Inter- and intrapopulation variation in thermal reaction norms for growth rate: evolution of latitudinal compensation in ectotherms with a genetic constraint

In ectotherms, lower temperatures in high-latitude environments would theoretically reduce the annual growth rates of individuals. If slower growth and resultant smaller body size reduce fitness, individuals in higher latitudes may evolve compensatory responses. Two alternative models of such latitudinal compensation are possible: Model I: thermal reaction norms for growth rates of high-latitude individuals may be horizontally shifted to a lower range of temperatures, or Model II: reaction norms may be vertically shifted so that high-latitude individuals can grow faster across all temperatures. Model I is expected when annual growth rates in the wild are only a function of environmental temperatures, whereas Model II is expected when individuals in higher latitudes can only grow during a shorter period of a year. A variety of mixed strategies of these two models are also possible, and the magnitude of horizontal versus vertical variation in reaction norms among latitudinal populations will be indicative of the importance of "temperature" versus "seasonality" in the evolution of latitudinal compensation. However, the form of latitudinal compensation may be affected by possible genetic constraints due to the genetic architecture of reaction norms. In this study, we examine the inter- and intrapopulation variations in thermal reaction norms for growth rate of the medaka fish Oryzias latipes. Common-environment experiments revealed that average reaction norms differed primarily in elevation among latitudinal populations in a manner consistent with Model II (adaptation to "seasonality"), suggesting that natural selection in high latitudes prefers individuals that grow faster even within a shorter growing season to individuals that have longer growing seasons by growing at lower temperatures. However, intrapopulation variation in reaction norms was also vertical: some full-sibling families grew faster than others across all temperatures examined. This tendency in intrapopulation genetic variation for thermal reaction norms may have restricted the evolution of latitudinal compensation, irrespective of the underlying selection pressure.     

An empirical test for a zone of canalization in thermal reaction norms

Theoretical models on the evolution of phenotypic plasticity predict a zone of canalization where reaction norms cross, and genetic variation is minimized in the environment a population most frequently encounter. Empirical tests of this prediction are largely missing, in particular for life‐history traits. We addressed this prediction by quantifying thermal reaction norms of three life‐history traits (somatic growth rate, age and size at maturation) of a Norwegian population of Daphnia magna and testing for the occurrence of an intermediate temperature (T_m) at which genetic variance in the traits is minimized. Size at maturation changed relatively little with temperature compared to the other traits, and there was no genetic variance in the shape of the reaction norm. Consequently, age at maturation and somatic growth rate were strongly negatively correlated. Both traits showed a strong genotype–environment interaction, and the estimated T_m was 14 °C for both age at maturation and growth rate. This value of T_m corresponds well with mean summer temperatures experienced by the population and suggests that the population has evolved under stabilizing selection in temperatures that fluctuate around this mean temperature. These results suggest local adaptation to temperature in the studied population and allow predicting evolutionary trajectories of thermal reaction norms under changing thermal regimes.     

The influence of diet on the duration and thermal sensitivity of development in the linden bug Pyrrhocoris apterus L. (Heteroptera: Pyrrhocoridae)

Temperature and nutrition are crucial environmental variables that determine rates of growth and development in insects. However, the simultaneous effect of these factors on life‐history traits is rarely addressed. In the present study, the influence of two diets (linden fruit and sunflower seeds) on the duration of immature stages and thermal reaction norms for development is tested in the bug Pyrrhocoris apterus L. (Heteroptera: Pyrrhocoridae). Eggs and larvae are reared at five constant temperatures (20, 22, 24, 26 and 28 °C) under an LD 20 : 4 h photocycle. Development rates deviate from linearity in the studied thermal range, especially in larvae; therefore, a nonlinear (power‐law) approximation is also attempted. Parental diet causes no change in thermal reaction norms for egg development. However, the progeny of sunflower‐fed bugs are more variable in terms of their development time, suggesting a transgenerational effect. Larval mortality rates increase in cooler conditions and are always higher on sunflower seeds. This is accompanied by more variable, less temperature‐dependent and generally slower larval development. A review of previously published case studies on temperature–diet interactions in the control of insect development leads to two general conclusions. First, there are two approaches for assessing the temperature‐dependent development in insects: one based on the concept of the sum of degree‐days and the other based on the concept of reaction norm. Despite an obvious non‐exclusiveness, the two approaches appear to have developed in isolation from each other. Second, three principal patterns of temperature–diet interactions can be recognized. The pattern found in P. apterus (the direct effects of diet are stronger at higher temperatures and much weaker or absent at lower temperatures) appears to be the most widespread.     

Latitudinal and temperature-dependent variation in embryonic development and growth in Rana temporaria

Variation in seasonal time constraints and temperature along latitudinal gradients are expected to select for life history trait differentiation, but information about the relative importance of these factors in shaping patterns of divergence in embryonic traits remains sparse. We studied embryonic survival, growth and development rates in the common frog (Rana temporaria) along a 1,400-km latitudinal gradient across Sweden by raising embryos from four populations in the laboratory at seven temperatures (9 degrees C, 12 degrees C, 15 degrees C, 18 degrees C, 21 degrees C, 24 degrees C, 27 degrees C). We found significant differences in mean values of all traits between the populations and temperature treatments, but this variation was not latitudinally ordered. In general, embryonic survival decreased at the two highest temperatures in all populations, but less so in the southernmost as compared to the other populations. The northernmost population developed slowest at the lowest temperature, while the two mid-latitude populations were slowest at the other temperatures. Hatchling size increased with increasing temperature especially in the two northern populations, whereas the two southern populations showed peak hatchling size at 15 degrees C. Analyses of within-population genetic variation with a half-sib design revealed that there was significant additive genetic variation in all traits, and egg size-related maternal effects were important in the case of hatchling size. Overall, our results indicate that unlike larval growth and development, variation in embryonic development and growth in R. temporaria cannot be explained in terms of a latitudinal gradient in season length. While adaptation to a latitudinal variation in temperature might have contributed to the observed differentiation in embryonic performance, the effects of other, perhaps more local environmental factors, seem to have overridden them in importance.     

Size‐induced phenotypic reaction norms in a parasitoid wasp: an examination of life‐history and behavioural traits

The amount of resources available during development often affects body size, causing phenotypic variation in life‐history traits and reproductive behaviours. However, past studies have seldom examined the reaction norms of both life‐history and behavioural traits versus body size. We measured the phenotypic plasticity of several life‐history (age‐specific egg load, egg size, longevity) and behavioural (oviposition rate, host marking rate, walking speed) traits of the egg parasitoid Telenomus podisi Ashmead (Hymenoptera: Scelionidae) in response to body size variation. We predicted that life‐history traits would show more evidence of size compensation than behavioural traits, resulting in fewer positively‐sloped size versus trait reaction norms among the former. As predicted by life‐history models, smaller wasps appear to shift resource allocation towards early‐life reproduction, having a similar egg load to large individuals 9 days after emergence. Surprisingly, longevity was unaffected by body size. However, egg size, the number of offspring produced during oviposition bouts, and the rate of subsequent egg synthesis were greater for larger individuals. In addition, as predicted, the reaction norms of behavioural traits versus body size were all positively sloped. Thus, despite possible adaptive compensatory plasticity of life‐history traits by small individuals, behavioural constraints directly related to body size would contribute to maintaining a positive size–fitness relationship.     

Pathways to fitness: carry‐over effects of late hatching and urbanisation on lifetime mating success

Life history theory and most empirical studies assume carry‐over effects of larval ­conditions to shape adult fitness through their impact on metamorphic traits (age and mass at metamorphosis). Yet, very few formal tests of this connection across metamorphosis exist, because this entails longitudinal studies from the egg stage and requires measuring fitness in (semi)natural conditions. In a longitudinal one‐year common‐garden rearing experiment consisting of an outdoor microcosm part for the larval stage and a large outdoor insectary part for the adult stage, we studied the effects of two factors related to time constraints in the larval stage (egg hatching period and urbanisation) on life history traits and lifetime mating success in the males of the damselfly Coenagrion puella. We reared early‐ and late‐hatched larvae from each of three rural and three urban populations from the egg stage throughout their adult life. Key findings were that both the hatching period and urbanisation shaped adult fitness, yet through different pathways. As expected, the more time‐constrained late‐hatched individuals accelerated their larval life history and this was associated with a lower lifetime mating success. A path analysis revealed this carry‐over effect was mediated by the changes in the two metamorphic traits (reduced age and lower mass at emergence). Notably, urban males had a 50% lower lifetime mating success, which was not mediated by age and mass at emergence, and possibly driven by their shorter lifespan. Our results point to long‐term carry‐over effects of the usually ignored natural variation in egg hatching dates, and further contribute to the limited evidence showing fitness costs of adjusting to an urban lifestyle.     

Latitudinal and voltinism compensation shape thermal reaction norms for growth rate

Latitudinal variation in thermal reaction norms of key fitness traits may inform about the response of populations to climate warming, yet their adaptive nature and evolutionary potential are poorly known. We assessed the contribution of quantitative genetic, neutral genetic and environmental effects to thermal reaction norms of growth rate for populations of the damselfly Ischnura elegans. Among populations, reaction norms differed primarily in elevation, suggesting that time constraints associated with shorter growth seasons in univoltine, high-latitude as well as multivoltine, low-latitude populations selected for faster growth rates. Phenotypic divergence among populations is consistent with selection rather than drift as Q(ST) was greater than F(ST) in all cases. Q(ST) estimates increased with experimental temperature and were influenced by genotype by environment interactions. Substantial additive genetic variation for growth rate in all populations suggests that evolution of trait means in different environments is not constrained. Heritability of growth rates was higher at high temperature, driven by increased genetic rather than environmental variance. While environment-specific nonadditive effects also may contribute to heritability differences among temperatures, maternal effects did not play a significant role (where these could be accounted for). Genotype by environment interactions strongly influenced the adaptive potential of populations, and our results suggest the potential for microevolution of thermal reaction norms in each of the studied populations. In summary, the observed latitudinal pattern in growth rates is adaptive and results from a combination of latitudinal and voltinism compensation. Combined with the evolutionary potential of thermal reaction norms, this may affect populations' ability to respond to future climate warming.     

Stable and variable life‐history responses to temperature and photoperiod in the beet webworm,Loxostege sticticalis

The beet webworm, Loxostege sticticalis L. (Lepidoptera: Crambidae), unlike many temperate insects and despite its wide distribution range, has a geographically stable value of the critical photoperiod for diapause induction. It has thus been hypothesized that the species could adjust its life cycle to different climates in an alternative way, which should be reflected in geographical variation and/or environmental plasticity of some other ecophysiological trait. Three remote populations of the beet webworm were studied. The insects were reared from egg to adult at several combinations of temperature and photoperiod in order to measure development times of all the immature stages and pre‐pupal body weight, and to characterize the sensitivity of these life‐history traits to the two ecological factors. The thermal reaction norms for immature development appeared to be significantly different in the three populations. There was also a significant effect of photoperiod on development time as well as on the thermal sensitivity and lower temperature threshold for larval development. Pre‐pupae from the northernmost population were heavier and their body mass was more strongly affected by photoperiod than in the other two, but attainment of a greater weight under short‐day conditions, especially combined with higher temperatures, was common for all the three populations. Nevertheless, all the discovered geographical and environmentally induced differences in life‐history traits were very small and their adaptive significance remains problematic.     

Integrating large‐scale geographic patterns in flight morphology,flight characteristics and sexual selection in a range‐expanding damselfly

While geographic trait variation along environmental clines is widespread, associated patterns in sexual selection remain largely unexplored. Geographic patterns in sexual selection may be expected if 1) phenotypes vary geographically and sexual selection is dependent on the local phenotypes in the population, and if 2) sexual selection is influenced by geographically structured environmental conditions. We quantified geographic variation in flight‐related traits and flight performance in mated and unmated males and tested for geographic variation in sexual selection on these traits in the poleward range‐expanding damselfly Coenagrion scitulum across a set of eleven core and edge populations ordered along thermal gradients in the larval and in the adult stage. We found little support for trait differentiation between core and edge populations, instead we found considerable geographic trait variation along the larval and adult thermal gradients. As expected under time constraints, body mass decreased with shorter larval growth seasons. Lower temperatures during the adult flight period were associated with a higher body mass, a higher flight speed and a higher fat content; these traits likely evolved to buffer flight ability at suboptimal temperatures and to optimize starvation resistance. Across the large geographic scale, we found a consistent higher flight duration in mated males. Instead, sexual selection for higher fat content was stronger in populations with lower adult flight temperatures and sexual selection for lower body mass acted only in edge populations. Our results indicate sexual selection on flight performance to be consistent over a large geographic scale and this despite the clear geographic patterns in sexual selection on the underlying morphological traits. Our results highlight that to fully understand the fitness implications of geographically changing trait patterns, researchers should consider the entire phenotype–performance–fitness axis and incorporate effects of geographically structured life‐stage specific environmental conditions on this axis.     

A large‐scale latitudinal pattern of life‐history traits in a strictly univoltine damselfly

1. Variation in thermal conditions and season length along latitudinal gradients affect body size‐related traits over different life stages. Selection is expected to optimise these size traits in response to the costs and benefits. 2. Egg, hatchling, larval and adult size in males and females were estimated along a latitudinal gradient of 2730 km across Europe in the univoltine damselfly Lestes sponsa, using a combination of field‐collection and laboratory‐rearing experiments. In the laboratory, individuals were grown in temperatures and photoperiod simulating those at the latitude of origin, and in common‐garden conditions. 3. The size of adults sampled in nature was negatively correlated with latitude. In all populations the females were larger than the males. Results from simulated and common‐garden rearing experiments supported this pattern of size difference across latitudes and between sexes, suggesting a genetic component for the latitudinal size trend and female‐biased size dimorphism. In contrast, hatchling size showed a positive relationship with latitude, but egg size, although differing between latitudes, showed no such relationship. 4. The results support a converse Bergmann cline, i.e. a negative body size cline towards the north. This negative cline in body size is probably driven by progressively stronger seasonal time and temperature constraints towards the higher latitudes and by the obligate univoltine life cycle of L. sponsa. As egg size showed no relationship with latitude, other environmental factors besides temperature, such as desiccation risk, probably affect this trait.     

Experimental evolution for generalists and specialists reveals multivariate genetic constraints on thermal reaction norms

Theory predicts the emergence of generalists in variable environments and antagonistic pleiotropy to favour specialists in constant environments, but empirical data seldom support such generalist–specialist trade‐offs. We selected for generalists and specialists in the dung fly Sepsis punctum (Diptera: Sepsidae) under conditions that we predicted would reveal antagonistic pleiotropy and multivariate trade‐offs underlying thermal reaction norms for juvenile development. We performed replicated laboratory evolution using four treatments: adaptation at a hot (31 °C) or a cold (15 °C) temperature, or under regimes fluctuating between these temperatures, either within or between generations. After 20 generations, we assessed parental effects and genetic responses of thermal reaction norms for three correlated life‐history traits: size at maturity, juvenile growth rate and juvenile survival. We find evidence for antagonistic pleiotropy for performance at hot and cold temperatures, and a temperature‐mediated trade‐off between juvenile survival and size at maturity, suggesting that trade‐offs associated with environmental tolerance can arise via intensified evolutionary compromises between genetically correlated traits. However, despite this antagonistic pleiotropy, we found no support for the evolution of increased thermal tolerance breadth at the expense of reduced maximal performance, suggesting low genetic variance in the generalist–specialist dimension.     

Thermal plasticity in postembryonic life history traits of a widely distributed Collembola: Effects of macroclimate and microhabitat on genotypic differences

Life history traits in many ectotherms show complex patterns of variation among conspecific populations sampled along wide latitudinal or climatic gradients. However, few studies have assessed whether these patterns can be explained better by thermal reaction norms of multiple life history traits, covering major aspects of the life cycle. In this study, we compared five populations of a Holarctic, numerically dominant soil microarthropod species, Folsomia quadrioculata, sampled from a wide latitudinal gradient (56–81°N), for growth, development, fecundity, and survival across four temperatures (10, 15, 20, and 25°C) in common garden experiments. We evaluated the extent to which macroclimate could explain differences in thermal adaptation and life history strategies among populations. The common garden experiments revealed large genotypic differences among populations in all the traits, which were little explained by latitude and macroclimate. In addition, the life history strategies (traits combined) hardly revealed any systematic difference related to latitude and macroclimate. The overall performance of the northernmost population from the most stochastic microclimate and the southernmost population, which remains active throughout the year, was least sensitive to the temperature treatments. In contrast, performance of the population from the most predictable microclimate peaked within a narrow temperature range (around 15°C). Our findings revealed limited support for macroclimate‐based predictions, and indicated that local soil habitat conditions related to predictability and seasonality might have considerable influence on the evolution of life history strategies of F. quadrioculata. This study highlights the need to combine knowledge on microhabitat characteristics, and demography, with findings from common garden experiments, for identifying the key drivers of life history evolution across large spatial scales, and wide climate gradients. We believe that similar approaches may substantially improve the understanding of adaptation in many terrestrial ectotherms with low dispersal ability.     

Rapid evolution of larval life history,adult immune function and flight muscles in a poleward‐moving damselfly

Although a growing number of studies have documented the evolution of adult dispersal‐related traits at the range edge of poleward‐expanding species, we know little about evolutionary changes in immune function or traits expressed by nondispersing larvae. We investigated differentiation in larval (growth and development) and adult traits (immune function and flight‐related traits) between replicated core and edge populations of the poleward‐moving damselfly Coenagrion scitulum. These traits were measured on individuals reared in a common garden experiment at two different food levels, as allocation trade‐offs may be easier to detect under energy shortage. Edge individuals had a faster larval life history (growth and development rates), a higher adult immune function and a nearly significant higher relative flight muscle mass. Most of the differentiation between core and edge populations remained and edge populations had a higher relative flight muscle mass when corrected for latitude‐specific thermal regimes, and hence could likely be attributed to the range expansion process per se. We here for the first time document a higher immune function in individuals at the expansion front of a poleward‐expanding species and documented the rarely investigated evolution of faster life histories during range expansion. The rapid multivariate evolution in these ecological relevant traits between edge and core populations is expected to translate into changed ecological interactions and therefore has the potential to generate novel eco‐evolutionary dynamics at the expansion front.     

Local adaptation of a Drosophila parasitoid: habitat‐specific differences in thermal reaction norms

Local climate is an important source of selection on thermal reaction norms that has been well investigated in cline studies, where populations sampled along altitudinal or latitudinal gradients are compared. Several biotic factors vary with climate, but are rarely integrated as alternative agents of selection to climatic factors. We tested the hypothesis that habitat may select for thermal reaction norms and magnitude of phenotypic plasticity in a drosophila parasitoid, independently of the climate of origin. We sampled populations of Leptopilina boulardi, a Drosophila parasitoid in two different habitats, orchards and forests. Orchards offer laying opportunities over small distances for parasitoids, with a low variability in the number of hosts per patch, while forests offer more dispersed and more variable patches. The sampling was realized in a temperate and a Mediterranean climate. We measured egg load, volume of eggs, longevity and lipid content for parasitoids reared at two temperatures. Reaction norms were opposite for populations from forests and orchards for investment in reproduction, independently of the climate of origin. The maximal investment of resources in reproduction occurred at the lower temperature in orchards and the higher temperature in forests. Host distribution differences between habitats may explain these opposite reaction norms. We also observed a flatter reaction norm for egg load in forests than in orchards. This relative canalization may have been selected in response to the higher variability in laying opportunities observed in forests. Our results demonstrate the potential role of resource distribution in evolution of thermal plasticity.     

Geographic variation in body size, sexual size dimorphism and fitness components of a seed beetle: local adaptation versus phenotypic plasticity

Variation in body size, growth and life history traits of ectotherms along latitudinal and altitudinal clines is generally assumed to represent adaptation to local environmental conditions, especially adaptation to temperature. However, the degree to which variation along these clines is due to adaptation vs plasticity remains poorly understood. In addition, geographic patterns often differ between females and males – e.g. sexual dimorphism varies along latitudinal clines, but the extent to which these sex differences are due to genetic differences between sexes vs sex differences in plasticity is poorly understood. We use common garden experiments (beetles reared at 24, 30 and 36°C) to quantify the relative contribution of genetically‐based differentiation among populations vs phenotypic plasticity to variation in body size and other traits among six populations of the seed‐feeding beetle Stator limbatus collected from various altitudes in Arizona, USA. We found that temperature induces substantial plasticity in survivorship, body size and female lifetime fecundity, indicating that developmental temperature significantly affects growth and life history traits of S. limbatus. We also detected genetic differences among populations for body size and fecundity, and genetic differences among populations in thermal reaction norms, but the altitude of origin (and hence mean temperature) does not appear to explain these genetic differences. This and other recent studies suggest that temperature is not the major environmental factor that generates geographic variation in traits of this species. In addition, though there was no overall difference in plasticity of body size between males and females (when averaged across populations), we did find that the degree to which dimorphism changed with temperature varied among populations. Consequently, future studies should be extremely cautious when using only a few study populations to examine environmental effects on sexual dimorphism.     

Comparative study of thermal reaction norms for development in ants

This is the first compilation of our research and data from the literature on the duration and thermal reaction norms for development in ants. Altogether, 97 regression lines characterizing the linear dependence of ant brood (egg, larval, prepupal and pupal) development on temperature were obtained for 33 species from 15 genera and three subfamilies. Significant positive correlations between the durations of different immature stages were revealed. We found differences between thermal reaction norms for development for various immature stages, some taxonomic groups, and southern and northern groups of species. All immature stages appeared to be shorter on average at 25°C in northern ants compared to southern species; this difference is insignificant only for eggs. Highly significant negative correlations were revealed between the temperature threshold for development (TTD) and the sum of degree‐days (SDD) for all immature stages. The latitudinal trends in intraspecific variation of thermal constants appeared to be opposite to those we observed at the interspecific level. At the latter, the coefficient of thermal sensitivity of development (i.e. the coefficient of linear regression of development rate on temperature) and TTD tends to decrease and SDD to increase from the south to the north. In contrast, at the intraspecific level, development became more temperature‐sensitive in northern populations, i.e. characterized by higher slopes of regression lines of development rate on temperature, and higher TTDs. The species of the genus Formica are characterized by the shortest and the most temperature‐sensitive immature development among all the ant species studied.     

Population differences in early life‐history traits in grayling

In this paper we test population differences in early life‐history traits in three grayling Thymallus thymallus populations. The grayling shared ancestors some 80–90 years ago. We performed common‐garden experiments at three temperatures (mimicking population‐specific summer temperatures), and measured survival and growth rates during early development. We found significant additive genetic variance in size (length and yolk‐sac volume) measured at hatching, swim‐up and termination of the experiment, and significantly different reaction norms for growth rate and survival during the period of first feeding. In general, each population did best at the temperature experienced in nature. These differences in early life‐history traits suggest that natural selection has resulted in local adaptation in a time period of 13–18 generations.     

Local divergence of thermal reaction norms among amphibian populations is affected by pond temperature variation

Although temperature variation is known to cause large‐scale adaptive divergence, its potential role as a selective factor over microgeographic scales is less well‐understood. Here, we investigated how variation in breeding pond temperature affects divergence in multiple physiological (thermal performance curve and critical thermal maximum [CT_max]) and life‐history (thermal developmental reaction norms) traits in a network of Rana arvalis populations. The results supported adaptive responses to face two main constraints limiting the evolution of thermal adaptation. First, we found support for the faster–slower model, indicating an adaptive response to compensate for the thermodynamic constraint of low temperatures in colder environments. Second, we found evidence for the generalist–specialist trade‐off with populations from colder and less thermally variable environments exhibiting a specialist phenotype performing at higher rates but over a narrower range of temperatures. By contrast, the local optimal temperature for locomotor performance and CT_max did not match either mean or maximum pond temperatures. These results highlight the complexity of the adaptive multiple‐trait thermal responses in natural populations, and the role of local thermal variation as a selective force driving diversity in life‐history and physiological traits in the presence of gene flow.     

Parental effects on embryonic viability and growth in Arctic charr Salvelinus alpinus at two incubation temperatures

The parental influences on three progeny traits (survival to eyed‐embryo stage, post‐hatching body length and yolk‐sac volume) of Arctic charr Salvelinus alpinus were studied under two thermal conditions (2 and 7° C) using a factorial mating design. The higher temperature resulted in elevated mortality rates and less advanced development at hatching. Survival was mostly attributable to maternal effects at both temperatures, but the variation among families was dependent on egg size only at the low temperature. No additive genetic variation (or pure sire effect) could be observed, whereas the non‐additive genetic effects (parental combination) contributed to offspring viability at 2° C. In contrast, any observable genetic variance in survival was lost at 7° C, most likely due to the increased environmental variance. Irrespective of temperature, dam and sire–dam interaction contributed significantly to the phenotypic variation in both larval length and yolk size. A significant proportion of the variation in larval length was also due to the sire effect at 2° C. Maternal effects were mediated partly through egg size, but as a whole, they decreased in importance at the high temperature, enabling a concomitant increase in non‐additive genetic effects. For larval length, however, the additive component, like maternal effects, decreased at 7° C. The present results suggest that an exposure to thermal stress during incubation can modify the genetic architecture of early developmental traits in S. alpinus and presumably constrain their short‐term adaptive potential and evolvability by increasing the amount of environmentally induced variation.     

Temperature-induced shifts in selective pressure at a critical developmental transition

Selective mortality within a population, based on the phenotype of individuals, is the foundation of the theory of natural selection. We examined temperature-induced shifts in the relationships among early life history traits and survivorship over the embryonic and larval stages of a tropical damselfish, Pomacentrus amboinensis. Our experiments show that temperature determines the intensity of selective mortality, and that this changes with ontogeny. The size of energy stores determined survival through to hatching, after which egg size became a good indicator of fitness as predicted by theoretical models. Yet, the benefits associated with egg size were not uniform among test temperatures. Initial egg size positively influenced larval survival at control temperature (29 °C). However, this embryonic trait had no effect on post-hatching longevity of individuals reared at the higher (31 °C) and lower (25 °C) end of the temperature range. Overall, our findings indicate that the outcome of selective mortality is strongly dependent on the interaction between environment conditions and intrinsic developmental schedules.