Descent with modification: the unity underlying homology and homoplasy as seen through an analysis of development and evolution

Homology is at the foundation of comparative studies in biology at all levels from genes to phenotypes. Homology is similarity because of common descent and ancestry, homoplasy is similarity arrived at via independent evolution. However, given that there is but one tree of life, all organisms, and therefore all features of organisms, share some degree of relationship and similarity one to another. That sharing may be similarity or even identity of structure and the sharing of a most recent common ancestor--as in the homology of the arms of humans and apes--or it may reflect some (often small) degree of similarity, such as that between the wings of insects and the wings of birds, groups whose shared ancestor lies deep within the evolutionary history of the Metazoa. It may reflect sharing of entire developmental pathways, partial sharing, or divergent pathways. This review compares features classified as homologous with the classes of features normally grouped as homoplastic, the latter being convergence, parallelism, reversals, rudiments, vestiges, and atavisms. On the one hand, developmental mechanisms may be conserved, even when a complete structure does not form (rudiments, vestiges), or when a structure appears only in some individuals (atavisms). On the other hand, different developmental mechanisms can produce similar (homologous) features. Joint examination of nearness of relationship and degree of shared development reveals a continuum within an expanded category of homology, extending from homology --> reversals --> rudiments --> vestiges --> atavisms --> parallelism, with convergence as the only class of homoplasy, an idea that turns out to be surprisingly old. This realignment provides a glimmer of a way to bridge phylogenetic and developmental approaches to homology and homoplasy, a bridge that should provide a key pillar for evolutionary developmental biology (evo-devo). It will not, and in a practical sense cannot, alter how homoplastic features are identified in phylogenetic analyses. But seeing rudiments, reversals, vestiges, atavisms and parallelism as closer to homology than to homoplasy should guide us toward searching for the common elements underlying the formation of the phenotype (what some have called the deep homology of genetic and/or cellular mechanisms), rather than discussing features in terms of shared or independent evolution.     

DEVELOPMENTAL MECHANISMS UNDERLYING THE FORMATION OF ATAVISMS

1. Atavisms emerge as evidence of localized modifications in development of an organ or of one of its parts. Different developmental processes can be triggered within the same organ rudiment, presumably in response to the same stimulus. We saw that that stimulus can have a genetic basis in a mutational event, which can be selected for. We also saw that atavism can be produced by experimental manipulation within developing systems -increased growth of the chick fibula, enamel production from avian ectoderm, and balancer formation in amphibians. Such atavisms are not based on heritable genetic changes. They indicate the developmental plasticity that exists within embryos and the relative ease with which development can be switched from one programme to another. 2. Examination of mutants (wingless chicks), limbless vertebrates and experimental manipulation of embryos, shows that cell death, inductive tissue interactions and altered patterns of growth are developmental mechanisms used in the formation of atavisms. 3. Differential development mechanisms can be triggered within the same organ at the same time to produce atavisms. In the guinea pig, formation of atavistic digit V involves prolongation of growth of metatarsal V whereas formation of atavistic digit I involves development of a new metatarsal I. 4. Secondary functional modifications ensure that the atavism is integrated with the other components of the functional unit, as illustrated by extra digits in horses or guinea pigs and fibulae in birds. Atavistic 2nd and 4th digits in horses arise by continued growth of their primordia. A consequent reduction in the growth rate of digit 3, the normal single functional digit, enables all three digits to attain approximately equal lengths and so potentially to function. The altered functional load transmitted to the limbs results in secondary but correlated alterations in muscles and skeletal elements in other portions of the limbs. The fact that embryonic digit 2 normally develops to a more advanced state than digit 4 explains why digit 2 more often develops atavistically, for if variation in growth rate is the basis for the atavistic digit, digit 2 has an advantage over digit 4. 5. Atavisms should not be an embarrassment to the evolutionary biologist. They are the outward and visible sign of a hidden potential for morphology change possessed by all organisms. Neither basic capacity to form the organ nor patterning information is lost. Modification of components of inductive tissue interactions helps to explain how organs are lost during evolution (also see Regal, 1977); retention of the basic mechanism explains how structures can be revived as atavisms (also see Rachootin & Thomson, 1981). Frequency of atavisms thus provides an indication of the degree of modification or loss of the underlying developmental programme.     

Charles Darwin,embryology, evolution and skeletal plasticity

Darwin provided us with the theory of evolutionary change through natural selection. Just as important to the science of biology was Darwin’s recognition that all organisms could be classified and were related to one another because they arose from a single common universal ancestor – what we know as the universal tree of life (UtoL). All the features of the skeletal biology of fish therefore can be explained, both in an evolutionary framework (ultimate causation) and in the framework of development, growth and physiology (proximate causation). Neither approach is complete without the other. I will outline the elements of Darwin’s theories on evolution and classification and, as importantly, discuss what was missing from Darwin’s theories. An important class of evidence for evolution used by Darwin came from embryology, both comparative embryology and the existence of vestiges and atavisms. After discussing this evidence I examine some fundamental features of skeletal development and evolution These include: the presence of four skeletal systems in all vertebrates; the existence of two skeletons, one based on cartilage, the other on bone and dentine; the modular nature of skeletal development and evolution; and the plasticity of the skeleton in response to either genetic or environmental changes.     

Possible atavisms of genitalia in two species of earwig (Dermaptera), Proreus simulans (Chelisochidae) and Euborellia plebeja (Anisolabididae)

Male and female genitalia generally show a rapid evolutionary rate, which raises the problems related to homologization and the determination of the polarities of evolutionary changes. In earwigs (Dermaptera), multiple or branched female sperm-storage organs (spermathecae) have been reported for members of the Karschiellidae, Pygidicranidae, and Diplatyidae, collectively termed the “basal” Dermaptera. Whether the complicated spermathecae represent a plesiomorphy or an apomorphy has not been resolved. Here I report the occurrence of multiple or branched spermathecae in gamma-irradiated samples of two earwig species, Euborellia plebeja (Dohrn, 1863) (Anisolabididae) and Proreus simulans (Stål, 1860) (Chelisochidae), which belong to the “higher” Dermaptera (Apachyidae, Labiduridae, Anisolabididae, Spongiphoridae, Chelisochidae, and Forficulidae). Females belonging to the higher Dermaptera normally have a single-unbranched spermatheca. I discuss examples of possible atavisms in relation to the evolutionary pathways of spermathecal morphology. Possible atavisms in the number of male organs for sperm transfer (virgae) are also reported.     

Ecological interactions drive evolutionary loss of traits

Loss of traits can dramatically alter the fate of species. Evidence is rapidly accumulating that the prevalence of trait loss is grossly underestimated. New findings demonstrate that traits can be lost without affecting the external phenotype, provided the lost function is compensated for by species interactions. This is important because trait loss can tighten the ecological relationship between partners, affecting the maintenance of species interactions. Here, we develop a new perspective on so-called `compensated trait loss' and how this type of trait loss may affect the evolutionary dynamics between interacting organisms. We argue that: (1) the frequency of compensated trait loss is currently underestimated because it can go unnoticed as long as ecological interactions are maintained; (2) by analysing known cases of trait loss, specific factors promoting compensated trait loss can be identified and (3) genomic sequencing is a key way forwards in detecting compensated trait loss. We present a comprehensive literature survey showing that compensated trait loss is taxonomically widespread, can involve essential traits, and often occurs as replicated evolutionary events. Despite its hidden nature, compensated trait loss is important in directing evolutionary dynamics of ecological relationships and has the potential to change facultative ecological interactions into obligatory ones.     

Pattern and process: evidence for the evolution of photosynthetic traits in natural populations

The patterns of interspecific variation identified by comparative studies provide valuable hypotheses about the role of physiological traits in evolutionary adaptation. This review covers tests of these hypotheses for photosynthetic traits that have used a microevolutionary perspective to characterize physiological variation among and within populations. Studies of physiological differentiation among populations show that evolutionary divergence in photosynthetic traits is common within species, and has a pattern that supports many adaptive hypotheses. These among-population studies imply that selection has influenced photosynthetic traits in some way, but they are not designed to identify the traits targeted by selection or the environmental agents that cause selection. Analyses of genetic and phenotypic variation within populations address these questions. Studies that have quantified genetic variation within populations show that levels of heritable variation can be adequate for evolutionary change in photosynthetic traits. Other studies have measured phenotypic selection for these traits by analyzing how the variation within populations is correlated with fitness. This work has shown that selection for photosynthetic traits may often operate indirectly via correlations with other traits, and emphasizes the importance of viewing the phenotype as an integrated function of growth, morphology, life-history and physiology. We also outline some methodological problems that may be encountered for ecophysiological traits by these types of studies, provide some potential solutions, and discuss future directions for the field of plant evolutionary ecophysiology.     

Restoring avian wing digits

The precise identification of the digits of the avian wing is of importance in evolutionary studies. If the digits are numbered two, three and four, this has been taken to suggest that birds are not descended directly from dinosaurs. If the digits are numbered one, two and three, dinosaur origins become more plausible. Studies of the development of the avian wing have failed to resolve this dilemma. However, in some instances, it is possible to deduce information about evolutionary morphologies by manipulating development experimentally. We grafted beads loaded with fibroblast growth factor 4 into the distal tip of chick wing buds at times when the apical ectodermal ridge is regressing. The consequence was that the cartilage structure conventionally labelled ''element 5'' increased dramatically in size and acquired a digit-like morphology in some instances. Corresponding changes in soft tissue morphology were also observed. We conclude that it may be possible to resolve the issue of avian digit homology by the induction of experimental atavisms of this kind.     

Single rays as indicators of fin vestigialization in coho salmon: patterns and perspectives

Although finrays in salmon normally contain a pair of elements (biramous), finrays with a single element (uniramous) occasionally develop. Exposure to chronic stress during character development has been shown to increase fluctuating asymmetry, suggesting the occurrence of single finrays may be stress‐induced. On the other hand, single finrays may be evolutionary atavisms, reflecting fin vestigialization caused by reduced selection pressure. To assess the merits of these hypotheses, cleared and stained paired and median fins were examined for single finrays in juvenile coho salmon (Oncorhynchus kisutch Walbaum) from two compatible hatchery stocks and their reciprocal hybrids which had been exposed to different patterns of chronic temperature fluctuation throughout embryogenesis. In the median fins, uniramous secondary finrays predominated, and single primary finrays were infrequent. Single finrays in the median fins did not respond to thermal treatment or cross, suggesting the fins were evolutionarily stable and under strong developmental control. The paired fins were observed to contain only primary finrays. Frequencies of single pelvic finrays increased under thermal stress, as did fluctuating asymmetry, suggesting increased sensitivity to stress due to reduced developmental control in this fin. However, the presence or absence of single finrays in the paired fins did not alter the statistical significance of the conclusions regarding levels of fluctuating asymmetry, the number of asymmetric fish, or the contribution to meristic variation from asymmetry. Locations of single finrays in the paired fins were unaffected by thermal treatment or cross. Single finrays were most commonly observed in the trailing margins of both paired fins, a finding consistent with vestigialization theory. Frequency histograms of single pectoral finray locations revealed a second peak in the leading quarter of the fin. The esults support the hypothesis that single primary finrays are evolutionary atavisms, and that reduced selection pressure is differentially influencing the paired fins.     

Fire as an evolutionary pressure shaping plant traits

Traits, such as resprouting, serotiny and germination by heat and smoke, are adaptive in fire-prone environments. However, plants are not adapted to fire per se but to fire regimes. Species can be threatened when humans alter the regime, often by increasing or decreasing fire frequency. Fire-adaptive traits are potentially the result of different evolutionary pathways. Distinguishing between traits that are adaptations originating in response to fire or exaptations originating in response to other factors might not always be possible. However, fire has been a factor throughout the history of land-plant evolution and is not strictly a Neogene phenomenon. Mesozoic fossils show evidence of fire-adaptive traits and, in some lineages, these might have persisted to the present as fire adaptations.     

Prudent males,group adaptation,and the tragedy of the commons

For almost five decades three threads have coexisted in the evolutionary and ecological literature, with their links only recently becoming visible and some of them still not properly addressed. These are the levels of selection debate, the metaphor of the tragedy of the commons, and the evolutionary study of sexual conflict. We analyze the eco‐evolutionary dynamics of a curious system where an asexual all‐female fish species (the Amazon molly Poecilia formosa) requires sperm from other species as a developmental trigger, without utilizing the genes from sperm. The dynamics of such a system bear strong resemblance to host–parasite dynamics, and populations of the sexual ‘host’ species persist much better if males avoid mating with Amazons. However, such avoidance may compromise their current mating success, and if this is the case, prudent mating becomes an altruistic trait that helps to keep an accumulating problem of a competing species at bay, and Amazon‐free space can be seen to form a common good that a population should maintain for future generations. A model shows that the evolution of altruistic mating restraint is possible but selection for short‐term gains means that it will remain less than perfect. This helps to explain why the anomalous gynogenetic system can persist, but it also raises questions about what kinds of traits can be classified as adaptations when optimization is not perfect and traits evolve to achieve short‐term goals better than long‐term performance. Contributing to the levels of selection debate, we encourage researchers to study the implications of the different timescales involved in the eco‐evolutionary process.     

Understanding Language Evolution: Beyond Pan-Centrism

Language does not fossilize but this does not mean that the language's evolutionary timeline is lost forever. Great apes provide a window back in time on our last prelinguistic ancestor's communication and cognition. Phylogeny and cladistics implicitly conjure Pan (chimpanzees, bonobos) as a superior (often the only) model for language evolution compared with earlier diverging lineages, Gorilla and Pongo (orangutans). Here, in reviewing the literature, it is shown that Pan do not surpass other great apes along genetic, cognitive, ecologic, or vocal traits that are putatively paramount for language onset and evolution. Instead, revived herein is the idea that only by abandoning single-species models and learning about the variation among great apes, there might be a chance to retrieve lost fragments of the evolutionary timeline of language.     

Social traits,social networks and evolutionary biology

The social environment is both an important agent of selection for most organisms, and an emergent property of their interactions. As an aggregation of interactions among members of a population, the social environment is a product of many sets of relationships and so can be represented as a network or matrix. Social network analysis in animals has focused on why these networks possess the structure they do, and whether individuals’ network traits, representing some aspect of their social phenotype, relate to their fitness. Meanwhile, quantitative geneticists have demonstrated that traits expressed in a social context can depend on the phenotypes and genotypes of interacting partners, leading to influences of the social environment on the traits and fitness of individuals and the evolutionary trajectories of populations. Therefore, both fields are investigating similar topics, yet have arrived at these points relatively independently. We review how these approaches are diverged, and yet how they retain clear parallelism and so strong potential for complementarity. This demonstrates that, despite separate bodies of theory, advances in one might inform the other. Techniques in network analysis for quantifying social phenotypes, and for identifying community structure, should be useful for those studying the relationship between individual behaviour and group‐level phenotypes. Entering social association matrices into quantitative genetic models may also reduce bias in heritability estimates, and allow the estimation of the influence of social connectedness on trait expression. Current methods for measuring natural selection in a social context explicitly account for the fact that a trait is not necessarily the property of a single individual, something the network approaches have not yet considered when relating network metrics to individual fitness. Harnessing evolutionary models that consider traits affected by genes in other individuals (i.e. indirect genetic effects) provides the potential to understand how entire networks of social interactions in populations influence phenotypes and predict how these traits may evolve. By theoretical integration of social network analysis and quantitative genetics, we hope to identify areas of compatibility and incompatibility and to direct research efforts towards the most promising areas. Continuing this synthesis could provide important insights into the evolution of traits expressed in a social context and the evolutionary consequences of complex and nuanced social phenotypes.     

Incomplete loss of a conserved trait: function,latitudinal cline,and genetic constraints

Retention of nonfunctional traits over evolutionary time is puzzling, because the cost of trait production should drive loss. Indeed, several studies have found nonfunctional traits are rapidly eliminated by selection. However, theory suggests that complex genetic interactions and a lack of genetic variance can constrain evolution, including trait loss. In the mustard family Brassicaceae the conserved floral condition includes four long and two short stamens, but we show that short stamens in the highly self‐pollinating mustard Arabidopsis thaliana do not significantly increase selfed seed set, suggesting that the trait has lost most or all of its function after the transition to selfing. We find that short stamen loss is common in native populations. Loss is incomplete and decreases with increasing latitude, a cline unexplained by correlations with flowering time or ovule count (which also vary with latitude). Using recombinant inbred lines derived from a cross between plants at the latitudinal extremes of the native range, we found three QTLs affecting short stamen number, with epistasis among them constraining stamen loss. Constraints on stamen loss from both epistasis and low genetic variance may be augmented by high selfing rates, suggesting that these kinds of constraints may be common in inbred species.     

EVOLUTIONARY NOVELTY AND ATAVISM IN THE SEMIONOTUS COMPLEX: RELAXED SELECTION DURING COLONIZATION OF AN EXPANDING LAKE

Fishes of the genus Semionotus diversified in the rift lakes of eastern North America during the Mesozoic (Newark Supergroup). Like the well-known cichlid fishes of the African great lakes, diverse complexes of semionotids were apparently endemic to a number of different lakes. Semionotid fishes show considerable morphological diversity in body shape and in a modified row of scales termed “dorsal ridge scales.” A number of distinct dorsal-ridge-scale patterns characterize groups of species from the Newark Supergroup. Interestingly, about 5.5% of individuals examined have anomalous scales mixed in with otherwise stereotypic dorsal-ridge-scale patterns. In this study, I take advantage of nearly annual stratigraphic resolution to determine whether dorsal-ridge-scale anomalies are concentrated stratigraphically in the early phase of lake formation and colonization by semionotids. More than 1,700 specimens of semionotid fish were collected from a single lake deposit (cycle P4), representing approximately 21,000 years, in the Early-Jurassic Towaco Formation of the Newark Basin. Dorsal-ridge-scale anomalies are significantly more frequent in older than in younger lake sediments, which I interpret as being the result of relaxed selection during the early colonization of the lake. Anomalous variation parallels variation in dorsal ridge scales between species-groups. Some anomalies are atavisms, while others are unique or foreshadow future evolutionary events. One type of anomaly is incorporated into the dorsal-ridge-scale series of two new species that gave rise to a radiation in a subsequent lake filling the same topographic basin. Because both novelties and atavisms occur in the dorsal-ridge-scale series of single individuals, I argue that the disruptions of the same “developmental program” produced both atavistic and novel traits.     

Variation,selection and evolution of function-valued traits

We describe an emerging framework for understanding variation, selection and evolution of phenotypic traits that are mathematical functions. We use one specific empirical example – thermal performance curves (TPCs) for growth rates of caterpillars – to demonstrate how models for function-valued traits are natural extensions of more familiar, multivariate models for correlated, quantitative traits. We emphasize three main points. First, because function-valued traits are continuous functions, there are important constraints on their patterns of variation that are not captured by multivariate models. Phenotypic and genetic variation in function-valued traits can be quantified in terms of variance-covariance functions and their associated eigenfunctions: we illustrate how these are estimated as well as their biological interpretations for TPCs. Second, selection on a function-valued trait is itself a function, defined in terms of selection gradient functions. For TPCs, the selection gradient describes how the relationship between an organism's performance and its fitness varies as a function of its temperature. We show how the form of the selection gradient function for TPCs relates to the frequency distribution of environmental states (caterpillar temperatures) during selection. Third, we can predict evolutionary responses of function-valued traits in terms of the genetic variance-covariance and the selection gradient functions. We illustrate how non-linear evolutionary responses of TPCs may occur even when the mean phenotype and the selection gradient are themselves linear functions of temperature. Finally, we discuss some of the methodological and empirical challenges for future studies of the evolution of function-valued traits.     

Evolutionary rates for multivariate traits: the role of selection and genetic variation

A fundamental question in evolutionary biology is the relative importance of selection and genetic architecture in determining evolutionary rates. Adaptive evolution can be described by the multivariate breeders'' equation (), which predicts evolutionary change for a suite of phenotypic traits () as a product of directional selection acting on them (β) and the genetic variance–covariance matrix for those traits (G). Despite being empirically challenging to estimate, there are enough published estimates of G and β to allow for synthesis of general patterns across species. We use published estimates to test the hypotheses that there are systematic differences in the rate of evolution among trait types, and that these differences are, in part, due to genetic architecture. We find some evidence that sexually selected traits exhibit faster rates of evolution compared with life-history or morphological traits. This difference does not appear to be related to stronger selection on sexually selected traits. Using numerous proposed approaches to quantifying the shape, size and structure of G, we examine how these parameters relate to one another, and how they vary among taxonomic and trait groupings. Despite considerable variation, they do not explain the observed differences in evolutionary rates.     

Evolutionary predictions should be based on individual-level traits

Recent theoretical studies have analyzed the evolution of habitat specialization using either the logistic or the Ricker equation. These studies have implemented evolutionary change directly in population-level parameters such as habitat-specific intrinsic growth rates r or carrying capacities K. This approach is a shortcut to a more detailed analysis where evolutionary change is studied in underlying morphological, physiological, or behavioral traits at the level of the individual that contribute to r or K. Here we describe two pitfalls that can occur when such a shortcut is employed. First, population-level parameters that appear as independent variables in a population dynamical model might not be independent when derived from processes at the individual level. Second, patterns of covariation between individual-level traits are usually not conserved when mapped to the level of demographic parameters. Nonlinear mappings constrain the curvature of trade-offs that can sensibly be assumed at the population level. To illustrate these results, we derive a two-habitat version of the logistic and Ricker equations from individual-level processes and compare the evolutionary dynamics of habitat-specific carrying capacities with those of underlying individual-level traits contributing to the carrying capacities. Finally, we sketch how our viewpoint affects the results of earlier studies.     

Estimating evolutionary parameters when viability selection is operating

Some individuals die before a trait is measured or expressed (the invisible fraction), and some relevant traits are not measured in any individual (missing traits). This paper discusses how these concepts can be cast in terms of missing data problems from statistics. Using missing data theory, I show formally the conditions under which a valid evolutionary inference is possible when the invisible fraction and/or missing traits are ignored. These conditions are restrictive and unlikely to be met in even the most comprehensive long-term studies. When these conditions are not met, many selection and quantitative genetic parameters cannot be estimated accurately unless the missing data process is explicitly modelled. Surprisingly, this does not seem to have been attempted in evolutionary biology. In the case of the invisible fraction, viability selection and the missing data process are often intimately linked. In such cases, models used in survival analysis can be extended to provide a flexible and justified model of the missing data mechanism. Although missing traits pose a more difficult problem, important biological parameters can still be estimated without bias when appropriate techniques are used. This is in contrast to current methods which have large biases and poor precision. Generally, the quantitative genetic approach is shown to be superior to phenotypic studies of selection when invisible fractions or missing traits exist because part of the missing information can be recovered from relatives.     

OPTIMIZATION MODELS,QUANTITATIVE GENETICS,AND MUTATION

The process of selection on a multivariate set of characters subject to functional constraints is considered from the points of view of both evolutionary optimization theory and quantitative genetics. Special attention is given to life-history characteristics. It is shown that, under suitable conditions (including weak selection), useful approximate formulas for the relations between the functional constraints and the additive genetic variance-covariance matrix can be derived. These can be used to show that the conditions for equilibrium under selection according to the two different approaches are approximately equivalent. Although large negative genetic correlations are to be expected between some pairs of life-history traits in populations at equilibrium under selection, in general some small negative genetic correlations and some positive genetic correlations will also be present. Thus, the observation of a positive genetic correlation between a pair of life-history traits does not necessarily refute the possibility of trade-offs among a multivariate set of traits that contains the pair in question. The relation between the pattern of functional constraints and the genetic correlations is often complex, and little insight into the former can be derived from the latter. The effects of mutations that lower the overall efficiency of resource utilization, thereby creating a positive component to the genetic covariances among life-history traits, are also considered for a specific model. Although such mutations can have a substantial effect on the form of the life history, extreme conditions seem to be needed for them to produce a large effect on the pattern of genetic correlations in a random-mating population. They can, however, cause the appearance of positive correlations following inbreeding, due to the exposure of deleterious recessive or partially recessive mutations. The analysis also suggests that the population means of individual components of a constrained multivariate system may often equilibrate at values that are far from the optima that would be attained if they were selected in isolation from the other members of the system.