Twenty years ago, Bulmer and Bull suggested that disruptive selection, produced by environmental fluctuations, can result in an evolutionary transition from environmental sex determination (ESD) to genetic sex determination (GSD). We investigated the feasibility of such a process, using mutation-limited adaptive dynamics and individual-based computer simulations. Our model describes the evolution of a reaction norm for sex determination in a metapopulation setting with partial migration and variation in an environmental variable both within and between local patches. The reaction norm represents the probability of becoming a female as a function of environmental state and was modeled as a sigmoid function with two parameters, one giving the location (i.e., the value of the environmental variable for which an individual has equal chance of becoming either sex) and the other giving the slope of the reaction norm for that environment. The slope can be interpreted as being set by the level of developmental noise in morph determination, with less noise giving a steeper slope and a more switchlike reaction norm. We found convergence stable reaction norms with intermediate to large amounts of developmental noise for conditions characterized by low migration rates, small differential competitive advantages between the sexes over environments, and little variation between individual environments within patches compared to variation between patches. We also considered reaction norms with the slope parameter constrained to a high value, corresponding to little developmental noise. For these we found evolutionary branching in the location parameter and a transition from ESD toward GSD, analogous to the original analysis by Bulmer and Bull. Further evolutionary change, including dominance evolution, produced a polymorphism acting as a GSD system with heterogamety. Our results point to the role of developmental noise in the evolution of sex determination.

Accumulation of deleterious mutations has important consequences for the evolution of mating systems and the persistence of small populations. It is well established that consanguineous mating can purge a part of the mutation load and that lethal mutations can also be purged in small populations. However, the efficiency of purging in natural populations, due to either consanguineous mating or to reduced population size, has been questioned. Consequences of consanguineous mating systems and small population size are often equated under “inbreeding” because both increase homozygosity, and selection is though to be more efficient against homozygous deleterious alleles. I show that two processes of purging that I call “purging by drift” and “purging by nonrandom mating” have to be distinguished. Conditions under which the two ways of purging are effective are derived. Nonrandom mating can purge deleterious mutations regardless of their dominance level, whereas only highly recessive mutations can be purged by drift. Both types of purging are limited by population size, and sharp thresholds separate domains where purging is either effective or not. The limitations derived here on the efficiency of purging are compatible with some experimental studies. Implications of these results for conservation and evolution of mating systems are discussed.

The floras of the Mediterranean-climate areas of southern Africa and southwestern Australia are remarkably species rich. Because the two areas are at similar latitudes and in similar positions on their respective continents, they have probably had similar Cenozoic climatic histories. Here we test the prediction that the evolution of the species richness in the two areas followed a similar temporal progression by comparing the rates of lineage accumulation for African and Australian Restionaceae. Restionaceae (Poales) are typical and often dominant elements in the fynbos vegetation of the Cape Floristic Region of southern Africa and the kwongan vegetation of the Southwestern Floristic Province of Western Australia. The phylogeny of the family was estimated from combined datasets for rbcL and trnL-F sequences and a large morphological dataset; these datasets are largely congruent. The monophyly of Restionaceae is supported and a basal division into an African clade (∼350 species) and an Australian clade (146 species) corroborated. There is also support for a futher subdivision of these two large sister-clades, but the terminal resolution within the African clade is very weak. Fossil pollen records provided a minimum age of the common ancestor of Australian and African Restionaceae as 64–71 million years ago, and this date was used to calibrate a molecular clock. A molecular clock was rejected by a likelihood ratio test; therefore, rate changes between the lineages were smoothed using nonparametric rate smoothing. The rate-corrected ages were used to construct a plot of lineages through time. During the Palaeogene the Australian lineage diversity increased consistent with the predictions of the constant birthrate model, while the African lineage diversity showed a dramatic increase in diversification rate in the Miocene. Incomplete sampling obscures the patterns in the Neogene, but extending the trends to the modern extant diversity suggests that this acceleration in the speciation rate continued in the African clade, whereas the Australian clade retained a constant diversification rate. The substantial morphological and anatomical similarity between the African and Australian Restionaceae appear to preclude morphological innovations as possible explanations for the intercontinental differences. Most likely these differences are due to the greater geographical extent and ecological variation in temperate Australia than temperate Africa, which might have provided refugia for basal Restionaceae lineages, whereas the more mountainous terrain of southern Africa might have provided the selective regimes for a more rapid, recent speciation.

To critically examine the relationship between species recognized by phylogenetic and reproductive compatibility criteria, we applied phylogenetic species recognition (PSR) to the fungus in which biological species recognition (BSR) has been most comprehensively applied, the well-studied genus Neurospora. Four independent anonymous nuclear loci were characterized and sequenced from 147 individuals that were representative of all described outbreeding species of Neurospora. We developed a consensus-tree approach that identified monophyletic genealogical groups that were concordantly supported by the majority of the loci, or were well supported by at least one locus but not contradicted by any other locus. We recognized a total of eight phylogenetic species, five of which corresponded with the five traditional biological species, and three of which were newly discovered. Not only were phylogenetic criteria superior to traditional reproductive compatibility criteria in revealing the full species diversity of Neurospora, but also significant phylogenetic subdivisions were detected within some species. Despite previous suggestions of hybridization between N. crassa and N. intermedia in nature, and the fact that several putative hybrid individuals were included in this study, no molecular evidence in support of recent interspecific gene flow or the existence of true hybrids was observed. The sequence data from the four loci were combined and used to clarify how the species discovered by PSR were related. Although species-level clades were strongly supported, the phylogenetic relationships among species remained difficult to resolve, perhaps due to conflicting signals resulting from differential lineage sorting.

We critically examined methods for recognizing species in the model filamentous fungal genus Neurospora by comparing traditional biological species recognition (BSR) with more comprehensive applications of both BSR and phylogenetic species recognition (PSR). Comprehensive BSR was applied to a set of 73 individuals by performing extensive crossing experiments and delineating biological species based on patterns of reproductive success. Within what were originally considered two species, N. crassa and N. intermedia, we recognized four reproductively isolated biological species. In a concurrent study (Dettman et al. 2003), we used genealogical concordance of four independent nuclear loci to recognize phylogenetic species in Neurospora. Overall, the groups of individuals identified as species were similar whether recognized by reproductive success or by phylogenetic criteria, and increased genetic distance between parents was associated with decreased reproductive success of crosses, suggesting that PSR using genealogical concordance can be used to reliably recognize species in organisms that are not candidates for BSR. In one case, two phylogenetic species were recognized as a single biological species, indicating that significant phylogenetic divergence preceded the development of reproductive isolation. However, multiple biological species were never recognized as a single phylogenetic species. Each of the putative N. crassa × N. intermedia hybrids included in this study was confidently assigned to a single species, using both PSR and BSR. As such, no evidence for a history of hybridization in nature among Neurospora species was observed. By performing reciprocal mating tests, we found that mating type, parental role, and species identity of parental individuals could all influence the reproductive success of matings. We also observed sympatry-associated sexual dysfunction in interspecific crosses, which was consistent with the existence of reinforcement mechanisms.

We compared pollen removal and deposition by hummingbirds and bumblebees visiting bird-syndrome Penstemon barbatus and bee-syndrome P. strictus flowers. One model for evolutionary shifts from bee pollination to bird pollination has assumed that, mostly due to grooming, pollen on bee bodies quickly becomes unavailable for transfer to stigmas, whereas pollen on hummingbirds has greater carryover. Comparing bumblebees and hummingbirds seeking nectar in P. strictus, we confirmed that bees had a steeper pollen carryover curve than birds but, surprisingly, bees and birds removed similar amounts of pollen and had similar per-visit pollen transfer efficiencies. Comparing P. barbatus and P. strictus visited by hummingbirds, the bird-syndrome flowers had more pollen removed, more pollen deposited, and a higher transfer efficiency than the bee-syndrome flowers. In addition, P. barbatus flowers have evolved such that their anthers and stigmas would not easily come into contact with bumblebees if they were to forage on them. We discuss the role that differences in pollination efficiency between bees and hummingbirds may have played in the repeated evolution of hummingbird pollination in Penstemon.

Drier periods from the late Pleistocene and early Holocene have been hypothesized to have caused the disappearance of various rainforest species over large geographical areas in South America and restricted the extant populations to mesic sites. Subsequent improvement in climatic conditions has been associated with recolonization. Changes in population size associated with these extinction-recolonization events should have affected genetic diversity within species. However, these historical hypotheses and their genetic consequences have rarely been tested in South America. Here, we examine the diversity of the chloroplast and nuclear genomes in a Neotropical rainforest tree species, Vouacapoua americana (Leguminosae, Caesalpinioideae) in French Guiana. The chloroplast diversity was analyzed using a polymerase chain reaction–restriction fragment length polymorphism method (six pairs of primers) in 29 populations distributed over most of French Guiana, and a subset of 17 populations was also analyzed at nine polymorphic microsatellite loci. To determine whether this species has experienced extinction-recolonization, we sampled populations in areas supposedly not or only slightly affected by climatic changes, where the populations would not have experienced frequent extinction, and in areas that appear to have been recently recolonized. In the putatively recolonized areas, we found patches of several thousands of hectares homogeneous for chloroplast variation that can be interpreted as the effect of recolonization processes from several geographical origins. In addition, we observed that, for both chloroplast and nuclear genomes, the populations in newly recolonized areas exhibited a significantly smaller allelic richness than others. Controlling for geographic distance, we also detected a significant correlation between chloroplast and nuclear population differentiation. This result indicates a cytonuclear disequilibrium that can be interpreted as a historical signal of a genetic divergence between fragmented populations. In conclusion, the spatial genetic structure of contemporary V. americana populations shows evidence that this species has experienced large extinction-recolonization events, which were possibly caused by past climatic change.

Reproduction has classically been viewed as a predominantly cooperative process. However, over the last 20 years this concept has steadily yielded ground to one of continual conflict in which the interests of the sexes are typically discordant. Within this framework, males and females are seen to be locked into a perpetual arms race, each adaptation by one sex promoting the evolution of countermeasures in the other sex. However, under strict genetic monogamy, the interests of the sexes become congruent, and hence antagonistic coevolution does not occur. We subjected the fly Sepsis cynipsea, a species with conspicuous sexual conflict, to experimentally enforced monogamy or polyandry for 29 generations and evaluated the microevolutionary consequences. We found that there were longevity costs to females consistent with sexually antagonistic coevolution. However, our measure of female fitness, offspring emergence, did not differ between treatments, even though life-history characters such as fertility and fecundity did. Results are discussed in terms of costs and benefits of sexual selection and sexual conflict.

Phenotypic plasticity of wing size and shape of Drosophila simulans was analyzed across the entire range of viable developmental temperatures with Procrustes geometric morphometric method. In agreement with previous studies, size clearly decreases when temperature increases. Wing shape variation was decomposed into its allometric (24%) and nonallometric (76%) components, and both were shown to involve landmarks located throughout the entire wing blade. The allometric component basically revealed a progressive, monotonous variation along the temperature. Surprisingly, nonallometric shape changes were highly similar at both extremes of the thermal range, suggesting that stress, rather than temperature per se, is the key developmental factor affecting wing shape.

Evolution Canyon in Lower Nahal Oren, Mount Carmel, Israel has been identified as a location promoting sympatric speciation. Several previous studies on Drosophila melanogaster populations from the two disparate slopes of the canyon suggest that these two populations are experiencing incipient speciation. However, recent microsatellite data did not reveal the expected level of population differentiation. Given the importance of this system for studying speciation, we set out to test two predictions of the incipient speciation hypothesis—genetic differentiation and sexual isolation. We sequenced six different Acp genes from isofemale lines from the south-facing slope (11 lines) and north-facing slope (nine lines) of Evolution Canyon. We found no evidence of genetic differentiation between the two slopes (F_ST = −0.03). We also conducted mate choice tests, using intraslope F₁ hybrids between different isofemale lines. There was no significant departure from random mating in mixtures of flies from the two slopes. Our results provide further indication that it is unlikely these two populations are experiencing incipient speciation. We discuss our results in light of the discrepancies that have been published on this enigmatic D. melanogaster system from Evolution Canyon.

Several laboratory studies of malaria parasites (Plasmodium sp.) and some field observations suggest that parasite virulence, defined as the harm a parasite causes to its vertebrate host, is positively correlated with transmission. Given this advantage, what limits the continual evolution of higher parasite virulence? One possibility is that while more virulent strains are more infectious, they are also more lethal to mosquitoes. In this study, we tested whether the virulence of the rodent malaria parasite P. chabaudi in the laboratory mouse was correlated with the fitness of mosquitoes it subsequently infected. Mice were infected with one of seven genetically distinct clones of P. chabaudi that differ in virulence. Weight loss and anemia in infected mice were monitored for 16–17 days before Anopheles stephensi mosquitoes were allowed to take a blood meal from them. Infection virulence in mice was positively correlated with transmission to mosquitoes (infection rate) and weakly associated with parasite burden (number of oocysts). Mosquito survival fell with increasing oocyst burden, but there was no overall statistically significant relationship between virulence in mice and mosquito mortality. Thus, there was no evidence that more virulent strains are more lethal to mosquitoes. Both vector survival and fecundity depended on parasite clone, and contrary to expectations, mosquitoes fed on infections more virulent to mice were more fecund. The strong parasite genetic effects associated with both fecundity and survival suggests that vector fitness could be an important selective agent shaping malaria population genetics and the evolution of phenotypes such as virulence in the vector.

Termites (Isoptera) comprise a large and important group of eusocial insects, yet, in contrast to the eusocial Hymenoptera (ants, bees, wasps), the breeding systems of termites remain poorly understood. In this study, I inferred the breeding system of the subterranean termite Reticulitermes flavipes based on colony and population genetic structure as determined from microsatellite and mitochondrial DNA markers. Termites were sampled from natural wood debris from three undisturbed, forested sites in central North Carolina. In each site, two transects separated by 1 km were sampled at approximately 15-m intervals. A total of 1272 workers collected from 57 collection points were genotyped at six microsatellite loci, and mitochondrial DNA haplotype was determined for a subset of these individuals using either restriction fragment length polymorphism or sequence variation in the AT-rich region. Colonies appeared to be localized: workers from the 57 collection points represented 56 genetically distinct colonies with only a single colony occupying two collection points located 15 m apart. Genetic analysis of family structure and comparisons of estimates of F-statistics (F_IT, F_IC, F_CT) and coefficients of relatedness (r) among nestmate workers with results of computer simulations of potential breeding systems suggested that 77% of all colonies were simple families headed by outbred monogamous pairs, whereas the remaining colonies were extended (inbred) families headed by low numbers of neotenics (about two females and one male) who were the direct offspring of the colony founders. There was no detectable isolation by distance among colonies along transects, suggesting that colony reproduction by budding is not common and that dispersal of reproductives during mating flights is not limited over this distance. Higher-level analysis of the microsatellite loci indicated weak but significant differentiation among sites (F_ST = 0.06), a distance of 16–38 km, and between transects within sites (F_ST = 0.06), a distance of 1 km. No significant differentiation at either the transect or site level was detected in the mitochondrial DNA sequence data. These results indicate that the study populations of R. flavipes have a breeding system characterized by monogamous pairs of outbred reproductives and relatively low levels of inbreeding because most colonies do not live long enough to produce neotenics, and those colonies that do generate neotenics contain an effectively small number of them.

Calcrete aquifers in arid inland Australia have recently been found to contain the world's most diverse assemblage of subterranean diving beetles (Coleoptera: Dytiscidae). In this study we test whether the adaptive shift hypothesis (ASH) or the climatic relict hypothesis (CRH) is the most likely mode of evolution for the Australian subterranean diving beetles by using a phylogeny based on two sequenced fragments of mitochondrial genes (CO1 and 16S-tRNA-ND1) and linearized using a relaxed molecular clock method. Most individual calcrete aquifers contain an assemblage of diving beetle species of distantly related lineages and/or a single pair of sister species that significantly differ in size and morphology. Evolutionary transitions from surface to subterranean life took place in a relatively small time frame between nine and four million years ago. Most of the variation in divergence times of the sympatric sister species is explained by the variation in latitude of the localities, which correlates with the onset of aridity from the north to the south and with an aridity maximum in the Early Pliocene (five mya). We conclude that individual calcrete aquifers were colonized by several distantly related diving beetle lineages. Several lines of evidence from molecular clock analyses support the CRH, indicating that all evolutionary transitions took place during the Late Miocene and Early Pliocene as a result of aridification.

Long-term geographic isolation can result in reproductive incompatibilities due to forces such as mutation, genetic drift, and differential selection. In the Sonoran topminnow, molecular genetic studies of mtDNA, microsatellites, and MHC genes have shown that the endangered Gila and Yaqui topminnows are substantially different, suggesting that divergence took place approximately two million years ago. Here we examined hybrid crosses and backcrosses between these two allopatric taxa to evaluate the accumulation of postmating barriers to reproduction. These results are then compared with results from a previous study where male topminnows were shown to mate assortatively with conspecific females. Despite their preference for conspecific mates, both types of interspecific crosses successfully produced offspring. There was evidence of reduced hybrid fitness, including smaller mean brood size and male-biased sex ratio, for some classes of backcrosses. Brood sizes and interbrood intervals varied significantly when hybrids were subdivided into different cross categories. Our results illustrate the importance of distinctly defining hybrid classes in studies of reproductive isolation. To our knowledge, this is the first such detailed evolutionary analysis in endangered fish taxa.

How do species that interbreed become reproductively isolated? If hybrids are less fit than parental types, natural selection should promote reproductive isolation by favoring the evolution of premating mechanisms that prevent hybridization (a process termed reinforcement). Although reinforcement should generate a decline in hybridization over time, countervailing forces of gene flow and recombination are thought to preclude natural selection from enhancing and finalizing reproductive isolation. Here, I present recent estimates of hybridization frequency between two species of spadefoot toad, Spea multiplicata and S. bombifrons. I compare these recent measures of hybrid frequency with previously published estimates and show that hybridization between these species has declined precipitously over the past 27 years. Although previous studies suggest that reinforcement possibly accounts for this decline in hybrids over time, three alternative hypotheses also can explain the observed decrease in hybridization. First, if one of the two interacting species becomes rare, opportunities for and incidence of hybridization may decrease. Second, if one of the two interacting species is initially rare, hybridization may be initially common if the rare species has difficulty locating conspecific mates. Third, if hybrids are produced only in particular environments, hybrid frequency may decline if habitat changes result in loss of those environments that promote hybrid formation. I found no support for these three alternative explanations of the decline in hybrids. Instead, reinforcement appears to best account for the evolution of enhanced reproductive isolation between these species. Moreover, the finding that hybridization declined precipitously in only 27 years suggests that many systems that have undergone reinforcement may be overlooked because reproductive isolation between the interacting populations or species may already be complete.

The house finch (Carpodacus mexicanus) is a native songbird of western North America that was introduced to the eastern United States and Hawaiian Islands in historic times. As such, it provides an unusually good opportunity to test the ability of molecular markers to recover recent details of a known population history. To investigate this prospect, genetic variation in 172 individuals from 16 populations in the western and eastern United States, southeastern Canada, Hawaiian Islands, and Mexico, as well as genetic variation in the closely related purple finch (Carpodacus purpureus) and Cassin's finch (Carpodacus cassinii) was studied by a semi-automated fluorescence-labeled amplified fragment length polymorphism (AFLP) marker system. A total of 363 markers were generated, of which 258 (71.2%) were polymorphic among species, 166 (61.4%) polymorphic among house finch subspecies, and 157 (60.2%) polymorphic among populations within the frontalis subspecies complex. Heterozygosities and interpopulation divergences revealed by the analysis appeared relatively low at all taxonomic levels, but there are few similar studies in avian populations with which to compare results. Whereas the known population history predicts that both eastern and Hawaiian finches should have been derived from within western populations, tree analysis using both populations and individuals as units suggests weak monophyly of eastern populations and indicates that Hawaiian populations are not clearly derived from California populations. However, the genetic distinctiveness of native and recently founded populations was disclosed by analyses of molecular variance as well as by a model-based assignment approach in which 98%, 94%, and 99% individuals from western, Hawaiian, and eastern regions, respectively, were assigned correctly to their populations without using prior information on population of origin, suggesting that these recent introductions have resulted in detectable differentiation without substantial loss of AFLP diversity. Our results indicate that AFLPs are a useful tool for population genetic and evolutionary studies of birds, particularly as a prelude to finding molecular markers linked to traits subjected to recent adaptive evolution.

We measured genetic differentiation among species of large white-headed gulls using mitochondrial (cytochrome b haplotypes) and nuclear (microsatellites) markers. Additional information was added using a previously published study of allozymes on the same species. Levels of differentiation among species at nuclear markers are much lower than would be expected for avian species and are not concordant with the level of differentiation in mitochondrial markers. This discrepancy is best explained by a combination of recent species origin and interspecific gene flow after speciation. The data also suggest that female-mediated gene flow is reduced compared to male-mediated gene flow, either due to behavioral bias or due to stronger counterselection of female hybrids in accordance with Haldane's rule for ZW species. Whatever the reasons for the low differentiation of the species' nuclear gene pools, the extensive similarity of their nuclear genome demonstrates that selection on a limited number of characters is an important factor in establishing and maintaining clear-cut phenotypic differences between these species and suggests that the number of loci involved in this process is quite low. This situation may not be exceptional in birds, indeed a number of studies have found similarly low level of differentiation in nuclear markers among congeneric bird species, although usually based on a single set of markers. Because hybridization is a widespread phenomenon in birds, many of these cases might be due to interspecific gene flow.

Van Tienderen recently published a method that links selection gradients between a phenotypic trait and multiple fitness components with the effects of these fitness components on the population growth rate (mean absolute fitness). The method allows selection to be simultaneously estimated across multiple fitness components in a population dynamic framework. In this paper we apply the method to a population of red deer living in the North Block of the Isle of Rum, Scotland. We show that (1) selection on birth date and birth weight can operate through multiple fitness components simultaneously; (2) our estimates of the response to selection are consistent with the observed change in trait values that we cannot explain with environmental and phenotypic covariates; (3) selection on both traits has fluctuated over the course of the study; (4) selection operates through different fitness components in different years; and (5) no environmental covariates correlate with selection because different fitness components respond to density and climatic variation in contrasting ways.

In this study, we compare rDNA substitution rates for a group of closely related plant species in the western Pacific that exist in different biomes. The results of this comparison indicate higher rates of substitution for species living in habitats with greater biologically available energy. We interpret that finding as potentially important in understanding evolution because of its implication that substitution rate may be a function of biologically available energy and its correlate, productivity. The relevance of this research is twofold. First, contrasting closely related species across different biomes allows for a comparison between rates of molecular evolution across different energetic/productivity regimes while controlling for phylogenetically influenced variation. Second, the research indicates some of the design parameters for future studies that are required to explore the importance of this relationship among different groups of related organisms. If higher rates of molecular evolution where there is greater available energy are found to be widespread this might bring an additional dimension to the understanding of macroevolutionary pattern and process.

Many microbial pathogens can switch to new hosts or adopt alternative transmission routes as environmental conditions change, displaying unexpected flexibility in their infection pathways and often causing emerging diseases. In contrast, parasitic worms that must develop through a fixed series of host species appear less likely to show phenotypic plasticity in their transmission pathways. Here, I demonstrate experimentally that a trematode parasite, Coitocaecum parvum, can accelerate its development and rapidly reach precocious maturity in its crustacean intermediate host in the absence of chemical cues emanating from its fish definitive host. Juvenile trematodes can also mature precociously when the mortality rate of their intermediate hosts is increased. Eggs produced by precocious adults hatch into viable larvae, capable of pursuing the parasite's life cycle. In the absence of chemical cues from fish hosts, the size of eggs released by precocious trematodes in their intermediate hosts becomes more variable, possibly indicating a bet-hedging strategy. These results illustrate that parasitic worms with complex life cycles have development and transmission strategies that are more plastic than commonly believed, allowing them to skip one host in their cycle when they perceive limited opportunities for transmission.

Conspicuous coloration is often used in combination with chemical defenses to deter predators from attacking. Experimental studies have shown that the avoidance inducing effect of conspicuous prey coloration increases with increasing size of pattern elements and with increasing body size. Here we use a comparative approach to test the prediction from these findings, namely that conspicuous coloration will evolve in tandem with body size. In our analysis, we use a previously published mitochondrial DNA-based phylogeny and comparative analysis of independent contrasts to examine if evolutionary shifts in color pattern have been associated with evolutionary changes in body size in aposematic poison frogs (Anura: Dendrobatidae). Information on body size (snout to vent length) and coloration were obtained from the literature. Two different measures of conspicuousness were used, one based on rankings by human observers and the other based on computer analysis of digitized photographs. The results from comparative analyses using either measure of coloration indicated that avoidance inducing coloration and body size have evolved in concert in poison frogs. Results from reconstruction of character change further indicate that the correlated evolution of size and coloration has involved changes in both directions within each of the different clades of the phylogenetic tree. This finding is consistent with the hypothesis that selection imposed by visually guided predators has promoted the evolution of larger body size in species with conspicuous coloration, or enhanced evolution of conspicuous coloration in larger species.

Studies of inbreeding and interspecific hybridization are generally pursued separately with different metrics. There is a need to integrate them because they have the common goal of seeking an understanding of the genetic and ecological basis of fitness variation in populations. We use mean expected heterozygosity as an axis of variation on which to compare the fitness of inbreeding and hybridizing Darwin's finches (Geospiza scandens and G. fortis) relative to the fitness of matched outbred controls. We find that relative fitness of inbred finches is less than one in the 1991 cohorts of both species. Inbreeding depression is stronger in the species (G. scandens) with the lower genetically effective population size. Relative fitness of hybrids (backcrosses) in the same cohort of G. scandens is greater than one. Evidence of heterosis in G. fortis is mixed. Thus the two interbreeding species displayed somewhat different fitness patterns under the same set of environmental conditions. Hybridization may enhance fitness to different degrees by counteracting the effects of inbreeding depression, by other additive and nonadditive genetic effects, and by producing phenotypes well suited to exploit particular ecological conditions.