THE DYNAMICS OF RECIPROCAL SELECTIVE SWEEPS OF HOST RESISTANCE AND A PARASITE COUNTER‐ADAPTATION IN DROSOPHILA

Host–parasite coevolution can result in consecutive selective sweeps of host resistance alleles and parasite counter‐adaptations. To illustrate the dynamics of this important but little studied form of coevolution, we have modeled an ongoing arms race between Drosophila melanogaster and the vertically transmitted sigma virus, using parameters we estimated in the field. We integrate these results with previous work showing that the spread of a resistance allele of the ref(2)P gene in the host was followed by the spread of a virus genotype, which overcomes this resistance. In line with these observations, our model predicts that there can be rapid selective sweeps in both the host and parasite, which can drive large changes in the prevalence of infection. The virus will tend to be ahead in the arms race, as incomplete dominance slows down host adaptation and selection for host resistance is weaker than selection for parasites to overcome resistance—the “life‐dinner” principle. This asymmetry in the adaptation rates results in a partial sweep of the host resistance allele, as it loses its advantage part way through the selective sweep. This well‐understood natural system illustrates how the outcome of host–parasite coevolution is determined by different population genetic parameters in the field.     

A TEST OF THE ROLE OF EPISTASIS IN DIVERGENCE UNDER UNIFORM SELECTION

Five populations of Drosophila melanogaster have previously been shown to be replicably different in their responses to artificial selection for knockdown resistance to ethanol fumes (Cohan and Hoffmann, 1986). The present study tests whether this divergence could be attributed to the epistatic mechanism assumed by Wright's shifting-balance model of evolution, in which alleles favored in the genetic background of one population are not favored in that of another. If this were the mechanism of divergence, crosses between selected lines from different populations would be expected to yield an epistatic loss of the selected phenotype. However, all such crosses showed a good fit to an additive model with dominance. Divergence by an epistatic mechanism may also be associated with epistatic variance within populations, but no evidence for such epistasis was found. The populations therefore appear to have responded in different ways to selection not because of epistasis but because knockdown-resistance alleles that were common in some populations were absent (or at least less common) in others.     

Differential responses to artificial selection on oviposition site preferences in Drosophila melanogaster and D. simulans

The preference–performance relationship in plant–insect interactions is a central theme in evolutionary ecology. Among many insects, eggs are vulnerable and larvae have limited mobility, making the choice of an appropriate oviposition site one of the most important decisions for a female. We investigated the evolution of oviposition preferences in Drosophila melanogaster Meigen and Drosophila simulans Sturtevant by artificially selecting for the preference for 2 natural resources, grape and quince. The main finding of our study is the differential responses of D. melanogaster and D. simulans. Although preferences evolved in the experimental populations of D. melanogaster, responses were not consistent with the selection regimes applied. In contrast, responses in D. simulans were consistent with expectations, demonstrating that this species has selectable genetic variation for the trait. Furthermore, crosses between D. simulans divergent lines showed that the genetic factors involved in grape preference appear to be largely recessive. In summary, our artificial selection study suggests that D. melanogaster and D. simulans possess different genetic architectures for this trait.     

Selection for starvation resistance in Drosophila melanogaster : physiological correlates,enzyme activities and multiple stress responses

Correlated responses to artificial selection for stress tolerance can provide insight into underlying genetic variation and the physiological basis of stress resistance . Lines of Drosophila melanogaster held in the absence of food or with an unsuitable resource, specifically decomposing lemon, responded to selection by becoming starvation resistant. The lemon-selected lines also adapted by evolving a resource-based induction response. Compared to control lines, the selected lines tended to store more lipid, develop slower and have a larger body size. Additional responses included resistance to desiccation and acetone fumes, suggesting multiple stress resistance is a correlated result of selection for starvation resistance. The specific metabolic rate was lower in the starvation selected lines and enzyme activities changed in response to selection. In particular, enzyme activities indirectly associated with lipid biogenesis increased in both types of selected lines. The correlated responses to the two selection regimes were sufficiently consistent to indicate a common basis for starvation resistance. Specific responses to starvation selection appeared to oppose the short-term phenotypic responses to starvation. Thus, a common response to stress selection may be to ameliorate the immediate physiological impact of the stress factor.     

EVOLUTIONARY GENETICS OF OVIPOSITION PREFERENCE IN SWALLOWTAIL BUTTERFLIES

Models of the evolution of host shifts and speciation in phytophagous insects are often built upon the assumption that host selection is under simple genetic control, perhaps even a single locus. The genetic basis for differences in host-plant preference by ovipositing insects was investigated using two closely related species of swallowtail butterfly, Papilio oregonius and P. zelicaon, which differ in the plant families on which females oviposit. Both species had been shown previously to vary within populations in host selection. A combination of analyses using reciprocal interspecific crosses and isofemale strains within populations indicated that oviposition preference in these species is determined significantly by one or more loci on the X chromosome, which female Lepidoptera inherit only paternally. Hence, preferences in hybrid females tended toward the paternal species. This is the first insect group for which partial control of oviposition preference has been localized onto a particular chromosome. In addition, one or more loci on another chromosome(s) appear to contribute to preference, as indicated by the partially intermediate preferences of some hybrid crosses. The overall differences in preference in the reciprocal interspecific hybrids were restricted to differences in the distribution of eggs laid among the local host plants of these two Papilio species; the reciprocal crosses did not differ in the small percentage of eggs laid on a novel potential host species. The variation in host selection found among the isofemale strains reinforces earlier results for these strains, indicating that there is genetic variation in host selection within these populations. Overall, the results indicate that the evolution of oviposition preference in these species involves genetic changes at two or more chromosomes with the X chromosome playing an important role in determining preference.     

Changes Across Development Influence Visible and Cryptic Natural Variation of <Emphasis Type="Italic">Drosophila melanogaster</Emphasis> Olfactory Response

Relative to an equivalent source of variation that do not present a hidden state, cryptic genetic variation is likely to be an effective source for possible adaptations at times of atypical environmental conditions. In addition to environmental perturbations, it has also been proposed that genetic disturbances can generate release of cryptic genetic variation. The genetic basis and physiology of olfactory response in Drosophila melanogaster is being studied profusely, but almost no analysis has addressed the question if populations harbor cryptic genetic variation for this trait that only manifests when populations experiences a typical or novel conditions. We quantified olfactory responses to benzaldehyde in both larval and adult lifecycle stages among samples of chromosome two substitution lines extracted from different natural populations of Argentina and substituted into a common inbred background. We also evaluated whether an effect of genetic background change, occurred during chromosome substitution, affect larval and adult olfactory response in terms of release of cryptic genetic variation. Results indicate the presence of genetic variation among chromosome substitution lines in both lifecycle stages analyzed. The comparative analyses between chromosome 2 substitution lines and isofemale lines used to generate the chromosome 2 substitution lines shown that only adults exhibited decanalizing process for olfactory response to benzaldehyde in natural populations of D. melanogaster, i.e., release of hidden genetic variation. We propose that this release of hidden genetic variation in adult flies is a consequence of the shift in genetic background context that happens in chromosome 2 substitution lines, that implies the disruption of natural epistatic interactions and generation of novel ones. All in all, we have found that changes across D. melanogaster development influence visible and cryptic natural variation of olfactory behavior. In this sense, changes in the genetic background can affect gene-by-gene interactions (epistasis) generating different or even novel phenotypes as consequence of phenotypic outcome of cryptic genetic variation.     

EVOLUTION OF RESISTANCE TO A MULTIPLE‐HERBIVORE COMMUNITY: GENETIC CORRELATIONS,DIFFUSE COEVOLUTION,AND CONSTRAINTS ON THE PLANT'S RESPONSE TO SELECTION

Although plants are generally attacked by a community of several species of herbivores, relatively little is known about the strength of natural selection for resistance in multiple‐herbivore communities—particularly how the strength of selection differs among herbivores that feed on different plant organs or how strongly genetic correlations in resistance affect the evolutionary responses of the plant. Here, we report on a field study measuring natural selection for resistance in a diverse community of herbivores of Solanum carolinense. Using linear phenotypic‐selection analyses, we found that directional selection acted to increase resistance to seven species. Selection was strongest to increase resistance to fruit feeders, followed by flower feeders, then leaf feeders. Selection favored a decrease in resistance to a stem borer. Bootstrapping analyses showed that the plant population contained significant genetic variation for each of 14 measured resistance traits and significant covariances in one‐third of the pairwise combinations of resistance traits. These genetic covariances reduced the plant's overall predicted evolutionary response for resistance against the herbivore community by about 60%. Diffuse (co)evolution was widespread in this community, and the diffuse interactions had an overwhelmingly constraining (rather than facilitative) effect on the plant's evolution of resistance.     

AN INTERACTION BETWEEN ENVIRONMENTAL TEMPERATURE AND GENETIC VARIATION FOR BODY SIZE FOR THE FITNESS OF ADULT FEMALE DROSOPHILA MELANOGASTER

Abstract. — Drosophila and other ectotherms show geographic genetic variation in body size, with larger individuals at higher latitudes and altitudes. Temperature is implicated as an important selective agent because long-term laboratory culture of Drosophila leads to the evolution of larger body size at lower temperatures. In this paper, we tested the hypothesis that, in Drosophila melanogaster, larger size is favored at lower temperatures in part because of selection on adult females. We used replicated lines of D. melanogaster artificially selected for increased and decreased wing area with constant cell area. The resulting size differences between the selected lines were due solely to differences in cell number, and thereby were similar to the cellular basis of clinal variation in body size in nature. We examined life-history traits of adult females at 18 and 25°C. Rearing for two generations at the two temperatures did not affect the extent of the size differences between lines from the different selection regimes. There was a strong interaction between temperature and size selection for both survival and lifetime reproductive success, with larger females living significantly longer and producing more offspring over their lifetime only when reared and tested in the colder environment. There was also an increase in average daily progeny production in large-line females relative to the control and small lines again, only in the colder environment. Thus, the females from the large selection lines were relatively fitter at the colder temperature. At both experimental temperatures, especially the lower one, the small- line females rescheduled their progeny production to later ages. Larger body size may have evolved at higher latitudes and altitudes because of the advantages to the adult female of being larger at lower temperatures.     

ANALYSIS OF THE GENETIC STRUCTURE OF LIFE HISTORY OF DROSOPHILA MELANOGASTER USING RECOMBINANT EXTRACTED LINES

A model for the evolution of senescence known as “antagonistic pleiotropy” makes the specific prediction that there should be a negative genetic correlation between early- and late-age traits associated with fitness. This model has previously been tested by classical quantitative-genetic means including sib-analysis and artificial selection. We used the approach of chromosome extraction, which has both advantages and disadvantages compared to classical techniques, to test the model further. From four isogenic lines of Drosophila melanogaster, four sets of recombinant extracted lines were constructed using standard balancer-chromosome techniques. The four parental lines and 53 recombinants were reared under controlled laboratory conditions and isolated as pairs for scoring daily fecundity and longevity. Even though the design is not optimal for estimating classical components of genetic variance, it afforded a uniquely direct test of the magnitude of environmental covariances, while giving a detailed genetic picture of part of the genome. There were clear differences among the recombinant series in the distribution of mean longevity and early fecundity. The genetic correlation between early fecundity (sum of egg production for the first five days posteclosion) and female longevity was significantly negative in only one of the recombinant series. When all lines were considered together, the phenotypic correlation between these traits was significantly negative (P < 0.02), while the broad-sense genetic correlation was –0.219 (P < 0.11). This result may be viewed as weakly consistent with the model of antagonistic pleiotropy, but other aspects of the data are at odds with the model.     

Directional selection reduces developmental canalization against genetic and environmental perturbations in Drosophila wings

Natural selection may enhance or weaken the robustness of phenotypes against genetic or environmental perturbations. However, important aspects of the relationship between adaptive evolution and canalization remain unclear. Recent work showed that the evolution of larger wing size in a high altitude natural population of Drosophila melanogaster was accompanied by decanalized wing development–‐specifically a loss of robustness to genetic perturbation. But this study did not address environmental robustness, and it compared populations that may have numerous biological differences. Here, we perform artificial selection on this same trait in D. melanogaster (larger wing length) and directly test whether this directional selection resulted in decanalization. We find that in general, size‐selected replicates show greater frequencies of wing defects than control replicates both after mutagenesis (genetic perturbation) and when subjected to high temperature stress (environmental perturbation), although the increase in defect frequency varies importantly among replicates. These results support the hypothesis that directional selection may result in the loss of both genetic and environmental robustness–offering a rare window into the relationship between adaptation and canalization.     

Genome-Wide Association Studies Reveal a Simple Genetic Basis of Resistance to Naturally Coevolving Viruses in Drosophila melanogaster

Variation in susceptibility to infectious disease often has a substantial genetic component in animal and plant populations. We have used genome-wide association studies (GWAS) in Drosophila melanogaster to identify the genetic basis of variation in susceptibility to viral infection. We found that there is substantially more genetic variation in susceptibility to two viruses that naturally infect D. melanogaster (DCV and DMelSV) than to two viruses isolated from other insects (FHV and DAffSV). Furthermore, this increased variation is caused by a small number of common polymorphisms that have a major effect on resistance and can individually explain up to 47% of the heritability in disease susceptibility. For two of these polymorphisms, it has previously been shown that they have been driven to a high frequency by natural selection. An advantage of GWAS in Drosophila is that the results can be confirmed experimentally. We verified that a gene called pastrel—which was previously not known to have an antiviral function—is associated with DCV-resistance by knocking down its expression by RNAi. Our data suggest that selection for resistance to infectious disease can increase genetic variation by increasing the frequency of major-effect alleles, and this has resulted in a simple genetic basis to variation in virus resistance.     

Non-linear selection response in Drosophila: a strategy for testing the rare-alleles model of quantitative genetic variability

Quantitative genetic theory predicts that variation due to rare alleles at many loci will generate a transient acceleration in the response to directional selection. We have tested this prediction by constructing experimental lines ofDrosophila melanogaster that carry positively selected ethanol resistance alleles at low frequencies, and then subjecting the lines to directional selection for ethanol resistance. Approximately 468,000 files were subjected to artificial selection over 30 generations. The predicted non-linear selection responses were observed in all experimental lines and replicates, on three genetic backgrounds. In contrast, un-selected controls and lines carrying random alleles at low frequencies on the same genetic backgrounds exhibited linear selection responses. These results demonstrate that non-linearities due to rare alleles are detectable and repeatable, provided that experiments are done on a sufficiently large scale. The results suggest that it may be possible to test for rare-alleles as a component of naturally occurring genetic variation by careful examination of selection response curves.     

The molecular genetics of clinal variation: a case study of ebony and thoracic trident pigmentation in Drosophila melanogaster from eastern Australia

Widespread pigmentation diversity coupled with a well‐defined genetic system of melanin synthesis and patterning in Drosophila provides an excellent opportunity to study phenotypes undergoing evolutionary change. Pigmentation variation is highly correlated with different ecological variables and is thought to reflect adaptations to different environments. Several studies have linked candidate genes from Drosophila melanogaster to intra‐population variation and interspecific morphological divergence, but less clearly to variation among populations forming pigmentation clines. We characterized a new thoracic trident pigmentation cline in D. melanogaster populations from eastern Australia, and applied a candidate gene approach to explain the majority of the geographically structured phenotypic variation. More melanized populations from higher latitudes tended to express less ebony than their tropical counterparts, and an independent artificial selection experiment confirmed this association. By partitioning temperature dependent effects, we showed that the genetic differences underlying clinal patterns for trident variation at 25 °C do not explain the patterns observed at 16 °C. Changes in thoracic trident pigmentation could be a common evolutionary response to climatically mediated environmental pressures. On the Australian east coast most of the changes appear to be associated with regulatory divergence of the ebony gene but this depends on temperature.     

THE INCREASED POTENTIAL FOR SELECTION OF THE LAC OPERON OF ESCHERICHIA COLI

The fitness effects of six lac operons from natural isolates of Escherichia coli were determined in chemostats, in a test of the idea that selective differences among natural alleles are greater in novel conditions than in the prevailing environment, resulting in a greater genetic variance in fitness in novel conditions. Fitnesses were determined in the common milk sugar lactose, the natural substrate of the lac operon, and in three rare β-galactosides, lactulose, galactosyl-arabinose, and methyl-galactopyranoside, that are novel for E. coli. Significantly greater fitness differences were observed among the lac alleles in each of the novel β-galactosides than in lactose. An alternative explanation of the experimental findings is discussed. General evolutionary causes and consequences of selection potentials are discussed, and an outline of the work necessary to further elucidate the physiological basis of the observed potential for selection of the lac operon of E. coli is presented.     

LATITUDINAL CLINE IN DROSOPHILA MELANOGASTER FOR KNOCKDOWN RESISTANCE TO ETHANOL FUMES AND FOR RATES OF RESPONSE TO SELECTION FOR FURTHER RESISTANCE

We have introduced a device for selecting Drosophila for increased resistance to very high concentrations of ethanol fumes. This device has enabled us to: 1) select quickly and easily over a thousand flies at a time, and 2) score the knockdown time of every fly in the distribution, while causing very little injury to the flies. A sample of nine west coast populations of Drosophila melanogaster showed a significant trend toward higher knockdown resistance in more northern populations. A population's level of knockdown resistance was virtually uncorrelated with its alcohol dehydrogenase (Adh) allele frequencies. Five of the above nine populations were then subjected to selection for further knockdown resistance. Each population was divided randomly into four groups of 256 flies: two lines to be selected, and two lines to remain unselected as control lines. In every generation each selected line was measured for knockdown resistance, and the last quartile of flies to be knocked down was saved to continue the selection cycle. Population sizes of the selected and unselected lines were all maintained at 256. Realized heritability, based on the responses to selection of the first four generations, was calculated for each selected line. The five populations were significantly heterogeneous for heritability estimates; the average heritability of the five populations pooled was 0.143 ± 0.019. Over the course of twelve generations, the ten selected lines increased their knockdown times by an average factor of 2.40. Before selection, the five populations were heterogeneous for knockdown resistance, and resistance was greatest among the most northern populations. The amount of change of knockdown resistance over the course of selection was also correlated with latitude: the most southern population increased its knockdown time by a factor of 2.23, and the most northern population increased it by a factor of 2.55. After ten generations of selection, the cline of knockdown resistance was about 4.5 times as steep as that before selection. Small phenotypic differences among populations before selection were thus exaggerated by the action of selection. The differences among populations in their rates of response to selection were attributed to genetic differences that existed before selection. The pattern of change of Adh frequencies over the course of selection was very inconsistent, both among and within populations. From this inconsistency of change of Adh alleles with selection, and the lack of correlation between Adh frequencies and knockdown resistance before selection, we concluded that Adh frequency changes could not have had much effect on the responses of the selected lines.     

PERSPECTIVE: A CRITIQUE OF SEWALL WRIGHT'S SHIFTING BALANCE THEORY OF EVOLUTION

We evaluate Sewall Wright's three-phase “shifting balance” theory of evolution, examining both the theoretical issues and the relevant data from nature and the laboratory. We conclude that while phases I and II of Wright's theory (the movement of populations from one “adaptive peak” to another via drift and selection) can occur under some conditions, genetic drift is often unnecessary for movement between peaks. Phase III of the shifting balance, in which adaptations spread from particular populations to the entire species, faces two major theoretical obstacles: (1) unlike adaptations favored by simple directional selection, adaptations whose fixation requires some genetic drift are often prevented from spreading by barriers to gene flow; and (2) it is difficult to assemble complex adaptations whose constituent parts arise via peak shifts in different demes. Our review of the data from nature shows that although there is some evidence for individual phases of the shifting balance process, there are few empirical observations explained better by Wright's three-phase mechanism than by simple mass selection. Similarly, artificial selection experiments fail to show that selection in subdivided populations produces greater response than does mass selection in large populations. The complexity of the shifting balance process and the difficulty of establishing that adaptive valleys have been crossed by genetic drift make it impossible to test Wright's claim that adaptations commonly originate by this process. In view of these problems, it seems unreasonable to consider the shifting balance process as an important explanation for the evolution of adaptations.     

LONG-TERM LABORATORY EVOLUTION OF A GENETIC LIFE-HISTORY TRADE-OFF IN DROSOPHILA MELANOGASTER. 2. STABILITY OF GENETIC CORRELATIONS

Experiments in laboratory populations of Drosophila melanogaster have shown a negative genetic correlation between early-life fecundity on the one hand and starvation resistance and longevity on the other. Selection for late-life reproductive success resulted in long-lived populations that had increased starvation resistance but diminished early-life fecundity relative to short-lived controls. This pattern of differentiation proved, however, to be unstable. When assayed in a standard high-fecundity environment, the relative early fecundity of the long- and short-lived stocks reversed over a decade. That is, the long-lived populations came to have greater relative early-life fecundity, late-life fecundity, longevity and starvation resistance. Nevertheless, when these populations were assayed in other assay environments, the original trade-off was still present. We investigated the genetic structure of the short- and long-lived populations, to ask whether the inconstancy of the trade-off, as inferred from among population comparisons, is reflected in the pattern of genetic correlations within populations. For this purpose, lines from each of the short- and long-lived populations that had been selected for starvation resistance were compared with unselected controls. The direct and correlated responses of these starvation selected populations suggest that (1) the original genetic trade-off was still present in the ancestral short- and long-lived populations, even when it was no longer apparent from their comparison; (2) the trade-off was present in both assay environments; and (3) selectable genotype × environment variation exists for early fecundity. We suggest that a failure of the pattern of differentiation among populations to reflect the pattern of genetic correlations, if common in natural populations, will prevent the reliable inference of genetic trade-offs from comparisons of most natural populations.     

Dissecting the Genetic Basis of a Complex cis-Regulatory Adaptation

Although single genes underlying several evolutionary adaptations have been identified, the genetic basis of complex, polygenic adaptations has been far more challenging to pinpoint. Here we report that the budding yeast Saccharomyces paradoxus has recently evolved resistance to citrinin, a naturally occurring mycotoxin. Applying a genome-wide test for selection on cis-regulation, we identified five genes involved in the citrinin response that are constitutively up-regulated in S. paradoxus. Four of these genes are necessary for resistance, and are also sufficient to increase the resistance of a sensitive strain when over-expressed. Moreover, cis-regulatory divergence in the promoters of these genes contributes to resistance, while exacting a cost in the absence of citrinin. Our results demonstrate how the subtle effects of individual regulatory elements can be combined, via natural selection, into a complex adaptation. Our approach can be applied to dissect the genetic basis of polygenic adaptations in a wide range of species.     

THE GENETIC BASIS OF SEXUAL ISOLATION BETWEEN DROSOPHILA MELANOGASTER AND D. SIMULANS

The genetic analysis of sexual isolation between the closely-related species Drosophila melanogaster and Drosophila simulans involved two experiments with no-choice tests. The efficiency of sexual isolation was measured by the frequency of courtship initiation and interspecific mating. We first surveyed the variation in sexual isolation between D. melanogaster strains and D. simulans strains of different geographic origin. Then, to investigate variation in sexual isolation within strains, we made F₁ diallel sets of reciprocal crosses within strains of D. melanogaster and D. simulans. The F₁ diallel progeny of one sex were paired with the opposite sex of the other species. The first experiment showed significant differences in the frequency of interspecific mating between geographic strains. There were more matings between D. simulans females and D. melanogaster males than between D. melanogaster females and D. simulans males. The second experiment uncovered that the male genotypes in the D. melanogaster diallel significantly differed in interspecific mating frequency, but not in courtship initiation frequency. The female genotypes in the D. simulans diallel were not significantly different in courtship initiation and interspecific mating frequency. Genetic analysis reveals that in D. melanogaster males sexual isolation was not affected by either maternal cytoplasmic effects, sex-linked effects, or epistatic interaction. The main genetic components were directional dominance and overdominance. The F₁ males achieved more matings with D. simulans females than the inbred males. The genetic architecture of sexual isolation in D. melanogaster males argues for a history of weak or no selection for lower interspecific mating propensity. The behavioral causes of variation in sexual isolation between the two species are discussed.     

COMPLEX TRADE-OFFS AND THE EVOLUTION OF STARVATION RESISTANCE IN DROSOPHILA MELANOGASTER

The measurement of trade-offs may be complicated when selection exploits multiple avenues of adaptation or multiple life-cycle stages. We surveyed 10 populations of Drosophila melanogaster selected for increased resistance to starvation for 60 generations, their paired controls, and their mutual ancestors (a total of 30 outbred populations) for evidence of physiological and life-history trade-offs that span life-cycle stages. The directly selected lines showed an impressive response to starvation selection, with mature adult females resisting starvation death 4–6 times longer than unselected controls or ancestors—up to a maximum of almost 20 days. Starvation-selected flies are already 80% more resistant to starvation death than their controls immediately upon eclosion, suggesting that a significant portion of their selection response was owing to preadult growth and acquisition of metabolites relevant to the stress. These same lines exhibited significantly longer development and lower viability in the larval and pupal stages. Weight and lipid measurements on one of the starvation-selected treatments (SB_1–5), its control populations (CB_1–5), and their ancestor populations (B_1–5) revealed three important findings. First, starvation resistance and lipid content were linearly correlated; second, larval lipid acquisition played a major role in the evolution of adult starvation resistance; finally, increased larval growth rate and lipid acquisition had a fitness cost exacted in reduced viability and slower development. This study implicates multiple life-cycle stages in the response to selection for the stress resistance of only one stage. Our starvation-selected populations illustrate a case that may be common in nature. Patterns of genetic correlation may prove misleading unless multiple pleiotropic interconnections are resolved.