, in trials where 357 females did mount males (n = 49), there was no evidence that acoustic treatment affected 358 spermatophore transfer, though males were more likely to transfer a spermatophore to 359 higher condition females (Supplementary Table S5 and Supplementary Figure S4B) . 

Unexpectedly, TGP affected adult, but not juvenile, traits (Figure 3 ). For example, the 368 acoustic treatment of mothers was not associated with locomotion and morphology of their 369 15-day-old and 45-day old juvenile offspring (Table 4, Supplementary Tables S6 and S7) . 370

However, adult offspring that experienced song themselves during rearing moved further 371 (WGP) (Table 5; Figure 3A) , and in the case of males, this WGP was considerably exaggerated 372 if their mothers had been raised without song (TGP) (Figure 3A, right) . It must be noted, this 373 WGP*TGP interaction was only marginally significant (p = 0.055; Table 5 ), though the effect 374 size appears non-trivial (mean movement differential of ca. 400 mm for adult song-reared 375 males; Figure 3A , right). 376 By contrast, acoustic effects on adult offspring morphology provided strong and 377 consistent evidence for WGP, and TGP also affected several aspects of adult offspring 378 morphology (Tables 6 and 7; Figure 3B ,C,D). TGP often did not affect traits in the same 379 direction as WGP; interactions between TGP and WGP combined to shape adult traits in a 380 way that suggests TGP activates WGP, or put another way, that the manifestation of TGP is 381 contingent on current environmental conditions. For example, female offspring reared 382 without song had similar pronotum lengths, but when they were reared with song, those 383 whose mothers experienced No Song grew to be larger than those whose mothers experienced 384 Song (Table 6 ; Figure 3B , left). Female somatic condition showed a similar crossing-over 385 effect (Table 6 ; Figure 3C , left). Also, TGP and WGP affected female investment in ovaries, 386 but in conflicting directions. Those raised in song developed heavier ovaries than those raised 387 without song, and offspring from the No Song maternal treatment developed heavier ovaries 388 than offspring from mothers who experienced Song (Table 7 ; Figure 3D , left). Adult males 389 raised in the presence of song developed higher somatic condition than those raised without 390 song, regardless of maternal treatment (Table 6 ; Figure 3B , right). We also found that 391 crickets raised without song attempted flight more than those raised with song, particularly 392 for females and lower condition individuals (Table 8; Figure 6 ; Supplementary Figure S5 ). 

Phenotypic plasticity's role in evolution stimulates vigorous debate, but one barrier to a 396 general resolution may be that plasticity is not a monolithic phenomenon. Influential verbal 397 models have suggested that "buffering" effects of plasticity can permit novel adaptations to 398 escape loss at low frequencies and subsequently spread under selection, but few empirical 399 studies have been able to assess the contributions and interactions of genetics and different 400 forms of phenotypic plasticity such as within-generation and transgenerational plasticity. 401

Here we demonstrate how all three inputs -genetics, WGP and TGP -interact to affect traits 402 that potentially facilitate the rapid evolution of a parasitoid-avoidance adaptation in 403

Hawaiian field crickets, male silence. 404

Genotype had a surprisingly large effect on juvenile behavior, but in the opposite 405 direction predicted. At very early juvenile stages, flatwing carriers moved nearly 80 body 406 lengths further than normal-wing carriers in a span of only 5 minutes. This genotypic 407 difference could result from pleiotropic effects of the fw mutation or genomic hitchhiking. 408

Genotypic differences in juvenile locomotion are consistent with phenotypic differences that 409 have been detected between fs and nw carriers in other sexually dimorphic adult traits 410 that the fw genotype may be exposed to selection at an earlier stage than previously 412 considered, for example through associated effects on foraging efficiency or predation risk. 413

Further, it is possible that increased locomotion of fw juveniles might have accelerated the 414 speed at which the mutation initially spread in the wild. Rather than facilitating local mating 415 aggregations as we initially hypothesized, greater movement activity may instead permit 416 silent crickets and females carrying male-silencing variants to encounter one another. The 417 ultimate fitness consequences of these differences remain to be tested, but more broadly, this to find a mate. Consistent effects of TGP and WGP on offspring would facilitate this, but instead we found that TGP and offspring WGP acted in opposing directions ( Figure 3D ). It is 443 therefore unlikely that this TGP in reproductive investment is an adaptive, anticipatory 444 effect to increase offspring fitness, but instead may be an incidentally-transmitted 445 physiological consequence of mothers responding to their social environment ("selfish TGP" 446 cf. Marshall and Uller 2007) or cross-generation spillover of parental condition ("condition-447 transfer effects" cf. Bonduriansky and Crean 2018). Lack of support for adaptive TGP is 448 consistent with a recent meta-analysis which recovered weak evidence for it across taxa 449 (Uller et al. 2013 whereas the social environment comprised of adult social cues is likely to be of greater 466 relevance to offspring when they are adults themselves. 467

In contrast to morphological traits, we found that behavioral traits related to 468 movement (locomotive activity and likelihood of flight) were more influenced by WGP than 469 TGP. This supports the prediction that traits whose expression remains flexible after 470 development are more strongly affected by WGP (Beaty et al. 2016) . Specifically, offspring of 471 both sexes were more active when reared in song. They may increase walking activity to 472 locate conspecifics they perceive to be abundant nearby, even in the absence of an immediate 473 acoustic cue. In contrast, crickets raised without song have no indication of nearby conspecifics and may decrease short-range mate-searching via walking to instead wait for an 475 acoustic cue. This trade-off is likely motivated by a high metabolic cost of mate-searching 476 (Hack 1998 ) and the resulting increase in predation risk (Bell 1990 ). Our results contrast 477

with previous studies in this species which found that adult males raised in song are less 478 active than those raised in silence (Balenger and Zuk 2015) and that females exhibit limited 479 flexibility in locomotive behavior in response to acoustic environment (Heinen-Kay et al. Our results support the idea that the effects of TGP can be contingent upon offspring 504 environment. Put another way, sometimes TGP potentiates WGP, and sometimes it 505 suppresses WGP. It is therefore necessary to consider the potentially conflicting effects of WGP and TGP when predicting how phenotypic plasticity influences adaptive evolution. 507

Theory predicts that, following rapid environmental change, populations may exhibit a 508 

Offspring size plasticity in response to intraspecific 526 competition: An adaptive maternal effect across life-history stages

Adjusting phenotypes via within-and across-529 generational plasticity

Mate choice plasticity in the field cricket Teleogryllus oceanicus: Effects of social 531 experience in multiple modalities

A neglected 534 conceptual problem regarding phenotypic plasticity's role in adaptive evolution: The importance of 535 genetic covariance and social drive

Density-Related Dispersal in Planthoppers : Effects of Interspecific 569

Natal dispersal and 571 personalities in great tits (Parus major)

The role of adaptive 574 trans-generational plasticity in biological invasions of plants

Achieving high sexual size dimorphism in insects: 577 Females add instars

The fitness costs of adaptation via phenotypic plasticity and 579 maternal effects

Explaining leptokurtic movement 581 distributions: Intrapopulation variation in boldness and exploration

Adaptive versus non-adaptive phenotypic 585 plasticity and the potential for contemporary adaptation in new environments

Development of intraspecific size variation in black coucals, white-browed coucals and ruffs from 589 hatching to fledging

The energetics of male mating strategies in field crickets

The cost of reproduction: the devil in the details

Mate choice by female crickets is influenced by predation risk

Direct and indirect effects of sexual signal loss model approach

Genetic integration of local dispersal 605 and exploratory behaviour in a wild bird

What Is an Adaptive Environmentally Induced Parental Effect?

Maternal effects in Daphnia: What mothers are telling their 609 offspring and do they listen?

Adaptation to an extraordinary environment by evolution of phenotypic plasticity 611 and genetic assimilation

The evolution of transgenerational integration of information in 613 heterogeneous environments

Quantity and quality of available mates alters female 615 responsiveness but not investment in the Pacific field cricket, Teleogryllus oceanicus

When is a maternal effect adaptive? Oikos

Maternal investment mediates offspring life history 620 variation with context-dependent fitness consequences

socially mediated plasticity in gene expression accompanies rapid adaptive evolution

Rapid evolution and gene expression: a rapidly evolving Mendelian trait that silences field crickets 631 has widespread effects on mRNA and protein expression

Field cricket genome reveals the footprint of recent

R: A Language and Environment for Statistical Computing

A silent orchestra: convergent song loss 639 in Hawaiian crickets is repeated, morphologically varied, and widespread

Release from intralocus sexual conflict? Evolved loss of a 642 male sexual trait demasculinizes female gene expression

Can behaviour impede evolution? Persistence of singing 645 effort after morphological song loss in crickets: Singing effort in mute crickets

The Influence of Phenotypic Modifications on Evolution : The Baldwin 648 Effect and Modern Perspectives

The ghosts of predators past: Population cycles and the role 650 of maternal programming under fluctuating predation risk

The auditory system of last instars in Gryllus bimaculatus DeGeer

Sex Differences in

Phenotypic Plasticity Affect Variation in Sexual Size Dimorphism in Insects: From Physiology to 656

Cricket Responses to Sexual Signals are Influenced More by Adult than Flexibility in male phonotaxis behavior and the loss of singing ability in the field cricket Teleogryllus 663 oceanicus

Measuring relative investment: A case study of testes investment 665 in species with alternative male reproductive tactics

Weak evidence for anticipatory parental effects in plants and 668 animals

Lévy 670 flight search patterns of wandering albatrosses

Developmental plasticity and evolution

The structure and function of auditory chordotonal organs in insects

Structure and development of the auditory system in the prothoracic leg of 676 the cricket Teleogryllus commodus (walker) -I. Adult structure. Zeitschrift für Zellforsch und 677 mikroskopische Anat

Effects of larval density on dispersal and fecundity of 679 western corn rootworm

Silent night: Adaptive disappearance of a sexual signal 682 in a parasitized population of field crickets

Calling Characteristics of Parasitized and Unparasitized

We are grateful for support from the Natural Environment Research Council to NWB