The evolution of premating isolation after secondary contact is primarily considered in the guise of reinforcement, which relies on low hybrid fitness as the driving force for mating preference divergence. Here I consider two additional forces that may play a substantial role in the adaptive evolution of premating isolation, direct selection on preferences and indirect selection against postmating, prezygotic incompatibilities. First, I argue that a combination of ecological character displacement and sensory bias can cause direct selection on preferences that results in the pattern of reproductive character displacement. Both analytical and numerical methods are then used to demonstrate that, as expected from work in single populations, such direct selection will easily overwhelm indirect selection due to low hybrid fitness as the primary determinant of preference evolution. Second, postmating, prezygotic incompatibilities are presented as a driving force in the evolution of premating isolation. Two classes of these mechanisms, those increasing female mortality after mating but before producing offspring and those reducing female fertility, are shown to be identical in their effects on preference divergence. Analytical and numerical techniques are then used to demonstrate that postmating, prezygotic factors may place strong selection on preference divergence. These selective forces are shown to be comparable if not greater than those produced by the low fitness of hybrids.

Corresponding Editor: M. Whitlock

The reasons that sex and recombination are so widespread remain elusive. One popular hypothesis is that sex and recombination promote adaptation to a changing environment. The strongest evidence that increased recombination may evolve because recombination promotes adaptation comes from artificially selected populations. Recombination rates have been found to increase as a correlated response to selection on traits unrelated to recombination in several artificial selection experiments and in a comparison of domesticated and nondomesticated mammals. There are, however, several alternative explanations for the increase in recombination in such populations, including two different evolutionary explanations. The first is that the form of selection is epistatic, generating linkage disequilibria among selected loci, which can indirectly favor modifier alleles that increase recombination. The second is that random genetic drift in selected populations tends to generate disequilibria such that beneficial alleles are often found in different individuals; modifier alleles that increase recombination can bring together such favorable alleles and thus may be found in individuals with greater fitness. In this paper, we compare the evolutionary forces acting on recombination in finite populations subject to strong selection. To our surprise, we found that drift accounted for the majority of selection for increased recombination observed in simulations of small to moderately large populations, suggesting that, unless selected populations are large, epistasis plays a secondary role in the evolution of recombination.

Corresponding Editor: A. Caballero

Both chromosomal rearrangements and negative interactions among loci (Dobzhansky-Muller incompatibilities) have been advanced as the genetic mechanism underlying the sterility of interspecific hybrids. These alternatives invoke very different evolutionary histories during speciation and also predict different patterns of sterility in artificial hybrids. Chromosomal rearrangements require drift, inbreeding, or other special conditions for initial fixation and, because heterozygosity per se generates any problems with gamete formation, F₁ hybrids will be most infertile. In contrast, Dobzhansky-Muller incompatibilities may arise as byproducts of adaptive evolution and often affect the segregating F₂ generation most severely. To distinguish the effects of these two mechanisms early in divergence, we investigated the quantitative genetics of hybrid sterility in a line cross between two members of the Mimulus guttatus species complex (M. guttatus and M. nasutus). Hybrids showed partial male and female sterility, and the patterns of infertility were not consistent with the action of chromosomal rearrangements alone. F₂ and F₁ hybrids exhibited equal decreases in pollen viability (> 40%) relative to the highly fertile parental lines. A large excess of completely pollen-sterile F₂ genotypes also pointed to the segregation of Dobzhansky-Muller incompatibility factors affecting male fertility. Female fertility showed a pattern similarly consistent with epistatic interactions: F₂ hybrids produced far fewer seeds per flower than F₁ hybrids (88.0 ± 2.8 vs. 162.9 ± 8.5 SE, respectively) and either parental line, and many F₂ genotypes were completely female sterile. Dobzhansky-Muller interactions also resulted in the breakdown of several nonreproductive characters and appear to contribute to correlations between male and female fertility in the F₂ generation. These results parallel and contrast with the genetics of postzygotic isolation in model animal systems and are a first step toward understanding the process of speciation in this well-studied group of flowering plants.

Corresponding Editor: S. Tonsor

The Mediterranean species complex of Senecio serves to illustrate evolutionary processes that are likely to confound phylogenetic inference, including rapid diversification, gene tree-species tree discordance, reticulation, interlocus concerted evolution, and lack of complete lineage sorting. Phylogeographic patterns of chloroplast DNA (cpDNA) haplotype variation were studied by sampling 156 populations (502 individuals) across 18 species of the complex, and a species phylogeny was reconstructed based on sequences from the internal transcribed spacer (ITS) regions of nuclear ribosomal DNA. For a subset of species, randomly amplified polymorphic DNAs (RAPDs) provided reference points for comparison with the cpDNA and ITS datasets. Two classes of cpDNA haplotypes were identified, with each predominating in certain parts of the Mediterranean region. However, with the exception of S. gallicus, intraspecific phylogeographic structure is limited, and only a few haplotypes detected were species-specific. Nuclear sequence divergence is low, and several unresolved phylogenetic groupings are suggestive of near simultaneous diversification. Two well-supported ITS clades contain the majority of species, amongst which there is a pronounced sharing of cpDNA haplotypes. Our data are not capable of diagnosing the relative impact of reticulation versus insufficient lineage sorting for the entire complex. However, there is firm evidence that S. flavus subsp. breviflorus and S. rupestris have acquired cpDNA haplotypes and ITS sequences from co-occurring species by reticulation. In contrast, insufficient lineage sorting is a viable hypothesis for cpDNA haplotypes shared between S. gallicus and its close relatives. We estimated the minimum coalescent times for these haplotypes by utilizing the inferred species phylogeny and associated divergence times. Our data suggest that ancestral cpDNA polymorphisms may have survived for ca. 0.4–1.0 million years, depending on molecular clock calibrations.

Corresponding Editor: D. Baum

We address how a conflict between pollinator attraction and avoidance of flower predation influences the evolution of flower shape in Polemonium viscosum. Flower shape in P. viscosum is the product of an isometric relationship between genetically correlated (r_A = 0.70) corolla flare and length. Bumblebee pollinators preferentially visit flowers that are more flared and have longer tubes, selecting for a funnel-shaped corolla. However, flower shape also influences nectar-foraging ants that sever the style at its point of attachment to the ovary. Surveys of ant damage show that plants having flowers with flared, short corollas are most vulnerable to ant predation. Consistent with this result, the ratio of corolla length to flare is significantly greater in a krummholz (high predation risk) population than in a tundra (low predation risk) population. To explicitly test whether the evolution of a better defended flower would exact a cost in pollination, we created tubular flowers by constricting the corolla during development. Performance of tubular flowers and natural controls was compared for defensive and attractive functions. In choice trials, ants entered control flowers significantly more often than tubular ones, confirming that the evolution of tubular flowers would reduce the risk of predation. However, in a bumblebee-pollinated population, tubular flowers received significantly less pollen and set fewer seeds than controls. A fitness model incorporating these data predicts that in the absence of the genetic correlation between corolla length and flare, intermittent selection for defense could allow tubular flowers to spread in the krummholz population. However, in the tundra, where bumblebees account for nearly all pollination, the model predicts that tubular flowers should always confer a fitness disadvantage.

Corresponding Editor: J. Conner

In reciprocal transplant experiments, Bertness and Gaines (1993) found that Semibalanus balanoides juveniles that had settled in an upper Narragansett Bay estuary survived better in that estuary that did juveniles from coastal localities. The observed pattern of survivorship led to the claim that local adaptation may result from a combination of limited gene flow between and strong selection within these habitats. Here we test the hypothesis that limited gene flow has led to habitat-specific population differentiation using sequence and restriction fragment length polymorphism analyses of the mitochondrial DNA D-loop region of S. balanoides. Samples were analyzed from replicated coastal and estuary localities in both Narragansett Bay, Rhode Island, and Damariscotta River, Maine. The patterns of F_ST indicate that gene flow between coast and estuary is extensive (Nm > 100) and is not lower in the estuary with lower flushing rates (Narragansett Bay). Given the high estimate of genetic exchange, adaptations for unpredictable environments seem more likely than local adaptation in this species because loci that respond to selection in one generation are essentially homogenized by the next seasons' settlement. Nevertheless, these estimates of neutral gene flow can help identify the strength of selection necessary for local adaptation to accumulate in Semibalanus.

Corresponding Editor: J. Neigel

We investigate the nature and duration of incompatibility between certain combinations of Acromyrmex leaf-cutting ants and symbiotic fungi, taken from sympatric colonies of the same or a related species. Ant-fungus incompatibility appeared to be largely independent of the ant species involved, but could be explained partly by genetic differences among the fungus cultivars. Following current theoretical considerations, we develop a hypothesis, originally proposed by S. A. Frank, that the observed incompatibilities are ultimately due to competitive interactions between genetically different fungal lineages, and we predict that the ants should have evolved mechanisms to prevent such competition between cultivars within a single garden. This requires that the ants are able to recognize unfamiliar fungi, and we show that this is indeed the case. Amplified fragment length polymorphism genotyping further shows that the two sympatric Acromyrmex species share each other's major lineages of cultivar, confirming that horizontal transfer does occasionally take place. We argue and provide some evidence that chemical substances produced by the fungus garden may mediate recognition of alien fungi by the ants. We show that incompatibility between ants and transplanted, genetically different cultivars is indeed due to active killing of the novel cultivar by the ants. This incompatibility disappears when ants are force-fed the novel cultivar for about a week, a result that is consistent with our hypothesis of recognition induced by the resident fungus and eventual replacement of incompatibility compounds during force-feeding.

Corresponding Editor: K. Ross

Genetic variance-covariance structures (G), describing genetic constraints on microevolutionary changes of populations, have a central role in the current theories of life-history evolution. However, the evolution of Gs in natural environments has been poorly documented. Resource quality and quantity for many animals and plants vary seasonally, which may shape genetic architectures of their life histories. In the mountain birch-insect herbivore community, leaf quality of birch for insect herbivores declines profoundly during both leaf growth and senescence, but remains stable during midsummer. Using six sawfly species specialized on the mountain birch foliage, we tested the ways in which the seasonal variation in foliage quality of birch is related to the genetic architectures of larval development time and body size. In the species consuming mature birch leaves of stable quality, that is, without diet-imposed time constraints for development time, long development led to high body mass. This was revealed by the strongly positive phenotypic and genetic correlations between the traits. In the species consuming growing or senescing leaves, on the other hand, the rapidly deteriorating leaf quality prevented the larvae from gaining high body mass after long development. In these species, the phenotypic and genetic correlations between development time and final mass were negative or zero. In the early-summer species with strong selection for rapid development, genetic variation in development time was low. These results show that the intuitively obvious positive genetic relationship between development time and final body mass is a probable outcome only when the constraints for long development are relaxed. Our study provides the first example of a modification in guild-wide patterns in the genetic architectures brought about by seasonal variation in resource quality.

Corresponding Editor: D. Roff

The likelihood of sympatric speciation is enhanced when assortative mating is a by-product of adaptation to different habitats. Pleiotropy of this kind is recognized as important in parasites that use their hosts as a long-range cue for finding mates, but is generally assumed to have limited applicability for most other organisms. In the larch budmoth, Zeiraphera diniana (Lepidoptera: Tortricidae), sympatric host races feed on larch or pine. Zeiraphera diniana females attract males (call) by releasing host-independent long-range pheromones. Pheromone composition differs strongly between host races, but we show in an experimental field study that cross-attraction can occur at a rate of 0.03–0.38. Cross-attraction to larch females increases when they call from neighborhoods (8-m radius) rich in pine or from pine trees. Cross-attraction to pine females similarly increases when calling from neighborhoods rich in larch, but there is no significant effect of calling substrate. Males, as well as females, of this species preferentially alight on their own host, and in neighborhoods where their own host is common. This effect of tree species and host neighborhood on assortative mating is therefore due, at least in part, to the numbers of males of each host race present within approximately 200 m² surrounding the female. This proximity effect is enhanced by the clumped distributions of the hosts themselves. Host chemistry might also affect pheromone production and/or response directly, but we have evidence neither for nor against this. This work provides empirical evidence that host adaptation has a pleiotropic effect on assortative mating in a species with host-independent long-range mating signals. Sympatric speciation via pleiotropy between ecological traits and assortative mating may thus be more common than generally supposed: Clumped resource distributions and habitat choice by adults are widespread.

Corresponding Editor: S. Strauss

Beetles in the weevil subfamilies Scolytinae and Platypodinae are unusual in that they burrow as adults inside trees for feeding and oviposition. Some of these beetles are known as ambrosia beetles for their obligate mutualisms with asexual fungi—known as ambrosia fungi—that are derived from plant pathogens in the ascomycete group known as the ophiostomatoid fungi. Other beetles in these subfamilies are known as bark beetles and are associated with free-living, pathogenic ophiostomatoid fungi that facilitate beetle attack of phloem of trees with resin defenses. Using DNA sequences from six genes, including both copies of the nuclear gene encoding enolase, we performed a molecular phylogenetic study of bark and ambrosia beetles across these two subfamilies to establish the rate and direction of changes in life histories and their consequences for diversification. The ambrosia beetle habits have evolved repeatedly and are unreversed. The subfamily Platypodinae is derived from within the Scolytinae, near the tribe Scolytini. Comparison of the molecular branch lengths of ambrosia beetles and ambrosia fungi reveals a strong correlation, which a fungal molecular clock suggests spans 60 to 21 million years. Bark beetles have shifted from ancestral association with conifers to angiosperms and back again several times. Each shift to angiosperms is associated with elevated diversity, whereas the reverse shifts to conifers are associated with lowered diversity. The unusual habit of adult burrowing likely facilitated the diversification of these beetle-fungus associations, enabling them to use the biomass-rich resource that trees represent and set the stage for at least one origin of eusociality.

Corresponding Editor: S. Pitnick

The extent and impact of introgressive hybridization was examined in the Gila robusta complex of cyprinid fishes using mitochondrial DNA (mtDNA) sequence variation. Lower Colorado River basin populations of G. robusta, G. elegans, and G. cypha exhibited distinct mtDNAs, with only limited introgression of G. elegans into G. cypha. The impact of hybridization was significant in upper Colorado River basin populations; most upper basin fishes sampled exhibited only G. cypha mtDNA haplotypes, with some individuals exhibiting mtDNA from G. elegans. The complete absence of G. robusta mtDNA, even in populations of morphologically pure G. robusta, indicates extensive introgression that predates human influence. Analysis of the geographic distribution of variation identified two distinctive G. elegans lineages; however, the small number of individuals and localities sampled precluded a comprehensive analysis. Analysis of haplotype and population networks for G. cypha mtDNAs from 15 localities revealed low divergence among haplotypes; however, significant frequency differences among populations within and among drainages were found, largely attributable to samples in the Little Colorado River region. This structure was not associated with G. cypha and G. robusta, as morphotypes from the same location are more similar than conspecific forms in other locations. This indicates that morphological and mtDNA variation are affected by different evolutionary forces in Colorado River Gila and illustrates how both hybridization and local adaptation can play important roles in evolution.

Corresponding Editor: J. Neigel

When selection acts on social or behavioral traits, the fitness of an individual depends on the phenotypes of its competitors. Here, we describe methods and statistical inference for measuring natural selection in small social groups. We measured selection on throat color alleles that arises from microgeographic variation in allele frequency at natal sites of side–blotched lizards (Uta stansburiana). Previous game-theoretic analysis indicates that two color morphs of female side–blotched lizards are engaged in an offspring quantity–quality game that promotes a density- and frequency–dependent cycle. Orange–throated females are r-strategists. They lay large clutches of small progeny, which have poor survival at high density, but good survival at low density. In contrast, yellow–throated females are K-strategists. They lay small clutches of large progeny, which have good survival at high density. We tested three predictions of the female game: (1) orange progeny should have a fitness advantage at low density; (2) correlational selection acts to couple color alleles and progeny size; and (3) this correlational selection arises from frequency-dependent selection in which large hatchling size confers an advantage, but only when yellow alleles are rare. We also confirmed the heritability of color, and therefore its genetic basis, by producing progeny from controlled matings. A parsimonious cause of the high heritability is that three alleles (o, b, y) segregate as one genetic factor. We review the physiology of color formation to explain the possible genetic architecture of the throat color trait. Heritability of color was nearly additive in our breeding study, allowing us to compute a genotypic value for each individual and thus predict the frequency of progeny alleles released on 116 plots. Rather than study the fitness of individual progeny, we studied how the fitness of their color alleles varied with allele frequency on plots. We confirmed prediction 1: When orange alleles are present in female progeny, they have higher fitness at low density when compared to other alleles. Even though the difference in egg size of the female morphs was small (0.02 g), it led to knife–edged survival effects for their progeny depending on local social context. Selection on hatchling survival was not only dependent on color alleles, but on a fitness interaction between color alleles and hatchling size, which confirmed prediction 2. Sire effects, which are not confounded by maternal phenotype, allowed us to resolve the frequency dependence of correlational selection on egg size and color alleles and thereby confirmed prediction 3. Selection favored large size when yellow sire alleles were rare, but small size when they were common. Correlational selection promotes the formation of a self-reinforcing genetic correlation between the morphs and life-history variation, which causes selection in the next density and frequency cycle to be exacerbated. We discuss general conditions for the evolution of self-reinforcing genetic correlations that arise from social selection associated with frequency–dependent sexual and natural selection.

Corresponding Editor: J. Losos

We compared reproductive allocation and variation in condition and survivorship of two heritable female throat color morphs (orange and yellow) in a free-living population of side-blotched lizards (Uta stansburiana). Using path analysis and structural equation modeling, we investigated how variation in the social environment affected clutch size and egg mass and two condition traits (postlaying mass, immunological condition) and how these traits in turn affected female field survival. In the presence of many neighbors, both morphs increased their clutch sizes, although these effects were only significant in yellow females. In addition, yellow females increased their egg mass in the presence of many orange neighbors. Orange females surrounded by many orange neighbors showed sign of stress in the form of immunosuppression, whereas this effect was less pronounced in yellow females. The morphs also differed in the impact of variation in clutch size and egg mass on both condition traits. Finally, female morphotype and immune responsiveness affected fitness interactively, and hence these two traits showed signs of fitness epistasis: Selection gradients on this trait were opposite in sign in the two morphs. The correlational selection gradient (γ_{throat×antibody response}) between female throat color and antibody responsiveness was −0.365. Our data thus reveal important interactive effects such as genotype-by-environment interaction toward the social environment and morph-specific trade-offs as well as the occurrence of correlational selection. We discuss the use of naturally occurring and conspicuous genetic polymorphisms in field studies of selection and life-history allocation.

Corresponding Editor: T. Garland Jr.

A previous study of the hybrid zone in western Panama between white-collared (Manacus candei) and golden-collared manakins (M. vitellinus) documented the unidirectional introgression of vitellinus male secondary sexual traits across the zone. Here, we examine the hybrid zone in greater genetic and morphological detail. Statistical comparisons of clines are performed using maximum-likelihood and nonparametric bootstrap methods. Our results demonstrate that an array of six molecular and two morphometric markers agree in cline position and width. Clines for male collar and belly color are similar in width to the first eight clines, but are shifted in position by at least five cline widths. The result is that birds in intervening populations are genetically and morphometrically very like parental candei, but males have the plumage color of parental vitellinus. Neither neutral diffusion nor nonlinearity of color scales appear to be viable explanations for the large cline shifts. Genetic dominance of vitellinus plumage traits is another potential explanation that will require breeding experiments to test. Sexual selection remains a plausible explanation for the observed introgression of vitellinus color traits in these highly dimorphic, polygynous, lek-mating birds. Two other clines, including a nondiagnostic isozyme locus, are similar in position to the main cluster of clines, but are broader in width. Thus, introgression at some loci is greater than that detected with diagnostic markers. Assuming that narrow clines are maintained by selection, variation in cline width indicates that selection is not uniform throughout the genome and that diagnostic markers are under more intense selective pressure. The traditional focus on diagnostic markers in studies of hybrid zones may therefore lead to underestimates of average introgression. This effect may be more pronounced in organisms with low levels of genetic divergence between hybridizing taxa.

Corresponding Editor: H. L. Gibbs

Three major types of bilateral asymmetry (fluctuating asymmetry, directional asymmetry, and antisymmetry) have long been recognized in the literature. Little, however, is known about transitions between asymmetry types, especially in natural populations. It is often assumed that directional asymmetry and antisymmetry have a larger genetic basis than fluctuating asymmetry. This leads many scientists to exclude traits or populations showing either directional asymmetry or antisymmetry from developmental instability studies, focusing attention on fluctuating asymmetry alone. This procedure may bias the findings and thus our understanding of patterns of bilateral asymmetry and the factors influencing it. To examine changes in bilateral asymmetry across the distribution range of a species, I studied the length of the third toe in 11 chukar partridge (Alectoris chukar) populations across a steep environmental gradient of 320 km within the species' range in Israel. This trait was selected due to its adaptive value in the chukar, a species that spends much of its activity walking, and due to its high measurement repeatability. Moving from the core toward the very extreme periphery of the range, the following four trends are detected: (1) the expression of the directional asymmetry component significantly increases; (2) the frequency of symmetrical individuals in the population significantly decreases, with a sharp decline at the steepest part of the climatic and environmental gradient studied, within the Mediterranean-desert ecotone; (3) mean asymmetry levels, as estimated using the unsigned difference between the right and left toe, significantly increases; and (4) the range of asymmetry increases such that the most asymmetrical individuals originate from the very edge of the range. These findings provide primary evidence that substantial shifts in asymmetry may occur across short geographical distances within a species' distribution range. They show a continuum between asymmetry types and support the notion that all three types of asymmetry can reflect developmental instability. Further studies of developmental instability should be designed so that they enable detection of transitions between asymmetry types across natural populations. Such a procedure may partly resolve some of the contradictions seen in the literature regarding the relationship between bilateral asymmetry and environmental stress.

Corresponding Editor: H. L. Gibbs

The effects of mutation on phenotypic expression are supposed to be mainly deleterious because mutations disrupt the expression of genes that function relatively well under current environmental conditions. Thus, mutations are assumed to give rise to deviant phenotypes that are generally selected against. Radioactive contamination in the Chernobyl region of Ukraine is associated with a significant increase by a factor two to 10 in mutation rate in microsatellite markers of the barn swallow, Hirundo rustica. Barn swallows from Chernobyl had a temporally constant, elevated frequency of partial albinism compared to the situation before radioactive contamination and compared to birds from a control area. Albinism disproportionately affected the carotenoid-based plumage of the head, suggesting that carotenoid metabolism is particularly susceptible to the effects of radiation. Individuals with partially albinistic plumage had, on average, lower mean phenotypic values than other birds, and this was particularly the case for males. Furthermore, differences in phenotypic variation, as determined using Levene's test, were significantly larger in partial albinos compared to nonalbinos in males, but not in females, even though the null expectation would be the opposite due to the lower mean phenotypic values of partial albinos. Although small phenotypes were commonly associated with germline mutations, there was no general decrease in overall body size during the period 1991–2000, implying that small individuals were selected against. Because partial albinism is disfavored by natural selection, the effects of mutations are deleterious, giving rise to a balance between mutation and selection.

Corresponding Editor: S. Edwards

Genetic markers that differ in mode of inheritance and rate of evolution (a sex-linked Z-specific microsatellite locus, five biparentally inherited microsatellite loci, and maternally inherited mitochondrial [mtDNA] sequences) were used to evaluate the degree of spatial genetic structuring at macro- and microgeographic scales, among breeding regions and local nesting populations within each region, respectively, for a migratory sea duck species, the spectacled eider (Somateria fisheri). Disjunct and declining breeding populations coupled with sex-specific differences in seasonal migratory patterns and life history provide a series of hypotheses regarding rates and directionality of gene flow among breeding populations from the Indigirka River Delta, Russia, and the North Slope and Yukon-Kuskokwim Delta, Alaska. The degree of differentiation in mtDNA haplotype frequency among breeding regions and populations within regions was high (ϕ_CT = 0.189, P < 0.01; ϕ_SC = 0.059, P < 0.01, respectively). Eleven of 17 mtDNA haplotypes were restricted to a single breeding region. Genetic differences among regions were considerably lower for nuclear DNA loci (sex-linked: ϕ_ST = 0.001, P > 0.05; biparentally inherited microsatellites: mean θ = 0.001, P > 0.05) than was observed for mtDNA. Using models explicitly designed for uniparental and biparentally inherited genes, estimates of spatial divergence based on nuclear and mtDNA data together with elements of the species' breeding ecology were used to estimate effective population size and degree of male and female gene flow. Differences in the magnitude and spatial patterns of gene correlations for maternally inherited and nuclear genes revealed that females exhibit greater natal philopatry than do males. Estimates of generational female and male rates of gene flow among breeding regions differed markedly (3.67 × 10⁻⁴ and 1.28 × 10⁻², respectively). Effective population size for mtDNA was estimated to be at least three times lower than that for biparental genes (30,671 and 101,528, respectively). Large disparities in population sizes among breeding areas greatly reduces the proportion of total genetic variance captured by dispersal, which may accelerate rates of inbreeding (i.e., promote higher coancestries) within populations due to nonrandom pairing of males with females from the same breeding population.

Corresponding Editor: H. L. Gibbs

Parasite resistance and body size are subject to directional natural selection in a population of feral Soay sheep (Ovis aries) on the island of St. Kilda, Scotland. Classical evolutionary theory predicts that directional selection should erode additive genetic variation and favor the maintenance of alleles that have negative pleiotropic effects on other traits associated with fitness. Contrary to these predictions, in this study we show that there is considerable additive genetic variation for both parasite resistance, measured as fecal egg count (FEC), and body size, measured as weight and hindleg length, and that there are positive genetic correlations between parasite resistance and body size in both sexes. Body size traits had higher heritabilities than parasite resistance. This was not due to low levels of additive genetic variation for parasite resistance, but was a consequence of high levels of residual variance in FEC. Measured as coefficients of variation, levels of additive genetic variation for FEC were actually higher than for weight or hindleg length. High levels of additive genetic variation for parasite resistance may be maintained by a number of mechanisms including high mutational input, balancing selection, antagonistic pleiotropy, and host-parasite coevolution. The positive genetic correlation between parasite resistance and body size, a trait also subject to sexual selection in males, suggests that parasite resistance and growth are not traded off in Soay sheep, but rather that genetically resistant individuals also experience superior growth.

Corresponding Editor: B. Crespi

Lynch (1999) proposed a method for estimation of genetic correlations from phenotypic measurements of individuals for which no pedigree information is available. This method assumes that shared environmental effects do not contribute to the similarity of relatives, and it is expected to perform best when sample sizes are large, many individuals in the sample are paired with close relatives, and heritability of the traits is high. We tested the practicality of this method for field biologists by using it to estimate genetic correlations from measurements of field-caught waterstriders (Aquarius remigis). Results for sample sizes of less than 100 pairs were often unstable or undefined, and even with more than 500 pairs only half of those correlations that had been found to be significant in standard laboratory experiments were statistically significant in this study. Statistically removing the influence of environmental effects (shared between relatives) weakened the estimates, possibly by removing some of the genetic similarity between relatives. However, the method did generate statistically significant estimates for some genetic correlations. Lynch (1999) anticipated the problems found, and proposed another method that uses estimates of relatedness between members of pairs (from molecular marker data) to improve the estimates of genetic correlations, but that approach has yet to be tested in the field.

Corresponding Editor: K. Ross

I present a model demonstrating that, in social Hymenoptera, split sex allocation can influence the evolution of reproductive partitioning (skew). In a facultatively polygynous population (with one to several queens per colony), workers vary in their relative relatedness to females (relatedness asymmetry). Split sex-ratio theory predicts that workers in monogynous (single-queen) colonies should concentrate on female production, as their relatedness asymmetry is relatively high, whereas workers in the polygynous colonies should concentrate on male production, as their relatedness asymmetry is relatively low. By contrast, queens in all colonies value males more highly per capita than they value females, because the worker-controlled population sex ratio is too female-biased from the queens' standpoint. Consider a polygynous colony in a facultatively polygynous population of perennial, social Hymenoptera with split sex ratios. A mutant queen achieving reproductive monopoly would gain from increasing her share of offspring but, because the workers would assess her colony as monogynous, would lose from the workers rearing a greater proportion of less-valuable females from the colony's brood. This sets an upper limit on skew. Therefore, in social Hymenoptera, skew evolution is potentially affected by queen-worker conflict over sex allocation.

Corresponding Editor: K. Ross

No Abstract Available

Book Review Editor: D. Futuyma

No Abstract Available

Book Review Editor: D. Futuyma