Present address: Department of Animal Ecology, Evolutionary Biology Centre (EBC), Uppsala University, Norbyvägen 18D, SE-752 36 Uppsala, Sweden.
Sexual conflict in wing size and shape in Drosophila melanogaster
Article first published online: 2 AUG 2010
© 2010 The Authors. Journal Compilation © 2010 European Society For Evolutionary Biology
Journal of Evolutionary Biology
Volume 23, Issue 9, pages 1989–1997, September 2010
How to Cite
ABBOTT, J. K., BEDHOMME, S. and CHIPPINDALE, A. K. (2010), Sexual conflict in wing size and shape in Drosophila melanogaster. Journal of Evolutionary Biology, 23: 1989–1997. doi: 10.1111/j.1420-9101.2010.02064.x
Dryad Digital Repository doi:10.5061/dryad.1754
- Issue published online: 20 AUG 2010
- Article first published online: 2 AUG 2010
- Received 27 April 2010; revised 30 June 2010; accepted 1 July 2010
- Drosophila melanogaster;
- experimental evolution;
- geometric morphometrics;
- intralocus sexual conflict;
- ontogenetic sexual conflict;
- sexual size dimorphism
- Top of page
- Supporting Information
Intralocus sexual conflict occurs when opposing selection pressures operate on loci expressed in both sexes, constraining the evolution of sexual dimorphism and displacing one or both sexes from their optimum. We eliminated intralocus conflict in Drosophila melanogaster by limiting transmission of all major chromosomes to males, thereby allowing them to win the intersexual tug-of-war. Here, we show that this male-limited (ML) evolution treatment led to the evolution (in both sexes) of masculinized wing morphology, body size, growth rate, wing loading, and allometry. In addition to more male-like size and shape, ML evolution resulted in an increase in developmental stability for males. However, females expressing ML chromosomes were less developmentally stable, suggesting that being ontogenetically more male-like was disruptive to development. We suggest that sexual selection over size and shape of the imago may therefore explain the persistence of substantial genetic variation in these characters and the ontogenetic processes underlying them.
- Top of page
- Supporting Information
The existence of sexual dimorphism is, in and of itself, evidence that the two sexes have had a history of disruptive selection. Recently, it has been suggested that constraints on the evolution of sexual dimorphism as a result of genetic correlations between the sexes may impose a substantial load on the fitness of one or both sexes (Rice, 1984; Prasad et al., 2007). This ‘gender load’ may sometimes be detectable as a negative intersexual genetic correlation for fitness, and evidence for such a pattern of covariation across the sexes has accumulated in the last decade in a variety of sexual organisms in both the laboratory and the field (reviewed in Bonduriansky & Chenoweth, 2009; and Cox & Calsbeek, 2009). Nonetheless, intralocus sexual conflict is, and will probably always be, difficult to measure because of (i) the composite nature of fitness and the virtual certainty of an admixture of trait-specific intersexual genetic correlations affecting it; (ii) the fact that maintenance of sexually antagonistic genetic variation requires specific, locus-dependent (i.e. autosomal or sex-linked) relationships between the selection coefficients on males and females; and (iii) a variety of environmental and genetic factors that will tend to make intersexual correlations positive (Bonduriansky & Chenoweth, 2009; Cox & Calsbeek, 2009).
One way to observe intralocus sexual conflict as an evolutionary force is to manipulate the relative intensity of selection on the two sexes. We followed the approach of Rice (1996) to eliminate female gene expression in Drosophila melanogaster by limiting virtually the entire genome (all but the dot chromosome IV; < 1% of the genome) to males. Under this male-limited (ML) experimental evolution scheme, the X-chromosome and both the major autosomes behave like a single large Y-chromosome in that they are transferred from father to son and are never expressed in females. This lets us harness the genomewide power of many loci to augment the benefits of sex limitation and allows loci polymorphic for male-benefit/female-detriment alleles to be positively selected. After a number of generations of ML evolution, the ML-selected chromosomes can then be expressed in both males and females to test their effects in a standardized genetic background. ML evolution should generate populations approaching the best masculine phenotypes available from that fraction of the standing variation in the ancestral populations. In accordance with the predictions from intralocus sexual conflict, it has previously been found that release from selection upon female function led to a burst of male-specific adaptation: the fitness of males increased and the fitness of females inheriting ML genotypes decreased (Prasad et al., 2007). These evolved fitness differences were accompanied by phenotypic shifts towards the male optimum (inferred from the direction of extant sexual dimorphism) in developmental time and body size (Prasad et al., 2007). Gains in male fitness were mediated by increased attractiveness and mating success (Bedhomme et al., 2008) and not by post-copulatory sexual selection (S. Bedhomme, unpublished data), therefore directing our attention to aspects of behaviour and the physical phenotype related to courtship and mating.
Because ML evolution resulted in a shift towards the male optimum for previously studied traits, this method should be useful for studying other traits exhibiting substantial sexual dimorphism in Drosophila, such as body size. Unlike vertebrates, sexual size dimorphism (SSD) in which females are larger than males is the rule rather than the exception in the Arthropoda and is proximately explained by differences in growth rate rather than development time (Blanckenhorn et al., 2007). The main hypotheses offered to explain this pattern are fecundity selection in females, female anautogeny (where females must feed before oviposition, Blanckenhorn et al., 2007), selection for protandry (Maklakov et al., 2004), and a higher cost of production of male gonadal tissue (Miller & Pitnick, 2003). A fifth hypothesis has occasionally been advanced, connecting small male size to direct benefits accruing from sexual selection, such as mate finding (Moya-Loraño et al., 2002). Drosophila melanogaster displays the typical arthropod pattern for SSD, but more strikingly, males are not only smaller than females but also take longer to mature, making them substantially slower growing (Blanckenhorn et al., 2007). There is evidence that fitness is positively associated with locomotor activity in males, and that this is a sexually antagonistic trait, with more active females experiencing reduced fitness (Long & Rice, 2007). One potential explanation for this result is that smaller males excel in chasing, harassment, or courtship displays involving speed or agility, but their daughters inherit only the negative effects of small size on fertility. A second related hypothesis is that whereas females benefit from rapid growth in terms of fertility selection, males benefit from slower growth because it promotes higher ontogenetic fidelity and resulting morphological quality. This latter ‘selection for perfection’ model (Chippindale et al., 2003) suggests that the risks of rapid growth are not just those associated with increased feeding rate and exposure to predators, but also risks associated with developmental accidents. In this model, the risks associated with rapid growth are outweighed by the benefits for females, but not for males, because male fitness may be substantially negatively impacted by developmental accidents that render them further from the optimal size or shape, and/or more asymmetrical.
Developmental stability is the ability of an organism to buffer its phenotype against genetic or environmental disturbances encountered during development and is usually measured as the inverse of the mean fluctuating asymmetry (FA, Clarke, 1998). The selection for perfection model predicts that this sort of developmental buffering should be more important for males than for females. More specifically, in the context of the ML evolution experiment, we expect that ML males will (i) be more symmetrical than Control males and (ii) evolve to be closer to the male phenotypic optimum inferred from extant sexual dimorphism in size and shape (i.e. have smaller wings that are more masculine in shape). To investigate these hypotheses, we carried out a geometric morphometric analysis of wing morphology. Wing morphology was chosen as an appropriate trait to measure when looking for evidence of intralocus sexual conflict as it is known to be subject to sexual selection in males (Taylor & Kekic, 1988) and lends itself well to landmark-based methods (Klingenberg & McIntyre, 1998) and FA analysis (Palmer, 1994; Palmer & Strobeck, 2002; Breuker et al., 2006).
- Top of page
- Supporting Information
We expressed ML and Control (C) haploid genomes (‘hemiclones’ consisting of the major autosomes and the X-chromosome) from four replicate lines in both sexes after 82 generations of experimental ML evolution (Prasad et al., 2007). We assayed fitness and investigated intralocus sexual conflict and developmental stability in wing morphology. For more details about ML evolution and the production of flies for fitness and morphological measurements, please see Supporting Information.
Female fitness was measured as follows: females were isolated as virgins and housed in groups of 10 along with five competitor females from a replica of the base stock (LHM) homozygous for the relatively benign recessive scarlet eye marker (called LHst) and were provided with 10 mg of yeast/vial. On day 12 post egg lay, females were combined with 20 males from LHst for 18 h, after which they were separated from the males and the ML females were allowed to oviposit for 20 h (LHst females were discarded). The progeny enclosing from these vials were counted 12 days later. Female fitness was therefore measured as total number of adult offspring produced after competition for a limited resource (yeast). Fifteen such vials were set up per population, and final sample size was 119 vials.
To measure male fitness, males were harvested 11 days post-oviposition. Ten males from ML (or C) populations were combined with 10 males from LHst population. Fifteen such vials were set up per population. On day 12 post egg lay, males were combined with 15 virgin clone-generator females and allowed to interact for 18 h after which the females were separated from the males and allowed to oviposit for 18 h. The progeny from the two types of males can be distinguished because of their eye colour. Twelve days later, the fraction of progeny sired by the focal males (ML or C) within each vial was scored, and this proportion was used as a fitness measure. Fifteen such vials were set up per population, and final sample size was 115 vials.
Male and female fitness were measured in different currency. To be able to include the two fitness measures in a same analysis, we calculated mean values for each sex within each replicate population (ML and C values pooled) and then divided the values for each sample by the appropriate mean to obtain sex-specific relative fitness values. Mean relative fitness values for each combination of sex, replicate population, and selection regime were calculated (N = 16) and then were analysed using a factorial anova in JMP, with sex (M or F), selection regime (C or ML), and their interaction (sex*sel) as fixed factors.
Individuals slated for morphological analysis were frozen and stored individually in Eppendorf tubes at −20 °C until they could be processed. Wings were mounted by hand on glass microscope slides using double-sided tape. Sample size was 965 individual flies (between 48 and 73 per population/sex/selection regime). After wing removal, flies were dried for at least 24 h in a 65 °C drying oven before being individually weighed to the nearest 0.0001 g on a Cahn C-31 microbalance. Eleven landmarks were selected for geometric morphometric analysis (Fig. 1a). These landmarks are similar to those used in other studies of wing morphology (Breuker et al., 2006; Gidaszewski et al., 2009). However, some landmarks on the proximal part of the wing that have been used in previous studies were not included here as it was sometimes difficult to remove the wing without damaging this area. Wings were photographed and digitized twice (nonsuccessively) to account for error because of distortion by camera/microscope lenses and variation in the placement of landmarks (Klingenberg & McIntyre, 1998). Unfortunately, it was not possible to entirely control for error caused by the mounting process, but individuals with wings that were damaged or creased in any way were excluded from the analysis. Also, because wings were mounted and digitized in a random order, improvements in mounting/digitizing technique over time cannot be the cause of any systematic differences between groups. Geometric morphometric analysis (digitization of landmarks, procrustes superimposition, relative warp analysis, and visualization of shape differences) was carried out in the tps suite of programs by F. James Rohlf (tpsUtil, tpsDig, tpsRelw, tpsRegr, and tpsSplin), which are freely available at http://life.bio.sunysb.edu/morph/.
Centroid size was used as a measure of wing size, and wing shape was analysed using relative warp scores. Note that centroid size, despite being a linear measure, is very highly correlated with wing area (r = 0.99, P < 0.0001) for this dataset. Wing loading was calculated as dry mass/wing centroid size, and allometric slopes were obtained by regressing wing size on body mass for each combination of sex, replicate population, and selection regime. Because previous results found differences in body mass between ML and Control flies (Prasad et al., 2007), we were interested in investigating allometric slopes to see whether differences in wing size could simply be attributed to the evolution of differences in body size.
Developmental stability in wing size was examined using FA analysis (Palmer, 1994; Palmer & Strobeck, 2002). Because male and female D. melanogaster differ substantially in size, size-standardized wing size asymmetry values were calculated via ln(R)-ln(L) (Palmer & Strobeck, 2002). We carried out analysis on both standardized data [i.e. using ln(R)-ln(L) values] and raw data (i.e. using raw size and shape values), but as results were qualitatively similar for both datasets, only the standardized analysis is presented in detail here. Before any tests of wing size FA were performed, an anova was carried out to quantify and test the different components of asymmetry: error, FA, and directional asymmetry (DA; see Palmer & Strobeck, 2002 for details). FA was large relative to error variance and therefore significant (F964, 1394 = 8034, P < 0.0001), and although there was significant DA (F1, 1394 = 63.77, P < 0.0001), this was probably mostly due to the large size of the dataset (Palmer & Strobeck, 2002). The side*wing size effect was very small (Cohen’s d = 0.0194), indicating that DA was much smaller than the average deviation around the mean. It was therefore not deemed necessary to correct for DA (Palmer & Strobeck, 2002). Signed asymmetry values were normally distributed. Mean absolute asymmetry values for each combination of sex, replicate population, and selection regime were calculated (N = 16) and then were analysed using a factorial anova in JMP, with sex (M or F), selection regime (C or ML), and their interaction (sex*sel) as fixed factors (this is equivalent to Levene’s test; Palmer & Strobeck, 2002).
Similarly, mean values for each combination of sex, replicate population, and selection regime were calculated (N = 16) for all other univariate traits (wing size, wing loading, body mass, allometry, and fitness) and then were analysed using a factorial anova in JMP, with sex (M or F), selection regime (C or ML), and their interaction (sex*sel) as fixed factors. This design is the same as that used for a previous analysis of data from these populations (Prasad et al., 2007). The mean values used in the analysis of univariate traits are reported in Supplementary Table S1. For the analysis of wing shape, we carried out a Mancova analysis of a similar design, but with centroid size included as a covariate to control for allometry. Because the Mancova was performed on mean values, there were too few degrees of freedom to calculate standard multivariate statistics for this analysis when carried out on the matrix of all partial warps plus the uniform component. We therefore analysed shape using relative warps (i.e. principal components of shape) and included as many in the model as possible, under the constraints provided by the limited number of degrees of freedom. We were able to include the first 11 relative warps (of 18) as dependent variables in the model, which explained over 95% of the variation in shape in our dataset.
- Top of page
- Supporting Information
We found evidence of phenotypic masculinization as a result of ML evolution for all univariate traits. Males had smaller wings than females (Table 1a, Fig. 2a), lower body mass (Table S2a, Fig. S1a), and lower wing loading (Table S2b, Fig. S1b), and parallel changes were seen as a result of ML evolution such that ML individuals of both sexes had smaller wings (Table 1a, Fig. 2a), lower body mass (Table S2a, Fig. S1a), and lower wing loading (Table S2b, Fig. S1b) than Controls. The difference between the sexes in the allometric relationship between wing size and body mass was not significant, but the change in this relationship as a result of ML evolution was still in the direction of extant sexual dimorphism (Table 1B, Fig. 2b), mostly due to an increase in slope in ML females. There were no significant sex*sel interactions for any of these traits, indicating that the degree of sexual dimorphism was unchanged as a result of ML evolution.
|(a) Wing size|
|(c) Relative fitness|
|(d) Wing size asymmetry|
Both the sexes and the selection treatments differed in wing shape (Table 2), and qualitatively similar patterns of phenotypic masculinization appeared to have been achieved via different evolutionary pathways. In males, the size of the proximal part of the wing was reduced and the distal part was increased relative to females (Fig. 1b). A similar pattern of reduction in the proximal part of the wing and increase in the distal part was seen in ML individuals relative to Controls (Fig. 1c), but this general result was achieved via a different pattern of displacement of wing vein intersections compared to the difference because of sexual dimorphism. Again, there was no indication of any change in the degree of sexual dimorphism in shape for ML individuals. This means that although the visualization in Fig. 1c was calculated using pooled data from both sexes, the pattern is the same even if the sexes are plotted separately (consistent with the nonsignificant sex*selection interaction term in Table 2).
|Effect||Num d.f.||Den d.f.||Wilks’λ||F-ratio||P-value|
We also found increased fitness in ML males and decreased fitness of females carrying ML-evolved chromosomes, consistent with earlier results from this system (Prasad et al., 2007; Table 1c, Fig. 2c). Interestingly, there was a significant sex*selection interaction effect in FA (Table 1d): the rank order of ML and C groups switched between the sexes (Fig. 2d) such that ML males had lower FA than C males, while the opposite was true for females. This pattern paralleled the changes seen in fitness (Fig. 2c) rather than size (Fig. 2a). ML-expressing males were more symmetrical for wing size than Control males were; however, females showed decreased developmental stability (higher size FA) when they carried ML chromosomes, despite being smaller than control females (Fig. 2a, Table 1).
- Top of page
- Supporting Information
We reproduce the earlier result that ML selection leads to increased total fitness of males and decreased fitness of females experimentally expressing ML chromosomes. We also found support for our two specific predictions about the evolution of size and wing morphology. First, ML males were indeed more symmetrical than C males, reflecting higher developmental stability. Second, we found that ML evolution proceeded in the direction of extant sexual dimorphism for all univariate traits and that wing shape evolution evolved in a manner qualitatively similar to the direction of sexual dimorphism. However, the change in wing shape as a result of ML evolution was achieved through a different pattern of displacement of wing vein intersections relative to the difference in shape between males and females. These results suggest that the average male in the ancestor or control populations is displaced from the optimal phenotype, presumably by counter selection in females because evolution in wing morphology occurred once selection on females was removed. Hence, although the effects of selection regime were still generally smaller than sex differences, we saw morphological evidence for a gender load resulting from intralocus sexual conflict.
Results on allometric relationship between wing size and body mass suggest both that a number of inter-related aspects of the developmental programme have changed as a result of ML evolution and that a reduction in body size is not the proximal explanation for the evolution of smaller wings in ML individuals. Our results also provide further experimental evidence that intersexual genetic correlations for wing size/shape and body mass traits must be high, as there was no change in the degree of SSD as a result of ML evolution for these traits (no significant sex*sel interactions, Table 1a,b, Table 2, and Table S2a,b). This is consistent with previous research on D. melanogaster, which has shown that intersexual genetic correlations for wing and body size traits generally range from 0.6 to 1 (Cowley et al., 1986; Cowley & Atchley, 1988; Reeve & Fairbairn, 1996; Karan et al., 1999, 2000), with a mean around 0.8 (Poissant et al., 2009; Supporting Information).
Previous analysis of wing shape in a number of Drosophila species suggests that wing morphology is relatively evolutionarily labile (Gidaszewski et al., 2009), and this is consistent with our results because differences in wing size, wing shape, wing loading, and allometry evolved on a short time scale. However, the lack of change of the degree of wing shape dimorphism as a result of ML evolution suggests that intersexual genetic correlations for shape are high. Shape changes should therefore evolve much more readily as a result of sexually congruent selection than as a result of sexually antagonistic selection. Wing loading is a trait that exhibits both plastic and genetic variation (Gilchrist & Huey, 2004; Frazier et al., 2008; Powell et al., 2010), so the observed change in wing loading on a short time scale seen here is consistent with previous results but is (to our knowledge) novel in detecting changes in wing loading because of sexual selection rather than because of ecological adaptation. The wing shape results also suggest that a functionally similar result (i.e. a decrease in the area of the proximal part of the wing and increase in the area of the distal part of the wing) has been achieved via different ontogenetic pathways. This is consistent with previous results for wing size evolution in Drosophila, where analogous clines in wing size are found in European and North American populations, but the clines are a result of size increases in different portions of the wing on each continent (Gilchrist et al., 2001). Similarly, differences in wing size can be a result of differences either in cell size or in cell number, and contrasting patterns have been found in natural populations (James et al., 1995) and as a result of selection experiments (Partridge et al., 1994). There do not seem to be strong constraints on the evolution of wing morphology in Drosophila (Mezey & Houle, 2005; Gidaszewski et al., 2009), so these examples of functionally similar trait values achieved in different ways (both from previous research and from our own results) are probably the result of differences in time scale. Divergence on short time scales (i.e. in the laboratory or in new environments) should proceed in the direction of the most readily available genetic variation (that is, along evolutionary lines of least resistance, Schluter, 1996), whereas divergence on longer (evolutionary) time scales should result in optimization of trait values.
Our results also raise several important questions about the genetic basis of developmental stability, as well as potential causal relationships between FA and fitness. Stressful conditions can increase FA (Parsons, 1992; Santos et al., 2006; Soto et al., 2008), so the increase in wing size FA in ML females is consistent with the idea that phenotypic masculinization is stressful for females. An alternative explanation for increased FA in females would be that the ML treatment alters the mutation-selection balance in populations, so that females are free to accumulate mutations at female sex-limited loci. This would make reduced fitness and increased FA a by-product of mutation accumulation at female-specific loci. While we cannot discount this hypothesis outright, only a small proportion of loci are expected to be female limited (Parisi et al., 2003), and a previous analysis of the effects of sex-specific selection indicated that most of the decline in the unselected sex could be attributed to a combination of sexually antagonistic loci and mutations that were deleterious in both sexes (Morrow et al., 2008). The consistency of results across independent replicate populations also argues against mutation accumulation at female-limited and female-biased loci as the sole explanation for a reduction in female fitness under ML, although it certainly may have played a role. Similarly, although the ML evolution laboratory protocol does not preclude adaptation to the Y-chromosome and the translocated chromosomes 2 and 3 found in the clone-generator females (see Supporting Information for more details), such adaptation would not explain the sex-specific nature of the fitness and FA results. The selection for perfection model suggests that males should be selected for increased developmental stability relative to females, but other studies have found higher FA in males in a number of different taxa (Söderman et al., 2006; Vishalakshi & Singh, 2006; Breuker et al., 2007; Davis & Grosse, 2008; Bonduriansky, 2009), and mean male wing size FA was indeed slightly higher than mean female wing size FA in our Control populations. This makes the increase in developmental stability we observed in ML males particularly striking, as it suggests that intralocus sexual conflict is an important factor in determining levels of developmental stability between the sexes.
The role of FA in mate choice has been widely discussed, and in particular, the application of this population parameter to the study of individual variation has been called into question (e.g. Houle, 1998; but see also Hansen et al., 2006). We unfortunately cannot deduce from the data at hand whether wing size FA contributed directly to increases in ML male fitness via female choice of more symmetrical males or increased success in intrasexual competition (Møller & Thornhill, 1998). Alternatively, FA may simply serve as an indicator trait of high genetic quality/attractiveness, for example, if FA is not under direct selection but is negatively correlated with other sexually selected traits (Markow & Ricker, 1992; Bonduriansky, 2009). ML males evolved increased fitness through higher mating frequency, and behavioural observations have shown that they obtain matings with females with lower courtship effort per copulation (Bedhomme et al., 2008). This does not appear to be related to differences between ML and C populations in CHCs (cuticular hydrocarbons; S. Bedhomme, A.K. Chippindale, N.G. Prasad, M. Delcourt, J.K. Abbott, M.A. Mallet and H.D. Rundle, unpublished data), so we can conclude that some other aspect of attractiveness or general vigour related to precopulatory sexual selection has improved. Interestingly, recent research has shown that in mice, loci coding for environmental robustness (insensitivity of the trait to environmental variation) are almost universally sex specific (Fraser & Schadt, 2010). Whether this is also true in Drosophila is currently unknown, but sex specificity of environmental robustness loci is certainly consistent with our results.
Intralocus sexual conflict will manifest itself when positive intersexual genetic correlations prohibit a response to disruptive selection on the sexes for different phenotypic optima. Consistent with this, ML selection not only led to smaller males, but also to increased development time, reflecting a decrease in growth rate through both of its components. At the same time, the wing generally evolved increased phenotypic masculinization (in terms of both size and shape), and the developmental stability of ML males increased. Both of these general results were consistent with our expectations from the selection for perfection model discussed earlier. Because we saw coordinated changes in female morphology when expressing ML chromosomes, but reduced fitness and lower levels of developmental stability, this provides experimental evidence of strong intersexual genetic correlations for the characters themselves but to differing mechanisms of homoeostasis in growth and ontogeny within the two sexes.
- Top of page
- Supporting Information
Financial support was provided by the Swedish Research Council (to JKA), NSERC (to AKC), a Lavoisier Award from the French government (to SB), and by Queen’s University ARC Awards (to JKA and SB). Thanks to Rhonda Snook three anonymous reviewers and Nelly Gidaszewski for helpful comments, to Göran Arnqvist for useful suggestions regarding the analysis of wing shape, and Lea Bond for the use of the microbalance.
- Top of page
- Supporting Information
- 2008. Reproductive behavior evolves rapidly when intralocus sexual conflict is removed. PLoS ONE 3: e2187. , , &
- 2007. Proximate causes of Rensch’s rule: does sexual size dimorphism in arthropods result from sex differences in development time? Am. Nat. 169: 245–257. , , , , , , , , , , , &
- 2009. Condition dependence of developmental stability in the sexually dimorphic fly Telostylinus angusticollis (Diptera: Neriidae). J. Evol. Biol. 22: 861–872.
- 2009. Intralocus sexual conflict. Trends Ecol. Evol. 24: 280–288. &
- 2006. A single basis for developmental buffering of Drosophila wing shape. PLoS ONE 1: 7. , &
- 2007. The association between wing morphology and dispersal is sex-specific in the glanville fritillary butterfly Melitaea cinxia (Lepidoptera: Nymphalidae). Eur. J. Entomol. 104: 445–452. , &
- 2003. The devil in the details of life-history evolution: instability and reversal of genetic correlations during selection on Drosophila development. J. Genet. 82: 133–145. , &
- 1998. The genetic basis of developmental stability. IV. Individual and population asymmetry parameters. Heredity 80: 553–561.
- 1988. Quantitative genetics of Drosophila melanogaster. II. Heritabilities and genetic correlations between sexes for head and thorax traits. Genetics 119: 421–433. &
- 1986. Quantitative genetics of Drosophila melanogaster. I. Sexual dimorphism in genetic parameters for wing traits. Genetics 114: 549–566. , &
- 2009. Sexually antagonistic selection, sexual dimorphism, and the resolution of intralocus sexual conflict. Am. Nat. 173: 176–187. &
- 2008. Measuring fluctuating asymmetry in plastron scutes of yellow-bellied sliders: the importance of gender, size and body location. Am. Midl. Nat. 159: 340–348. &
- 2010. The quantitative genetics of phenotypic robustness. PLoS ONE 5: e8635. &
- 2008. Cold-rearing improves cold-flight performance in Drosophila via changes in wing morphology. J. Exp. Biol. 211: 2116–2122. , , &
- 2009. Evolution of sexual diimorphism of wing shape in the Drosophila melanogaster subgroup. BMC Evol. Biol. 9: 110. , &
- 2004. Plastic and genetic variation in wing loading as a function of temperature within and among parallel clines in Drosophila subobscura. Integr. Comp. Biol. 44: 461–470. &
- 2001. Rapid evolution of wing size clines in Drosophila subobscura. Genetica 112–113: 273–286. , &
- 2006. On adaptive accuracy and precision in natural populations. Am. Nat. 168: 168–181. , &
- 1998. High enthusiasm and low r-squared. Evolution 52: 1872–1876.
- 1995. Cellular basis and developmental timing in a size cline of Drosophila melanogaster. Genetics 140: 659–666. , &
- 1999. Body size reaction norms in Drosophila melanogaster: temporal stability and genetic architecture in a natural population. Genet. Sel. Evol. 31: 491–508. , , , &
- 2000. The genetics of phenotypic plasticity. IX. Genetic architecture, temperature, and sex differences in Drosophila melanogaster. Evolution 54: 1035–1040. , , , , &
- 1998. Geometric morphometrics of developmental instability: analyzing patterns of fluctuating asymmetry with Procrustes methods. Evolution 52: 1363–1375. &
- 2007. Adult locomotory activity mediates intralocus sexual conflict in a laboratory-adapted population of Drosophila melanogaster. Proc. R. Soc. Lond. B Biol. Sci. 274: 3105–3112. &
- 2004. Sexual selection for increased male body size and protandry in a spider. Anim. Behav. 68: 1041–1048. , &
- 1992. Male size, developmental stability, and mating success in natural populations of three Drosophila species. Heredity 69: 122–127. &
- 2005. The dimensionality of genetic variation for wing shape in Drosophila melanogaster. Evolution 59: 1027–1038. &
- 2003. Sperm-female coevolution in Drosophila. Science 298: 1230–1233. &
- 1998. Bilateral symmetry and sexual selection: a meta-analysis. Am. Nat. 151: 174–192. &
- 2008. Assessing the extent of genome-wide intralocus sexual conflict via experimentally enforced gender-limited selection. J. Evol. Biol. 21: 1046–1054. , &
- 2002. Climbing to reach females: Romeo should be small. Evolution 56: 420–425. , &
- 1994. Fluctuating asymmetry analyses: a primer. In: Developmental Instability: Its Origins and Evolutionary Implications (T.A.Markow, ed.). Kluwer, Dordrecht, Netherlands, pp. 335–364.
- 2002. Fluctuating asymmetry analyses revisited. In: Developmental Instability: Causes and Consequences (M.Polak, ed.), pp. 279–319. Oxford University Press, Oxford, UK. &
- 2003. Paucity of genes on the Drosophila X chromosome showing male-biased expression. Science 299: 697–700. , , , , , , &
- 1992. Fluctuating asymmetry: a biological monitor of environmental and genomic stress. Heredity 68: 361–368.
- 1994. Evolution and development of body size and cell size in Drosophila melanogaster in response to temperature. Evolution 48: 1269–1276. , , &
- 2009. Sex-specific genetic variance and the evolution of sexual dimorphism: a systematic review of cross-sex genetic correlations. Evolution 64: 97–107. , &
- 2010. Phenotypic plasticity across 50 MY of evolution: drosophila wing size and temperature. J. Insect Physiol. 56: 380–382. , &
- 2007. An evolutionary cost of separate genders revealed by male-limited expression. Am. Nat. 169: 29–37. , , &
- 1996. Sexual size dimorphism as a correlated response to selection on body size: an empirical test of the quantitative genetic model. Evolution 50: 1927–1938. &
- 1984. Sex chromosomes and the evolution of sexual dimorphism. Evolution 38: 735–742.
- 1996. Sexually antagonistic male adaptation triggered by experimental arrest of female evolution. Nature 381: 232–234.
- 2006. Thermal evolution of pre-adult life history traits, geometic size and shape, and developmental stability in Drosophila subobscura. J. Evol. Biol. 19: 2006–2021. , &
- 1996. Adaptive radiation along genetic lines of least resistance. Evolution 50: 1766–1774.
- 2006. Environmental stress increases skeletal fluctuating asymmetry in the moor frog Rana arvalis. Oecologia 151: 593–604. , , &
- 2008. Wing morphology and fluctuating asymmetry depend on the host plant in cactophilic Drosophila. J. Evol. Biol. 21: 298–609. , , &
- 1988. Sexual selection in a natural population of Drosophila melanogaster. Evolution 42: 197–199. &
- 2006. Fluctuating asymmetry in certain morphological traits in laboratory populations of Drosophila ananassae. Genome 49: 777–785. &
- Top of page
- Supporting Information
Data S1 Methods.
Data S2 Results.
Table S1 Means for each combination of population, sex, and selection regime for all univariate traits. Loading is short for wing loading.
Table S2 Statistical significance of analysis of (A) Body mass, and (B) Wing loading.
Figure S1 Differences between the sexes and experimental groups in (a) Dry body mass, and (b) Wing loading.
As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.
|JEB_2064_sm_Table-S1-2-Figure-S1.doc||167K||Supporting info item|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.