Evolution Letters

The evolutionary potential of traits is governed by the amount of heritable variation available to selection. While this is typically quantified based on genetic variation in a focal individual for its own traits (direct genetic effects, DGEs), when social interactions occur, genetic variation in interacting partners can influence a focal individuals traits (indirect genetic effects, IGEs). Theory and studies on domesticated species have suggested IGEs can greatly impact evolutionary trajectories, but whether this is true more broadly remains unclear. Here we perform a systematic review and meta-analysis to quantify the amount of trait variance explained by IGEs and the contribution of IGEs to predictions of adaptive potential. We identified 180 effect sizes from 47 studies across 21 species and found that, on average, IGEs of a single social partner account for a small but statistically significant amount of phenotypic variation (0.03). As IGEs affect the trait values of each interacting group member and due to a typically positive - although statistically nonsignificant - correlation with DGEs (rDGE-IGE = 0.26), IGEs ultimately increase trait heritability substantially from 0.27 (narrow-sense heritability) to 0.45 (total heritable variance). This 66% average increase in heritability suggests IGEs can increase the amount of genetic variation available to selection. Furthermore, whilst showing considerable variation across studies, IGEs were most prominent for behaviours, and to a lesser extent for reproduction and survival, in contrast to morphological, metabolic, physiological, and development traits. Our meta-analysis therefore shows that IGEs tend to enhance the evolutionary potential of traits, especially for those tightly related to interactions with other individuals such as behaviour and reproduction. Lay SummaryPredicting evolutionary change is important for breeding better livestock and crops, for understanding how biodiversity arises and how populations respond to environmental change. Normally, these predictions are based on how the genetic variants in an organism influence its own traits (characteristics). However, when organisms socially interact, for instance by fighting or cooperating, then the genes in one individual can influence the traits of others, therefore affecting the potential for evolutionary change. We compared 47 studies across 21 animal species and found that the effect of the genes of a single social partner is small but statistically significant, while the total contribution of social genetic effects to evolutionary potential is large. These effects are particularly important for the evolution of animals behaviours and reproductive traits, but less so for other traits such as body size and physiology. We also found that, because an individual can interact with many others and influence them all, social interactions can substantially increase the potential for a population to evolve from generation to generation. Our results show how social interactions can potentially alter the evolution of those traits known to respond to social interactions in comparison to standard expectations.

The adaptive nature of phenotypic plasticity is widely documented in natural populations. However, little is known about the evolutionary forces that shape genetic variation in plasticity within populations. Here we empirically address this issue by testing the hypothesis that stabilizing selection shapes genetic variation in the anti-predator developmental plasticity of Daphnia pulex. The anti-predator morphological defense is characterized by pedestal and spikes that grow on the back of the Daphnia neck following exposure to predator cure. We characterized variation in this plasticity using a novel, high-throughput phenotyping method that describes the entire dorsal shape amongst >100 D. pulex strains originating from a natural population in the UK. We found low genetic diversity for morphological defenses among genetically diverse clones upon predation risk exposure. The strongest reduction in genetic variation was observed in areas of greatest phenotypic plasticity, which we interpret as evidence of stabilizing selection. By assessing among-clone variance in clonally related, field derived strains, we contrasted mutational variation (Vm) to standing variation (Vg). We found that Vg/Vm is lowest in areas of greatest plasticity. These data strongly suggest that stabilizing selection operates directly on phenotypic plasticity, providing a rare glimpse into the evolution of fitness related traits in natural populations.

Heritable variation in traits under natural selection is a prerequisite for evolutionary response. While it is recognised that trait heritability may vary spatially and temporally depending under which environmental conditions traits are expressed, less is known about the possibility that genetic variance contributing to the expected selection response in a given trait may vary at different stages of ontogeny. Specifically, whether different loci underlie the expression of a trait throughout development - thus providing an additional source of variation for selection to act on - is unclear. Here we show that the heritability (h2) of body size, an important life history trait, remains constant across ontogeny in a stickleback fish. Nevertheless, both analyses of quantitative trait loci (QTL) and genetic correlations across ages show that different chromosomes/loci contribute to this heritability in different ontogenic time-points. This suggests that body size can respond to selection at different stages of ontogeny but that this response is determined by different loci at different points of development. Hence, this illustrates the notion that diverse genetic architectures may underline similar (expected) phenotypic outcomes, and that similar selection pressures may lead to genetically heterogeneous responses depending on what life stage selection is acting on.

Hybridization is widely acknowledged as an important mechanism of acquiring adaptive variation. In Texas, the sunflower Helianthus annuus subsp. texanus is thought to have acquired herbivore resistance and morphological traits via introgression from a local congener, H. debilis. Here we test this hypothesis using whole genome sequencing data from across the entire range of H. annuus and possible donor species, as well as phenotypic data from a common garden study. We find that although it is morphologically convergent with H. debilis, H. a. texanus has conflicting signals of introgression. Genome wide tests (Pattersons D and TreeMix) only find evidence of introgression from H. argophyllus (sister species to H. annuus and also sympatric), but not H. debilis, with the exception of one individual of 109 analysed. We further scanned the genome for localized signals of introgression using PCAdmix and found minimal but non-zero introgression from H. debilis and significant introgression from H. argophyllus. Putative introgressions mainly occur in high recombination regions as predicted by theory if introgressed ancestry contains maladaptive alleles. To reconcile the disparate findings of our analyses, we discuss potential test-specific confounding features, including introgression from other taxa. Given the paucity of introgression from H. debilis, we argue that the morphological convergence observed in Texas is likely independent of introgression.

Hosts and their parasites and pathogens are locked in antagonistic co-evolution. The genetic consequence of this can be seen in the rates of adaptive evolution in immunologically important loci in many taxa. As the risk of disease transmission increases we might also expect to see greater rates of adaptive evolution on genes of immune function. The evolution of sociality and its elaborations in insects represent enormous shift in disease transmission risk. Here, we examine whether sociality in the bees corresponds to changes in the rate of adaptive evolution in both classical canonical immune genes, and genes with putative immune functions identified from meta-analyses of honey-bee transcriptomic responses to infection. We find that measures of gene-wide adaptive evolution do not differ among canonical immune, non-canonical candidate immune, and background gene sets, but that branch-site adaptive evolution does increase with sociality regardless of gene category. Solitary species have greater rates of adaptive evolution in canonical immune genes than background genes, supporting the suggestion that social immune mechanisms may instead be the site of host-pathogen co-evolution in social species. We identify three genes with putative roles in immunity that warrant further attention (Vitel-logenin Vg, disks large 1 tumour suppressor, and the uncharacterised protein LOC100577972). There are more gene family changes after the origin of sociality across all gene classes, with contractions occur-ring after the elaboration of sociality to complex eusociality. There are few genes or functions under adaptive selection that appear to be shared outside of specific lineages, suggesting that evolution of the immune system may be specific to individual species and their pathogen interactions. SignificanceInfectious disease drives rapid evolution of immune genes, but infection risk should be much higher in social species. To examine whether greater sociality drives faster immune system evolution we compared the rate of immune gene evolution in solitary, social, and highly eusocial bees. To account for possible novel immune genes in bees, we analysed classical immune genes alongside candidate immune genes inferred from other studies. Surprisingly, we find that solitary bees have the highest rate of immune gene evolution relative to background genes but that sociality is associated with rapid evolution across the whole genome. These findings suggest that 1) accelerated immune gene evolution is not universal, 2) immune gene evolution is moderated by sociality in that solitary species invest more into immune gene change, and 3) that social genomes are highly dynamic, which may obscure evolution at immunological loci. The types of immune genes and functions appear mostly lineage-specific, regardless of sociality, suggesting individual evolutionary his-tories exert more selection pressure than general patterns of greater pathogen exposure introduced by social living.

Seasonal polyphenisms are common across the animal and plant kingdom yet we understand the explicit interactions between genetics and environment for only a few taxa. Are the genomic regions and their variants associated with the trait the same or different across environments? Is the response to selection shared or different across different "background" selection environments? Offspring type in the sacoglossan sea slug Alderia willowi is a seasonally modulated interaction between genotype and phenotype that results in offspring of wildly different developmental trajectories. In a genome-wide association test I found 41 SNPs associated with offspring type. In an evolve and resequence experiment I found thousands of loci changed in frequency following selection. These loci were partially shared (37%) between low and high salinity. Of the 41 candidate SNPs identified in the GWAS only seven also showed significant allele frequency change across replicates in the selection experiments with four in high salinity, two in low and one in both. This reveals a broad pattern of allele frequency change that is largely unique to the environment in which selection for the same phenotype occurs. The results presented in this paper showcase the ability of phenotypic plasticity to move the phenotype independent of the genotype and thus maintain the polyphenism that is so striking in this species.

Understanding the genetic basis of behavioural variation among-individuals is vital for predicting if, when, and how quickly behaviour can evolve under selection. However, in heterogeneous environments, behavioural plasticity (a source of within-individual variation) may also contribute to the phenotypic variance that can be selected on. If so, a complete picture of evolutionary potential, requires estimation of genotype-by-environment interactions (GxE). Here we investigate the quantitative genetics of shy-bold behavioural variation in the red cherry shrimp, Neocaridina davidi, an emerging decapod model for behavioural, genetic, and ecotoxicological research. Using a suite of behaviours associated with shy-bold personality variation we demonstrate moderate to high behavioural repeatabilities and show how a multivariate approach allows characterising the shape, not just the amount, of variation. Using a half-sib full sib breeding design in which shrimp from known families were tested under either control conditions or with predator (fish) cues present, we jointly estimate the plastic response to elevated risk, and the contribution of genetic factors to phenotypic variance. We find that genetic variance does underpin among-individual differences in behaviour. We also find evidence of plasticity, with individual shrimp shifting towards a shyer, or more risk averse, average phenotype in the presence of fish cues. However, we found no variation in plasticity either among-individuals (IxE) or among-genotypes (GxE). This implies that average behaviour can evolve under predator-mediated selection, but further adaptive evolution of behavioural plasticity may be constrained by a lack of GxE.

The paradox of stasis - the unexpectedly slow evolution of heritable traits under direct selection - has been widely documented in the last few decades. This paradox is often particularly acute for body size, which is often heritable and where positive associations of size and fitness are frequently identified, but constraints to the evolution of larger body sizes are often not obvious. Here, we identify a trade-off between survival and size-dependent reproduction in Soay sheep (Ovis aries), contributes to selection against large body size. Using recently developed theory on non-linear developmental systems, then decompose total selection of ewe lamb mass along different causal paths to fitness. Larger lambs are more likely to become pregnant, which has a large viability cost. After controlling for this pathway, however, the association between lamb mass and subsequent lifetime fitness is positive. Thus this trade-off does not fully explain stasis of size in tis population, but it does substantially reduce the strength of positive directional selection of size that would otherwise occur. While selection currently favours reduced probability of early pregnancy, largely irrespective of body size, it is likely that the occurrence of early pregnancy could result from adaptation to conditions during a recent period during which population density was much lower.

Mating is rarely random in nature, but the effects of mate choice on offspring performance are still poorly understood. We sampled in total 47 wild lake char (Salvelinus umbla) during two breeding seasons and used their gametes to investigate the genetic consequences of different mating scenarios. In a first study, 1,464 embryos that resulted from sperm competition trials were raised singly in either a stress- or non-stress environment. Offspring growth turned out to be strongly reduced with increased genetic relatedness between the parents while male coloration (that reveal aspects of male health) was no significant predictor of offspring performance. In a second experiment one year later, block-wise full-factorial in vitro breeding was used to produce 3,094 embryos that were raised singly after sublethal exposures to a pathogen or water only. Offspring growth was again strongly reduced with increased genetic relatedness between the parents while male coloration was no significant predictor of offspring performance. We conclude that the genetic benefits of mate choice would be strongest if females avoided genetic similarity, while male breeding colors seem more relevant in intra-sexual selection. Impact SummaryMales and females usually compete for access to mating partners, and they usually choose their mates. Sexual selection is therefore a major force in evolution. It shapes sexual signals and mate preferences depending on the type of mating system. A comparatively simple mating system is when fertilization is external and neither males nor females care for their brood, as is the case in salmonid fish. A group of hypotheses then predicts that female mate preferences have evolved to maximize offspring growth and survival through indirect genetic benefits. There are two types of such indirect benefits. In models of good genes sexual selection, conspicuous signals reveal a males health and vigor because only males in good health can afford these costly traits. Females would then prefer males with strong signals. In compatible genes models, females would instead focus on signals that allow them to complement their own genotype to achieve high offspring viability. An example of the latter is inbreeding avoidance through odors that reveal kinship. We sampled wild lake char to compare the likely consequences of these two types of possible female preferences for offspring growth and survival. We experimentally crossed these fish in vitro and raised large numbers of offspring singly and for several months. Our first experiment revealed that offspring growth would be significantly increased if females would avoid mating with genetically more similar males, while preferring males with strong sexual ornaments (in this case: yellow skin colors) would not improve offspring performance. These results could be confirmed in a second experiment with a larger sample size. We conclude that the genetic benefit of mate choice is largest if females aim for compatible genes rather than focusing on the breeding colors that males display. These breeding colors are therefore likely to play a more important role in other contexts, e.g., in male-male competition.

Facultative sex, the ability to reproduce both sexually and asexually, is widespread across the tree of life. In anisogamous species, the frequency of sex modulates selection on traits with sex-specific expression. Current theory on conditional gene expression posits that the strength of selection on loci only expressed by a subset of individuals, and/or in a subset of environments, is proportional to the frequency of expressers in the population. We show here that this assumption does not hold when the subsets in question are males or females (because of the Fisher condition) and is most important in facultatively sexual populations. In this case, the proportion of sexually produced offspring is not determined by male frequency (sex ratio), but on relative female investment in sexual versus asexual reproduction. This breaks the link between the frequency of expressers and selection for loci with sex-specific expression. However, certain conditions can re-establish this link, for example male traits that predict mating success better under strong male-male competition, and sex ratio affecting the relative fecundity of sexual versus asexual females. Our work highlights the importance of the Fisher condition for understanding the efficiency of selection and has implications for differences in the genetic load among sexes.

Maternal inheritance allows selection to act on mtDNA-encoded effects in females but prevents direct selection on mtDNA in males. Mutations that are deleterious in males but neutral or beneficial in females can persist in populations. This predicts that mtDNA-based phenotypic variation should be more common among males than among females, a pattern referred to as Mothers Curse (MC). Most studies of MC place alternative mtDNAs on common homozygous nuclear chromosomal backgrounds, a condition not common in nature. Moreover, it is not known whether MC effects accumulate as mtDNAs acquire nucleotide substitutions between populations or species. We tested the MC hypothesis using mtDNAs from Drosophila melanogaster (OreR, Zimbabwe or w1118), D. simulans (siI and siII) and D. yakuba each placed on several D. melanogaster nuclear backgrounds heterozygous for different chromosomal deficiencies paired with a common w1118 chromosome set. Females and males were tested for starvation resistance, climbing speed, and flight performance. In the majority of chromosomal backgrounds the variance among mtDNA genotypes was greater in females than in males, opposite from the central prediction of Mothers Curse. This suggests that additive and dominance variation across the nuclear genome may provide nuclear blessings that can counter the curse of maternally inherited mtDNA. Teaser textMothers Curse (MC) posits that selection on mtDNA should be stronger in females than in males due to maternal inheritance of mtDNA. This predicts that phenotypic variation among mtDNA genotypes should be lower for females and higher for males. There is conflicting experimental evidence for MC. Most studies of MC have used a common, homozygous nuclear background and have not explored the influence of divergent mtDNAs as strong predictors of MC effects. We address both issues by assaying performance traits among mtDNAs of varying levels of divergence on heterozygous backgrounds. The data fail to support the MC hypothesis and even reveal the opposite effect that females have greater phenotypic variation across mtDNAs. MC may operate in some contexts, but it is not a consistent force in evolutionary genetics.

Experiments have shown that when one plant is attacked by a pathogen or herbivore, this can lead to other plants connected to the same mycorrhizal network upregulating their defence mechanisms. It has been hypothesised that this represents signalling, with attacked plants producing a signal to warn other plants of impending harm. We examined the evolutionary plausibility of this and other hypotheses theoretically. We found that the evolution of plant signalling about an attack requires restrictive conditions, and so will rarely be evolutionarily stable. The problem is that signalling about an attack provides a benefit to competing neighbours, even if they are kin, and so reduces the relative fitness of signalling plants. Indeed, selection is often more likely to push plant behaviour in the opposite direction - with plants signalling dishonestly about an attack that has not occurred, or by suppressing a cue that they have been attacked. Instead, we show that there are two viable alternatives that could explain the empirical data: (1) the process of being attacked leads to a cue (information about the attack) which is too costly for the attacked plant to fully suppress; (2) mycorrhizal fungi monitor their host plants, detect when they are attacked, and then the fungi signal this information to warn other plants in their network. Our results suggest the empirical work that would be required to distinguish between these possibilities. Significance statementExperiments have shown that when one plant is attacked by a herbivore, this can lead to other plants connected to the same mycorrhizal network upregulating their defence mechanisms. It has been hypothesised that this represents signalling, with attacked plants producing a signal to warn other plants of impending harm. We found theoretically that plant warning signals are rarely evolutionarily stable. Instead, we identify two viable alternatives that could explain the empirical data: (1) being attacked leads to a cue (information about the attack) which is too costly for the attacked plant to suppress; (2) mycorrhizal fungi monitor their host plants, detect when they are attacked, and then the fungi signal this information to warn other plants in their network.

Understanding the evolutionary and genetic underpinnings of susceptibility to pathogens is of fundamental importance across a wide swathe of biology. Much theoretical and empirical effort has focused on genetic variants of large effect, but pathogen susceptibility often appears to be a polygenic complex trait. Here we investigate the quantitative genetics of survival over 120 hours of exposure ("susceptibility") of C. elegans to three bacterial pathogens of varying virulence, along with the standard laboratory food, the OP50 strain of E. coli. We compare the genetic (co)variance input by spontaneous mutations accumulated under minimal selection to the standing genetic (co)variance in a set of 47 wild isolates. Three conclusions emerge. First, mutations increase susceptibility to pathogens, and susceptibility is uncorrelated with fitness in the absence of pathogens. Second, the orientation in trait space of the heritable (co)variance of wild isolates is sufficiently explained by mutation. However, with the possible exception of S. aureus, pathogen susceptibility is clearly under purifying, directional, selection of magnitude roughly similar to that of competitive fitness in the MA conditions. The results provide no evidence for fitness tradeoffs between pathogen susceptibility and fitness in the absence of pathogens.

Herbivory shapes plant trait evolution by altering allocation to growth and defense in ways that affect plant reproduction and fitness. Initiation of these trade-offs may be particularly strong in juvenile plants with high phenotypic plasticity. Herbivory costs are often measured in terms of plant size or flower numbers, but other herbivore-induced floral changes can alter interactions with pollinators and have important implications for mating systems. In mixed-mating plants that can both self-fertilize and outcross, herbivory can maintain mating system variation if herbivore damage and defensive induction change a plants likelihood of selfing versus outcrossing. Here, I use mixed-mating Datura stramonium to evaluate how early defensive induction and herbivory result in trade-offs among plant defense, growth and reproduction. I used a 2x2 factorial manipulation of early chemical defense induction and season-long insecticide in the field. Growth costs of chemical induction were seen even before plants received damage, indicating an inherent cost of defense. Induction and herbivory changed multiple aspects of floral biology associated with a plants selfing or outcrossing rate. This including reduced floral allocation, earlier flowering, and reduced anther-stigma separation (herkogamy). Although these floral changes are associated with decreased attractiveness to pollinators, plants exposed to natural herbivory did not have decreased seed set. This is likely because their floral morphologies became more conducive to selfing (via reduced herkogamy). These vegetative and floral changes following damage and defensive induction can impact interactions among plants (by altering mating environment) and interactions with pollinators (via changes in floral allocation and floral phenology).

The response of natural populations to selection and the role of genetic correlations in constraining or facilitating evolutionary change is fundamental to adaptation. We use artificial selection to investigate the evolutionary response of clonal reproduction in the common monkeyflower (Mimulus guttatus), a species with extensive life history variation. We first characterize the standing genetic variation in a single perennial population, and then conduct four generations of divergent artificial selection on stolon number--the mechanism of clonal reproduction in this species. To start, stolon number had moderate heritability (H{superscript 2}=0.25) and was negatively genetically correlated with reproductive traits. Artificial selection produced a clear but asymmetrical response. High selection lines made significantly more stolons, while low lines diverged less from controls. Analyses of G matrices revealed that selection not only changed trait means but also genetic correlations, with high lines diverging more in multivariate genetic architecture. Our results demonstrate that single populations harbor sufficient genetic variation to respond rapidly to selection on clonality, and the response is shaped by existing patterns of genetic covariation. The capacity for rapid evolution of clonal traits is particularly relevant as climate change alters selection and shifts the relative advantages of sexual versus clonal life-history strategies.

A populations ability to adapt is determined by its levels of additive genetic variance (VA), and while it is agreed that most organisms have genetic variation for most traits, the extent to which it varies between species is poorly characterised. Here we investigate this question by compiling 3209 and 1852 estimates of heritability and evolvability (the additive genetic variance divided by the square of the mean) estimates respectively, for a variety of traits, from 220 and 172 multicellular eukaryotic species. Using phylogenetic generalised linear mixed models, we find substantial and highly significant interspecific variation in evolvability. Much of the variation is explained by phylogenetic relatedness, with plants in our data having substantially higher evolvability than animals. While heritability also varies between species, the differences are more subtle, and plants are not exceptional. We investigate whether the variation in evolvability and heritability between species is due to variation in the mutation rate, effective population size, genome size, ploidy and recombination rate, but find little evidence of any factor being important. However, the confidence intervals are large suggesting that we have little power to detect any associations between these factors and our estimates of VA.

Closely related populations often differ in resistance to a given parasite, as measured by infection success or failure. Yet, the immunological mechanisms of these evolved differences are rarely specified. Does resistance evolve via changes to the hosts ability to recognize that an infection exists, actuate an effective immune response, or attenuate that response? We tested whether each of these phases of the host response contributed to threespine sticklebacks recently evolved resistance to their tapeworm Schistocephalus solidus. While marine stickleback and some susceptible lake fish permit fast-growing tapeworms, other lake populations are resistant and suppress tapeworm growth via a fibrosis response. We subjected lab-raised fish from three populations (susceptible marine ancestors, a susceptible lake, a resistant lake), to a novel immune challenge (injection of: 1) a saline control, 2) alum, a generalized pro-inflammatory adjuvant that causes fibrosis, 3) a tapeworm protein extract, and 4) a combination of alum and tapeworm protein). All three populations were capable of a robust fibrosis response to the alum treatments (but not the saline control). Yet, only the resistant population exhibited a fibrosis response to the tapeworm protein alone. Thus, these populations differed in their ability to recognize the tapeworm but shared an intact fibrosis pathway. However, the resistant population also initiated fibrosis faster, and was able to attenuate fibrosis, unlike the susceptible populations slow but longer-lasting response to alum. As fibrosis has presumed pathological side-effects, this difference may reflect adaptions to mitigate costs of immunity in the resistant population. Broadly, our results confirm that parasite detection, activation speed, and immune attenuation simultaneously contribute to adaptations to parasite infection in natural populations. IMPACT SUMMARYDramatic variation in parasite resistance is common in nature, even to the same parasite, yet we are still working to understand the mechanisms of how such differences evolve. Many evolution studies focus on the broad outcomes of infection (infected or not) when studying this question, without specifying what part of the immune response has evolved. Here, we experimentally partition different sequential stages in the host immune response (recognition, actuation, attenuation), to evaluate which stage(s) underly the evolution of host resistance to infection. This study compares three populations of threespine stickleback that naturally differ in their exposure and their ability to resist infections of a freshwater tapeworm. These include a "resistant" lake population, a "susceptible" lake population, and an ancestral marine population that is rarely exposed to the tapeworm in nature, but is susceptible when exposed in the lab. The resistant population exhibits a fibrosis immune response to infection, which has previously been linked to suppressed tapeworm growth and viability. We injected different immune challenges directly into the site of infection (peritoneal cavity) and measured the subsequent fibrosis response through time. We found that all populations were capable of producing fibrosis in response to a general immune stimulant (alum). But, only the resistant population was able to recognize and respond to tapeworm protein alone. This population also responded faster than the others, within 24 hours, and attenuated its fibrosis by 90 days post-injections whereas the other populations exhibited a slower response that did not attenuate in the study time-frame. We concluded that variation in parasite recognition, an early phase in the host response, shapes the evolution of the initiation and resolution of the physical response to infection. Broadly, our results support that parasite detection mechanisms could play a key role in the rapid evolution of parasite resistance.

Understanding evolutionary repeatability is a central question in biology, as it informs how predictably organisms respond to similar selection pressures. However, the extent to which phenotypic repeatability is recapitulated at the genetic level remains unclear, particularly for quantitative traits. The recurrent evolution of similar phenotypes in high altitude plant populations relative to their low altitude counterparts offers an ideal model for testing genetic repeatability, as these habitats are associated with shifts in complex suites of phenotypes. Here, we investigate the modularity and genetic architecture of life-history trait divergence across four independent transitions to high altitude habitats among closely related perennial taxa in the Mimulus guttatus species complex. High altitude taxa exhibit largely repeated phenotypic evolution in 40 univariate traits and suites of traits form correlated modules that are highly similar across taxa. Nonetheless, the genetic architecture underlying each trait was largely non-repeatable, a pattern consistent for both quantitative and genetically simple traits. Despite a general lack of overall repeatability, individual QTLs with larger effects and those that were associated with multiple traits were more likely to be repeatable than smaller-effect or single-trait associated loci. These findings suggest that evolution may follow distinct genetic pathways while repeatedly converging on functionally integrated trait modules. Additionally, although there may be several genetic routes to the same phenotypic outcomes, aspects of genetic architecture can influence the most likely genetic routes taken. Overall, our results provide insights into adaptation to high altitude environments and also advance our understanding of evolutionary repeatability of complex traits.

Microevolution is well documented in natural populations, yet its persistence as an adaptive process remains debated. Despite widespread directional selection on heritable traits, including life-history traits, evolutionary stasis often prevails, with no detectable microevolutionary response. However, evidence of microevolution in some populations raises a key question: do populations under similar ecological conditions and selective pressures exhibit parallel evolution in the same traits? To address this, we examined age at first reproduction (AFR) in three contemporary human populations considered semi-independent replicates, sharing a genetic, demographic, and historical background, with pedigree data available for [~]7 generations. Across all populations, we found strong directional selection favoring earlier AFR, yet quantitative genetic analyses revealed consistently low heritability (h2 {approx} 0.11). Only in the Charlevoix population did AFR show a negative genetic correlation with relative fitness, where more than 50% of the standardized phenotypic selection gradient was explained by the genetic selection gradient. Using the Breeders Equation and Robertsons Secondary Theorem of Selection, we predicted an evolutionary response to selection for AFR, which emerged only in Charlevoix. However, neither phenotypic nor breeding values of AFR showed temporal trends, indicating evolutionary stasis. These findings demonstrate that even under consistent directional selection and moderate additive genetic variation, microevolutionary responses may vary across replicate populations. Our results underscore the prevalence of evolutionary stasis, challenging assumptions about the inevitability of microevolution in response to natural selection.

Parasite defense is the ability of a host to minimize fitness loss to parasites, and it is among the most variable phenotypes in natural populations. We expect this variation in defense to facilitate rapid adaptation under parasite-mediated selection. What we do not know is what traits are most likely to evolve in response to this selection. A common assumption is that the most defended hosts are the most resistant, meaning they limit the establishment and growth of infecting parasites. Under this assumption, resistance traits should evolve readily under parasite selection. Resistance is, however, just one of many strategies hosts use to defend against parasites, and it does not consistently covary with parasite defense. We accordingly ask: which host traits covary with parasite defense and are thus likely to respond to parasite selection? We use controlled exposures to characterize genetic variation in defense of the nematode Caenorhabditis elegans against its natural microsporidian parasites. We report extensive variation in parasite defense among wild strains of C. elegans: some strains lost 60% of fecundity under parasite exposure, while others were unaffected. We then tested the covariance of defense with two prominent host traits, resistance and reproductive timing. Our results did not support the hypothesis that resistance covaries with defense: strains with lower parasite burden did not have higher relative fecundity under exposure. Our results instead supported the hypothesis that life history covaries with defense: host strains that reproduced quickly had higher relative fecundity under exposure, consistent with the idea that parasites diminish future reproductive opportunities. Moreover, we detected substantial heritability of fecundity traits but low heritability of resistance traits. Together, these findings indicate significant potential for adaptation of wild C. elegans populations to defend against their natural parasites. They further predict that life history traits will evolve rapidly in response to parasite selection. AUTHOR SUMMARYSome hosts fare much better than others in the face of parasite infection. What traits differentiate defended hosts from undefended hosts? The answer to this question is critical for identifying the strategies that best protect hosts from their parasites. It also allows us to predict and interpret the evolution of host populations over the course of epidemics. To address this question, we surveyed wild strains of a tractable model host, the nematode Caenorhabditis elegans, for their response to two species of microsporidian parasites. We found that, on average, parasite exposure substantially impaired the ability of hosts to reproduce. Host strains, however, varied widely: some experienced major losses in fecundity with exposure, while others were highly defended, showing little to no change. We identified reproductive timing as the trait that differentiated defended hosts from undefended hosts. Our results indicate that reproducing quickly was protective, because hosts were able to make most of their offspring before parasites impaired reproduction. We did not find evidence that resistance was protective - hosts with lower parasite burdens did not reproduce better than those with high parasite burdens. These findings give added weight to life history as a major component of host defense against parasites.