数量性状发育遗传模型及其分析方法的研究进展

发育遗传模型是同时反映性状遗传和发育本质、提供影响遗传变异及调整发育进程的有关因素的信息的模型。建立在群体遗传学基础上的直接效应模型适用于单一基因控制的简单性状。渐成模型将遗传变异分解成直接分量和渐成分量(母体效应和互作效应),能更好地反映有机体遗传和发育的生物学机制。生长轨迹模型有效地综合了复杂性状各分量的发育动态,可获得连续的、综合的、详细的、动态的发育信息。条件遗传分析方法不仅可以估算特定时间段的净效应,且可将净效应分解为不同遗传分量,了解各效应分量的相对贡献。 Abstract：Developmental genetic models and analysis methods for quantitative traits are presented.Developmental genetic models should reflect the genetic and developmental essence,and provide the information of the factors influencing the genetic variation and the developmental process.Direct effect models,which based on the population genetics,may be suitable to analyze simple traits with single gene.Epigenetic models can decompose the whole genetic variation into direct and epigenetic components (maternal effects and epigenetic interaction effects),so that biological mechanism can be better understood.Growth trace models effectively synthesize the developmental dynamics of components of complex traits.With them,continuous,compositive,detailed,and dynamic information of development is available.Conditional analysis method can not only estimate the net effects in a specific time interval,but also depose them into genetic components and help to appreciate the contributions of different effects.     

Pleiotropic effect, clock genes, and reproductive isolation

 The mechanism by which a clock gene pleiotropically controlling life history and behavioral traits causes reproductive isolation is explained using a model species, the melon fly, Bactrocera cucurbitae (Coquillett) (Diptera: Tephritidae). Melon flies mate once a day, at dusk. The population selected for life history traits exhibits correlated responses in the time of mating during the day. For example, the fly populations selected for faster (slower) development have an earlier (later) time of mating. A circadian rhythm controls the time of mating. The circadian periods in constant darkness were about 22 h in lines selected for a short developmental period and about 31 h in lines selected for a long developmental period. The data on crosses between the selected lines indicated that the developmental period is controlled by a polygene, whereas the circadian period may be controlled by a single clock gene. These results suggest a clock gene pleiotropically controls developmental and circadian periods in the melon fly. Reproductive isolation may often evolve as an indirect (pleiotropic) consequence of adaptation to different environments or habitats. For example, niches that are temporally or seasonally offset can select organisms with different developmental characteristics. These developmental differences can inadvertently cause reproductive isolation by a variety of means including shifts in mating activity patterns. The difference in time of mating between populations selected for developmental period translated into significant prezygotic isolation, as measured by mate choice tests. If the mating time between populations differed more than 1 h, the isolation index was significantly higher than zero. These findings indicate that premating isolation can be established by a pleiotropic effect of a clock gene. There are many examples in which the difference in timing of reproduction prevents gene flow between populations, such as the egg spawning time in marine organisms, the flowering time in angiosperms, and the time of mating in insects. In such organisms, if genetic correlations between circadian rhythm and reproductive traits exist, multifarious divergent selection for life history traits would often accelerate the evolution of reproductive isolation through clock genes. Natural populations may diverge in reproduction time through drift, direct natural selection for time of reproduction, or as a by-product effect of genetic correlations. In any case, clock genes are keys in reproductive isolation. Received: January 31, 2002 / Accepted: July 29, 2002 Acknowledgments I am grateful to Tetsuo Arai, Akira Matsumoto, Takashi Matsuyama, Toru Shimizu, Aya Takahashi, Teiichi Tanimura, Tetsuya Toyosato, and Yasuhiko Watari for useful discussion, and to the responsible editor and two anonymous reviewers for helpful suggestions. I also thank Yoshihiko Chiba, Norio Ishida, Emi Koyama, Kazuhiko Sakai, and Takaomi Sakai for useful information. My work on speciation has been supported by a Grant-in-Aid for Scientific Research (KAKENHI 14340244) from the Ministry of Education, Culture, Sports, Science and Technology of Japan.     

Correlated development, organism‐wide asymmetry and patterns of asymmetry in two moth species

Understanding the mechanisms that determine the development of a bilaterally symmetrical trait is crucial to the interpretation of patterns of fluctuating asymmetry (FA). Experimental and theoretical studies have indicated that feedback mechanisms both within and between developing traits, may participate in the developmental control of asymmetry. This study provides evidence that naturally occurring patterns of FA are affected by interactions between different traits. We found positive between‐trait correlations in signed FA values for tibia lengths on different legs, but not between wing and tibia FA in two moth species. Further research should investigate if trait functionality is related to this presumed correlated development. An extension of the Rashevsky–Turing model of morphogenesis further showed that correlations between the signed FA values can be generated by feedback mechanisms that regulate growth patterns between traits. We argue that such feedback mechanisms can be expected to be widespread and show that between‐trait correlations in the unsigned FA then become confounded with correlations in the signed FA. In addition, correlated development appeared to invalidate the use of the hypothetical repeatability to translate correlations between the unsigned FA values into correlations in the presumed underlying developmental instability. In conclusion, the presence of an organism‐wide asymmetry, which are most frequently found in morphologically integrated traits, may be even less common than previously thought. This revised version was published online in July 2006 with corrections to the Cover Date.     

Bayesian inference of genetic parameters for ultrasound scanning traits of Kivircik lambs

Ultrasound scanning traits have been adapted in selection programs in many countries to improve carcass traits for lean meat production. As the genetic parameters of the traits interested are important for breeding programs, the estimation of these parameters was aimed at the present investigation. The estimated parameters were direct and maternal heritability as well as genetic correlations between the studied traits. The traits were backfat thickness (BFT), skin+backfat thickness (SBFT), eye muscle depth (MD) and live weights at the day of scanning (LW). The breed investigated was Kivircik, which has a high quality of meat. Six different multi-trait animal models were fitted to determine the most suitable model for the data using Bayesian approach. Based on deviance information criterion, a model that includes direct additive genetic effects, maternal additive genetic effects, direct maternal genetic covariance and maternal permanent environmental effects revealed to be the most appropriate for the data, and therefore, inferences were built on the results of that model. The direct heritability estimates for BFT, SBFT, MD and LW were 0.26, 0.26, 0.23 and 0.09, whereas the maternal heritability estimates were 0.27, 0.27, 0.24 and 0.20, respectively. Negative genetic correlations were obtained between direct and maternal effects for BFT, SBFT and MD. Both direct and maternal genetic correlations between traits were favorable, whereas BFT–MD and SBFT–MD had negligible direct genetic correlation. The highest direct and maternal genetic correlations were between BFT and SBFT (0.39) and between MD and LW (0.48), respectively. Our results, in general, indicated that maternal effects should be accounted for in estimation of genetic parameters of ultrasound scanning traits in Kivircik lambs, and SBFT can be used as a selection criterion to improve BFT.     

The evolution of alternative developmental pathways: footprints of selection on life-history traits in a butterfly

Developmental pathways may evolve to optimize alternative phenotypes across environments. However, the maintenance of such adaptive plasticity under relaxed selection has received little study. We compare the expression of life-history traits across two developmental pathways in two populations of the butterfly Pararge aegeria where both populations express a diapause pathway but one never expresses direct development in nature. In the population with ongoing selection on both pathways, the difference between pathways in development time and growth rate was larger, whereas the difference in body size was smaller compared with the population experiencing relaxed selection on one pathway. This indicates that relaxed selection on the direct pathway has allowed life-history traits to drift towards values associated with lower fitness when following this pathway. Relaxed selection on direct development was also associated with a higher degree of genetic variation for protandry expressed as within-family sexual dimorphism in growth rate. Genetic correlations for larval growth rate across sexes and pathways were generally positive, with the notable exception of correlation estimates that involved directly developing males of the population that experienced relaxed selection on this pathway. We conclude that relaxed selection on one developmental pathway appears to have partly disrupted the developmental regulation of life-history trait expression. This in turn suggests that ongoing selection may be responsible for maintaining adaptive developmental regulation along alternative developmental pathways in these populations.     

National evaluation for calving ease,gestation length and birth weight by linear and threshold model methodologies

Data included 393,097 calving ease, 129,520 gestation length, and 412,484 birth weight records on 412,484 Gelbvieh cattle. Additionally, pedigrees were available on 72,123 animals. Included in the models were effects of sex and age of dam, treated as fixed, as well as direct, maternal genetic and permanent environmental effects and effects of contemporary group (herd-year-season), treated as random. In all analyses, birth weight and gestation length were treated as continuous traits. Calving ease (CE) was treated either as a continuous trait in a mixed linear model (LM), or as a categorical trait in linear-threshold models (LTM). Solutions in TM obtained by empirical Bayes (TMEB) and Monte Carlo (TMMC) methodologies were compared with those by LM. Due to the computational cost, only 10,000 samples were obtained for TMMC. For calving ease, correlations between LM and TMEB were 0.86 and 0.78 for direct and maternal genetic effects, respectively. The same correlations but between TMEB and TMMC were 1.00 and 0.98, respectively. The correlations between LM and TMMC were 0.85 and 0.75, respectively. The correlations for the linear traits were above.97 between LM and TMEB but as low as 0.91 between LM and TMMC, suggesting insufficient convergence of TMMC. Computing time required was about 2 hrs, 5 hrs, and 6 days for LM, TMEB and TMMC, respectively, and memory requirements were 169, 171, and 445 megabytes, respectively. Bayesian implementation of threshold model is simple, can be extended to multiple categorical traits, and allows easy calculation of accuracies; however, computing time is prohibitively long for large models.     

Developmental quantitative genetic models of evolutionary change

Discussions about evolutionary change in developmental processes or morphological structures are predicated on specific quantitative genetic models whose parameters predict whether evolutionary change can occur, its relative rate and direction, and if correlated change will occur in other related and unrelated structures. The appropriate genetic model should reflect the relevant genetical and developmental biology of the organisms, yet be simple enough in its parameters so that deductions can be made and hypotheses tested. As a consequence, the choice of the most appropriate genetic model for polygenically controlled traits is a complex tissue and the eventual choice of model is often a compromise between completeness of the model and computational expediency. Herein, we discuss several developmental quantitative genetic models for the evolution of development and morphology. The models range from the classical direct effects model to complex epigenetic models. Further, we demonstrate the algebraic equivalency of the Cowley and Atchley epigenetic model and Wagner's developmental mapping model. Finally, we propose a new multivariate model for continuous growth trajectories. The relative efficacy of these various models for understanding evolutionary change in developmental and morphological traits is discussed. © 1994 Wiley-Liss, Inc.     

LOCATING EVOLUTIONARY PRECURSORS ON A PHYLOGENETIC TREE

Conspicuous innovations in the history of life are often preceded by more cryptic genetic and developmental precursors. In many cases, these appear to be associated with recurring origins of very similar traits in close relatives (parallelisms) or striking convergences separated by deep time (deep homologies). Although the phylogenetic distribution of gain and loss of traits hints strongly at the existence of such precursors, no models of trait evolution currently permit inference about their location on a tree. Here we develop a new stochastic model, which explicitly captures the dependency implied by a precursor and permits estimation of precursor locations. We apply it to the evolution of extrafloral nectaries (EFNs), an ecologically significant trait mediating a widespread mutualism between plants and ants. In legumes, a species‐rich clade with morphologically diverse EFNs, the precursor model fits the data on EFN occurrences significantly better than conventional models. The model generates explicit hypotheses about the phylogenetic location of hypothetical precursors, which may help guide future studies of molecular genetic pathways underlying nectary position, development, and function.     

Lizards from Urban Areas Are More Asymmetric: Using Fluctuating Asymmetry to Evaluate Environmental Disturbance

The increase in human activities that leads to wildlife decline and species extinction poses an urgent need for simple indicators of environmental stress in animal populations. Several studies have suggested that fluctuating asymmetry (FA) can be an easy, direct measure of developmental instability because it is associated to environmental stress and, as such, it can be a useful indicator of population disturbance. We examined three different morphological traits in urban and rural populations of the common wall lizard (Podarcis muralis) to test whether anthropogenic disturbance causes an increase in FA. Compared to rural populations, urban ones showed higher levels of FA in all analyzed traits, thus providing evidence that FA can respond to anthropogenic disturbance. However, we also found significant differences in FA among traits, where femoral pores and subdigital lamellae, traits with a functional relevance, were more stable developmentally compared to supracilliar granules which have no evident function. Unsigned FA [abs(right-left)] exhibited significant, but weak, positive correlations among traits, indicating that developmental noise does not have a uniform effect across characters and thus questioning the view of developmental stability as an organism-wide property. The degree of signed FA (right-left) was more similar between structurally associated traits, possibly as an outcome of morphological integration. In conclusion, our results demonstrate that FA can be a reliable indicator of disturbance provided that it is analyzed on multiple traits simultaneously and examined at the population level.     

Individual differences in developmental precision and fluctuating asymmetry: a model and its implications

In many studies, fluctuating asymmetry (FA) has been used as a measure of individual differences in developmental imprecision. A model of how variation in developmental imprecision is associated with variation in asymmetry is described and applied to important issues about FA. If individual differences in developmental imprecision exist, asymmetry due to developmental error should be leptokurtically distributed. Moreover, the greater the magnitude of individual differences, the greater the leptokurtosis. Asymmetry purportedly due to developmental error in a variety of species is indeed leptokurtically distributed. The level of leptokurtosis suggests that the CV in individual differences in underlying developmental imprecision is generally 20–25, consistent with it being a fitness trait. In addition, data suggest that: (1) the individual differences that underlie the developmental imprecision of different traits are largely shared across traits and not trait-specific; (2) the heritability of these individual differences may average between 35 and 55%, despite small heritabilities of individual trait FAs; and (3) correlations between FA and fitness traits or components suggest high correlations between underlying variation in developmental precision and fitness in many species. Theoretical implications are discussed.     

THE EFFECT OF A VARIABLE ENVIRONMENT ON THE GENETIC CORRELATION STRUCTURE IN A FIELD CRICKET

The evolutionary trajectory of a trait depends not only on the presence of genetic variation, but also on the pattern of genetic correlations (r_g) among traits. Genetic correlations are most easily measured under homogeneous, controlled laboratory conditions, whereas natural populations typically experience a higher degree of environmental variability. The effect of environmental variability on genetic correlations in the cricket, Gryllus pennsylvanicus, was studied by measuring genetic correlations within and between two environments differing in levels of environmental heterogeneity. Within-environment r_g among morphological traits measured in the homogeneous laboratory environment were found to be reliable predictors of r_g measured in the experimental field environment. Laboratory measures of r_g involving life-history traits, though, were not found to reflect the same correlations measured in the heterogeneous environment. A significant negative genetic correlation between fecundity and developmental time was found in the field environment, yet was not detectable when measured in the laboratory. Phenotypic correlations may be obtained much more easily than genetic correlations, but their usefulness in evolutionary inference depends on the pattern of similarity between the two correlations. A comparison of genetic and phenotypic correlations revealed a close match between the two measures for morphological traits, but revealed only broad similarities when considering life-history traits. Male-female genetic correlations between morphological traits were high (all r_g > 0.73) and were consistently higher in the field environment than in the laboratory. The genetic correlations between the sexes in developmental time followed the same trend, but the male-female genetic correlation of gonad weights was low in both environments. Across-environment correlations were found to be strong for morphological traits and for gonad weight, whereas the genetic expression of developmental time was found to be dependent on the environment in which the crickets were raised.     

Intralocus tactical conflict: genetic correlations between fighters and sneakers of the dung beetle Onthophagus taurus

Males and females differ in their phenotypic optima for many traits, and as the majority of genes are expressed in both sexes, some alleles can be beneficial to one sex but harmful to the other (intralocus sexual conflict; ISC). ISC theory has recently been extended to intrasexual dimorphisms, where certain alleles may have opposite effects on the fitness of males of different morphs that employ alternative reproductive tactics (intralocus tactical conflict; ITC). Here, we use a half‐sib breeding design to investigate the genetic basis for ISC and ITC in the dung beetle Onthophagus taurus. We found positive heritabilities and intersexual genetic correlations for almost all traits investigated. Next, we calculated the intrasexual genetic correlation between males of different morphs for horn length, a sexually selected trait, and compared it to intrasexual correlations for naturally selected traits in both sexes. Intrasexual genetic correlations did not differ significantly between the sexes or between naturally and sexually selected traits, failing to support the hypothesis that horns present a reduction of intrasexual genetic correlations due to ITC. We discuss the implications for the idea of developmental reprogramming between male morphs and emphasize the importance of genetic correlations as constraints for the evolution of dimorphisms.     

COMPOSITE TRAITS,SELECTION RESPONSE,AND EVOLUTION

It is very difficult to relate macroevolutionary patterns to the microevolutionary processes described by quantitative-genetic models. Quantitative-genetic parameters are statistical abstractions. Their long-term significance and evolution might be understood if they can be related to development, physiology, and other biological properties. Most continuous traits are composites of other traits that may contribute differentially to selection response and long-term divergence. The operation of selection on continuous traits can be indirect, with intermediate optima caused by correlated fitness components. Few realistic models are available, and heritable maternal effects can further complicate selection response. Examples involving allometry of brain and body size in mammals suggest that prenatal and postnatal growth have contributed differently to body-size evolution, with different correlated changes in brain size. Several different models could explain these patterns, and interpretation is further complicated by statistical difficulties in comparative biology. Quantitative-genetic models may become more informative and predictive if variation in their parameters can be explained by developmental and other biological processes that have been shaped by the previous history of the population.     

Linear and threshold analysis of direct and maternal genetic effects for secondary sex ratio in Iranian buffaloes

The objective of this study was to estimate variance components and genetic parameters for secondary sex ratio (SSR) in Iranian buffaloes. Calving records from April 1995 to June 2010 comprising 15,207 calving events from the first three lactations of 1066 buffalo herds of Iran were analyzed using linear and threshold animal models to estimate variance components, heritabilities and genetic correlations between direct and maternal genetic effects for SSR. Linear and threshold animal models included direct and maternal genetic effects with covariance between them and maternal permanent environmental effects were implemented by Gibbs sampling methodology. Posterior means of direct and maternal heritabilities and repeatability for SSR obtained from linear animal model were 0.15, 0.10, and 0.17, respectively. Threshold estimates of direct and maternal heritabilities and repeatability for SSR were 0.48, 0.27, and 0.52, respectively. The results showed that the correlations between direct and maternal genetic effects of SSR were negative and high in both models. In addition, the ratios of maternal permanent environmental variance were low. Exploitable genetic variation in SSR can take advantage of sexual dimorphism for economically important traits which may facilitate greater selection intensity and thus greater response to selection, as well as reducing the replacement costs. Threshold animal model may be applied in selection programs where animals are to be genetically ranked for female rate.     

Building Developmental Integration into Functional Systems: Function-Induced Integration of Mandibular Shape

The mammalian mandible is a developmentally modular but functionally integrated system. Whether morphological integration can evolve to match the optimal pattern of functional integration may depend on the developmental origin of integration, specifically, on the role that direct epigenetic interactions play in shaping integration. These interactions are hypothesized to integrate modules and also to be highly conservative, potentially constraining the evolution of integration. Using the fox squirrel (Sciurus niger) mandible as a model system, we test five a priori developmental hypotheses that predict mandibular integration and we also explore for correlations between shapes of mandibular regions not anticipated by any of the developmental models. To determine whether direct epigenetic interactions are highly conserved in rodents, we examine the correlation structure of fluctuating asymmetry, and compare integration patterns between fox squirrels and prairie deer mice (Peromyscus maniculatus bairdii). In fox squirrels, we find a correlation structure unanticipated by all a priori developmental models: adjacent parts along the proximodistal jaw axis are correlated whereas more distant ones are not. The most notable exception is that the shape of the anterior incisor alveolus is correlated with the shape of the ramus (FA component) or coronoid (symmetric component). Those exceptions differ between species; in prairie deer mice, the molar alveolus is connected to more parts, and the incisor alveolus to fewer, than in fox squirrels. The structure of integration suggests that the mandible cannot be decomposed into parts but rather is a single connected unit, a result consistent with its functional integration. That match between functional and developmental integration may arise, at least in part, from function-induced growth, building developmental integration into the functional system and enabling direct epigenetic interactions to evolve when function does.     

Phenotypic Integration Without Modularity: Testing Hypotheses About the Distribution of Pleiotropic Quantitative Trait Loci in a Continuous Space

Theories of phenotypic integration have relied heavily on the concept of modularity in order to model the ways in which traits in an organism correlate and covary. Recent investigations suggest that, while some functional and developmental processes may be morphologically and ontogenetically localized, and thus modular in a developmental sense, there is a great deal of overlap among these influences on patterns of integration in the adult form. This can result in blurry boundaries between hypothesized modules constructed to test hypotheses about phenotypic integration. This investigation tests hypotheses about the contribution of pleiotropic quantitative trait loci (QTL) to phenotypic integration in the mouse mandible without using a priori categorical hypotheses about which traits constitute a module. We ask two main questions: (1) Are the effects of pleiotropic QTL localized to highly correlated traits or more spread out among traits than one might expect by chance? (2) Does the pattern of trait influence when all pleiotropic QTL are considered together deviate from what we might expect if QTL affect traits without regard for the correlations among traits? We find that a large proportion of pleiotropic QTL affect traits that are more highly correlated than we expect by chance with the remainder having effects that are distributed as if by chance. Furthermore, the overall distribution of the effects of pleiotropic QTL differs significantly from the null distribution of no association between pleiotropic effects on traits and correlations among traits. The main modular hypothesis used by earlier studies often does not predict the distribution of sets of traits sharing a common QTL. These results suggest that there is a clear tendency for pleiotropic effects of QTL to be localized but that the localization may be best thought of as occurring in a continuous space rather being clustered in discrete modules.     

A quantitative genetic test of adaptive decoupling across metamorphosis for locomotor and life-history traits in the pacific tree frog, Hyla regilla

Metamorphosis is assumed to be beneficial because it can break developmental links between traits in the different phases of a complex life-cycle and thereby allow larval and adult phases to adapt independently. I tested the prediction that correlations between the larval and adult phases are smaller than within stages. I estimated phenotypic and additive genetic variances and correlations for tadpole swimming speed, frog jump distance, body size, and larval period in a single population of the Pacific tree frog, Hyla regilla. These traits are known or reasonably assumed to be important for survival in this and other anuran species from temporary ponds. Only the three size variables were affected by sire identity. Heritabilities for locomotor performance, larval period, and size-independent performance were low (0.00-0.23) and not significant. Body size measurements showed somewhat higher and statistically significant heritabilities (0.24-0.34). Most traits were phenotypically correlated. On average, phenotypic correlations were larger between phases than within phases (0.41 vs. 0.28). Genetic correlations involving body-size traits were positive and large, and average within- and between-phase genetic correlation coefficients were identical (0.81). These results do not support the adaptive decoupling hypothesis, and they indicate that a paucity of additive genetic variation is a likely constraint on the evolution of traits measured for this population.     

Selection for developmental time in bean weevil (Acanthoscelides obtectus): correlated responses for other life history traits and genetic architecture of line differentiation

In this article we investigate the direct and correlated responses to selection for developmental time in order to discern differences between lines in several preadult and adult life history traits of Acanthoscelides obtectus (Coleoptera, Bruchidae). Selection for fast development was about five times as effective as selection for slow development, as judged by realized heritabilities. The correlated responses on the following life‐history traits were studied: egg size, hatching success, embryonic developmental time, egg‐to‐adult viability, body weight, first day of egg laying, total fecundity, and longevity. Analyses of the terminal generation of selection showed that all life history traits examined, except for hatching success, were affected by selection. The findings suggest that body weight, total fecundity, and longevity traded off to preadult developmental time. Unlike the adult traits, none of the preadult traits showed negative correlations with developmental time. We also present data concerning the underlying genetic basis that produces changes in preadult developmental time, body weight, and egg‐to‐adult viability in the lines selected for fast and slow preadult developmental time. Additive‐dominance genetic architecture for both preadult developmental time and body weight was found. In addition, it appears that the responses to selection for preadult developmental time involved between 10 and 28 loci, which were correlated with at least one to four genes for body weight. Epistasis makes a significant contribution to genetic divergence between fast and slow selected lines only with respect to preadult viability. The observed levels of dominance and epistasis underscore the important role of nonadditive genetic effects to the adaptive diversifications of bean weevil populations.     

Modular phenotypic plasticity: divergent responses of barnacle penis and feeding leg form to variation in density and wave-exposure

Traits can evolve both in response to direct selection and in response to indirect selection on other linked traits. Although the evolutionary significance of coupled traits (e.g., through shared components of developmental pathways, or through competition for shared developmental resources) is now well accepted, we know comparatively little about how developmental coupling may restrict the independent responses of two or more phenotypically plastic traits in response to conflicting environmental cues. Such studies are important because coupled development, if present, could act as an important limit to the evolution of functionally independent plasticity in multiple traits. I tested whether developmental coupling can restrict the direction of plastic responses by studying how penis form and leg form--both highly plastic traits of barnacles--varied in response to differences in conspecific density and water velocity. Penis length and leg length in Balanus glandula varied in parallel with variation in wave-exposure but varied in opposite directions with variation in conspecific density. This study represents one of the rare tests of developmental coupling between multiple (demonstrably adaptive) plastic traits: Barnacle legs and penises appear to exhibit modular development that can respond concurrently--yet in independent directions--to conflicting environmental cues.