Intraspecific mating system evolution and its effect on complex male secondary sexual traits: Does male–male competition increase selection on size or shape?

Sexual selection is generally held responsible for the exceptional diversity in secondary sexual traits in animals. Mating system evolution is therefore expected to profoundly affect the covariation between secondary sexual traits and mating success. Whereas there is such evidence at the interspecific level, data within species remain scarce. We here investigate sexual selection acting on the exaggerated male fore femur and the male wing in the common and widespread dung flies Sepsis punctum and S. neocynipsea (Diptera: Sepsidae). Both species exhibit intraspecific differences in mating systems and variation in sexual size dimorphism (SSD) across continents that correlates with the extent of male–male competition. We predicted that populations subject to increased male–male competition will experience stronger directional selection on the sexually dimorphic male foreleg. Our results suggest that fore femur size, width and shape were indeed positively associated with mating success in populations with male‐biased SSD in both species, which was not evident in conspecific populations with female‐biased SSD. However, this was also the case for wing size and shape, a trait often assumed to be primarily under natural selection. After correcting for selection on overall body size by accounting for allometric scaling, we found little evidence for independent selection on any of these size or shape traits in legs or wings, irrespective of the mating system. Sexual dimorphism and (foreleg) trait exaggeration is therefore unlikely to be driven by direct precopulatory sexual selection, but more so by selection on overall size or possibly selection on allometric scaling.     

QUANTITATIVE GENETIC DIVERGENCE AND STANDING GENETIC (CO)VARIANCE IN THERMAL REACTION NORMS ALONG LATITUDE

Although the potential to adapt to warmer climate is constrained by genetic trade‐offs, our understanding of how selection and mutation shape genetic (co)variances in thermal reaction norms is poor. Using 71 isofemale lines of the fly Sepsis punctum, originating from northern, central, and southern European climates, we tested for divergence in juvenile development rate across latitude at five experimental temperatures. To investigate effects of evolutionary history in different climates on standing genetic variation in reaction norms, we further compared genetic (co)variances between regions. Flies were reared on either high or low food resources to explore the role of energy acquisition in determining genetic trade‐offs between different temperatures. Although the latter had only weak effects on the strength and sign of genetic correlations, genetic architecture differed significantly between climatic regions, implying that evolution of reaction norms proceeds via different trajectories at high latitude versus low latitude in this system. Accordingly, regional genetic architecture was correlated to region‐specific differentiation. Moreover, hot development temperatures were associated with low genetic variance and stronger genetic correlations compared to cooler temperatures. We discuss the evolutionary potential of thermal reaction norms in light of their underlying genetic architectures, evolutionary histories, and the materialization of trade‐offs in natural environments.     

Geographic patterns of postzygotic isolation between two closely related widespread dung fly species (Sepsis cynipsea and Sepsis neocynipsea; Diptera: Sepsidae)

Identifying the contribution of pre‐ and postzygotic barriers to gene flow is a key goal of speciation research. The widespread dung fly species Sepsis cynipsea and Sepsis neocynipsea offer great potential for studying the speciation process over a range of opportunities for gene exchange within and across sister species (cross‐continental allopatry, continental parapatry and sympatry). We examined the role of postcopulatory isolating barriers by comparing female fecundity and egg‐to‐adult viability of F₁ and F₂ hybrids, as well as backcrosses of F₁ hybrids with the parental species, via replicated crosses of sym‐, para‐ and allopatric populations. Egg‐to‐adult viability was strongly but not totally suppressed in hybrids, and offspring production approached nil in the F₂ generation (hybrid breakdown), indicating yet unspecified intrinsic incompatibilities. Viable F₁ hybrid offspring showed almost absolute male (the heterogametic sex) sterility while females remained largely fertile, in accordance with Haldane's rule. Hybridization between the two species in European areas of sympatry (Swiss Alps) indicated only minor reinforcement based on fecundity traits. Crossing geographically isolated European and North American S. neocynipsea showed similar albeit weaker isolating barriers that are most easily explained by random genetic drift. We conclude that in this system with a biogeographic continuum of reproductive barriers, speciation is mediated primarily by genetic drift following dispersal of flies over a wide (allopatric) geographic range, with some role of natural or sexual selection in incidental or direct reinforcement of incompatibility mechanisms in areas of European sympatry. S(ubs)pecies status of continental S. neocynipsea appears warranted.     

Does thermal plasticity align with local adaptation? An interspecific comparison of wing morphology in sepsid flies

Although genetic and plastic responses are sometimes considered as unrelated processes, their phenotypic effects may often align because genetic adaptation is expected to mirror phenotypic plasticity if adaptive, but run counter to it when maladaptive. Because the magnitude and direction of this alignment has further consequences for both the tempo and mode of adaptation, they are relevant for predicting an organisms’ reaction to environmental change. To better understand the interplay between phenotypic plasticity and genetic change in mediating adaptive phenotypic variation to climate variability, we here quantified genetic latitudinal variation and thermal plasticity in wing loading and wing shape in two closely related and widespread sepsid flies. Common garden rearing of 16 geographical populations reared across multiple temperatures revealed that wing loading decreases with latitude in both species. This pattern could be driven by selection for increased dispersal capacity in the cold. However, although allometry, sexual dimorphism, thermal plasticity and latitudinal differentiation in wing shape all show similar patterns in the two species, the relationship between the plastic and genetic responses differed between them. Although latitudinal differentiation (south to north) mirrored thermal plasticity (hot to cold) in Sepsis punctum, there was no relationship in Sepsis fulgens. While this suggests that thermal plasticity may have helped to mediate local adaptation in S. punctum, it also demonstrates that genetic wing shape differentiation and its relation to thermal plasticity may be complex and idiosyncratic, even among ecologically similar and closely related species. Hence, genetic responses can, but do not necessarily, align with phenotypic plasticity induced by changing environmental selection pressures.     

Sexual selection on morphological and physiological traits and fluctuating asymmetry in the yellow dung fly

Previous univariate studies of the yellow dung fly (Scathophaga stercoraria) have demonstrated strong sexual selection, in terms of mating success, on male size (estimated as hind tibia length). To identify specific target(s) of selection on body size and possible conflicting selection pressures on particular body parts, two multivariate field studies of sexual selection were conducted. In one study using point samples from three populations, we assessed several morphological traits, including genital traits and measures of fluctuating asymmetry (FA) of all paired traits. There was sexual selection for large male size in general, confirming previous, univariate studies. With the possible exception of thorax width, which was selected in the opposite direction, no main target of selection was identified, as most morphological traits were highly correlated. There was no detectable sexual selection on the male external genital structures assessed. In a second study using multiple samples from one population, we included physiological measures of energy reserves (lipids, glucose and glycogen) known to affect mating success, in addition to trait size and FA of wings and legs. Inclusion of physiological traits is rare in phenomenological studies of selection. This study again confirmed the mating advantage of large males, and additionally showed independent positive influences of lipid and glucose but not glycogen levels. FA in paired traits generally did not affect male mating success, but was negatively correlated with energy reserves. Our study suggests that inclusion of physiological measures and genital traits in phenomenological studies of selection would be fruitful in other species.     

Genetic and environmental sources of covariance among internal reproductive traits in the yellow dung fly

Substantial inter- and intraspecific variation is found in reproductive traits, but the evolutionary implications of this variation remain unclear. One hypothesis is that natural selection favours female reproductive morphology that allows females to control mating and fertilization and that diverse male reproductive traits arise as counter adaptations to subvert this control. Such co-evolution predicts the establishment of genetic correlations between male and female reproductive traits that closely interact during mating. Therefore, we measured phenotypic and genetic correlations between male and female reproductive tract characteristics in the yellow dung fly, Scathophaga stercoraria (Diptera: Scathophagidae), using a nested half-sib breeding experiment. We found significant heritabilities for the size of most reproductive tract traits investigated in both females (spermathecae and their ducts, accessory glands and their ducts) and males (testis size but not sperm length). Within the sexes, phenotypic and genetic correlations were mostly nil or positive, suggesting functional integration of or condition-dependent investment in internal reproductive traits. Negative intrasexual genetic correlations, potentially suggestive of resource allocation trade-offs, were not evident. Intersexual genetic correlations were mostly positive, reflecting expected allometries between male and female morphologies. Most interestingly, testis size correlated positively with female accessory gland size and duct length, potentially indicative of a co-evolutionary arms race. We discuss these and alternative explanations for these patterns of genetic covariance.     

Temperature-dependent ovariole and testis maturation in the yellow dung fly

Temperature is one of the abiotic environmental factors most strongly affecting animal behaviour, physiology, and life history. In insects, lower temperatures generally slow down most physiological processes, reducing growth rate and prolonging the juvenile period. Here, we investigate temperature‐dependent ovariole and testis maturation in the anautogenous yellow dung fly, Scathophaga stercoraria L. (Diptera: Scathophagidae), and relate it to corresponding temperature effects on pre‐adult development time and the adult pre‐reproductive period. Flies were reared in the laboratory at three constant temperatures (18, 22, and 26 °C), and the size of the developing ovarioles and testes (reflecting sperm production) was measured over time (i.e., age). Ovariole size increased asymptotically over the first 12 days of adult life, while the testes continued to fill after day 10. In accordance with the temperature‐size rule, warmer temperatures resulted in smaller ovarioles (eggs) and smaller testes, independent of body size. Warmer temperatures also greatly reduced pre‐adult development time by more than half, from 12 to 25 °C, the larger males always taking 1–3 days longer than the females. Corresponding temperature effects on the adult pre‐reproductive period were small (<1 day between 15 and 25 °C), with males taking 5–6 days and females 10–13 days to first reproduction. Time lost by males during the pre‐adult stage, when ovaries and testes are produced, can thus be more than compensated‐for by time gained during the pre‐reproductive period, when eggs and sperm are produced, so males can nevertheless start reproducing sooner than females.