Some animal groups, such as birds, seem prone to extreme forms of sexual selection. One contributing factor may be sex linkage of genes affecting male displays and female preferences. Here we show that sex linkage can have substantial effects on the genetic correlation between these traits and consequently for Fisher's runaway and the good-genes mechanisms of sexual selection. Under some kinds of sex linkage (e.g. Z-linked preferences), a runaway is more likely than under autosomal inheritance, while under others (e.g., X-linked preferences and autosomal displays), the good-genes mechanism is particularly powerful. These theoretical results suggest empirical tests based on the comparative method.

Eight bacteriophages were adapted for rapid growth under similar conditions to compare their evolved, endpoint fitnesses. Four pairs of related phages were used, including two RNA phages with small genomes (MS2 and Qβ), two single-stranded DNA phages with small genomes (ϕX174 and G4), two T-odd phages with medium-sized, double-stranded DNA genomes (T7 and T3), and two T-even phages with large, double-stranded DNA genomes (T6 and RB69). Fitness was measured as absolute growth rate per hour under the same conditions used for adaptation. T7 and T3 achieved the highest fitnesses, able to increase by 13 billionfold and three-quarters billionfold per hour, respectively. In contrast, the RNA phages achieved low fitness maxima, with growth rates approximately 400-fold and 4000-fold per hour. The highest fitness limits were not attributable to high mutation rates or small genome size, even though both traits are expected to enhance adaptation for fast growth. We suggest that major differences in fitness limits stem from different “global” constraints, determined by the organization and composition of the phage genome affecting whether and how it overcomes potentially rate-limiting host processes, such as transcription, translation, and replication. Adsorption rates were also measured on the evolved phages. No consistent pattern of adsorption rate and fitness was observed across the four different types of phages, but within each pair of related phages, higher adsorption was associated with higher fitness. Different adsorption rate limits within pairs may stem from “local” constraints—sequence differences leading to different local optima in the sequence space.

Mating among the immediate products of meiosis (intratetrad mating) is a common feature of many organisms with parthenogenesis or with mating-type determination in the haploid phase. Using a three-locus deterministic model we show that intratetrad mating, unlike other systems of mating, allows sheltering of deleterious recessive alleles even if there is only partial linkage between a mating locus and a load locus. Moreover, modifiers that reduce recombination between the load and mating-type locus will spread to fixation, even when there is no linkage disequilibrium between these loci in the population as a whole. This seeming contradiction to classical expectation is because partial linkage generates linkage disequilibrium among segregating loci within a tetrad, which then acts as the “mating unit.”

A striking linear dominance relationship for uniparental mitochondrial transmission is known between many mating types of plasmodial slime mold Physarum polycephalum. We herein examine how such hierarchical cytoplasmic inheritance evolves in isogamous organisms with many self-incompatible mating types. We assume that a nuclear locus determines the mating type of gametes and that another nuclear locus controls the digestion of mitochondria DNAs (mtDNAs) of the recipient gamete after fusion. We then examine the coupled genetic dynamics for the evolution of self-incompatible mating types and biased mitochondrial transmission between them. In Physarum, a multiallelic nuclear locus matA controls both the mating type of the gametes and the selective elimination of the mtDNA in the zygotes. We theoretically examine two potential mechanisms that might be responsible for the preferential digestion of mitochondria in the zygote. In the first model, the preferential digestion of mitochondria is assumed to be the outcome of differential expression levels of a suppressor gene carried by each gamete (suppression-power model). In the second model (site-specific nuclease model), the digestion of mtDNAs is assumed to be due to their cleavage by a site-specific nuclease that cuts the mtDNA at unmethylated recognition sites. Also assumed is that the mtDNAs are methylated at the same recognition site prior to the fusion, thereby being protected against the nuclease of the same gamete, and that the suppressor alleles convey information for the recognition sequences of nuclease and methylase. In both models, we found that a linear dominance hierarchy evolves as a consequence of the buildup of a strong linkage disequilibrium between the mating-type locus and the suppressor locus, though it fails to evolve if the recombination rate between the two loci is larger than a threshold. This threshold recombination rate depends on the number of mating types and the degree of fitness reduction in the heteroplasmic zygotes. If the recombination rate is above the threshold, suppressor alleles are equally distributed in each mating type at evolutionary equilibrium. Based on the theoretical results of the site-specific nuclease model, we propose that a nested subsequence structure in the recognition sequence should underlie the linear dominance hierarchy of mitochondrial transmission.

Dioecy is reported to be correlated with a number of ecological traits, including tropical distribution, woody growth form, plain flowers, and fleshy fruits. Previous analyses have concentrated on determining whether dioecy is more likely to evolve in lineages possessing these traits, rather than considering the speciation and extinction rates of dioecious lineages with certain combinations of traits. To address the association between species richness in dioecious lineages as a function of the ecological traits, we compared the evolutionary success (i.e., relative species richness) of dioecious focal lineages with that of their nondioecious sister groups. This test was repeated for the evolutionary success of randomly chosen nondioecious lineages (control lineages) compared with their nondioecious sister groups. If the possession of certain ecological traits enhances the evolutionary success of dioecious lineages, we predict an association between the presence of these traits and relative species richness in the former, but not latter, set of sister-group comparisons. Dioecious focal lineages with a higher number of these traits experienced higher evolutionary success in sister-group comparisons, whereas no trend was found for the control focal lineages. The increase in evolutionary success was especially true for dioecious focal lineages that had a tropical distribution or fleshy fruit. We discuss how these results provide strong support for differential evolutionary success theories for the correlations between dioecy and the ecological traits considered.

Although pollinator-mediated natural selection has been measured on many floral traits and in many species, the extent to which selection is constrained from producing optimal floral phenotypes is less frequently studied. In particular, negative correlations between flower size and flower number are hypothesized to be a major constraint on the evolution of floral displays, yet few empirical studies have documented such a trade-off. To determine the potential for genetic constraints on the adaptive evolution of floral displays, I estimated the quantitative genetic basis of floral trait variation in two populations of Lobelia siphilitica. Restricted maximum likelihood (REML) analyses of greenhouse- grown half-sib families were used to estimate genetic variances and covariances for flower number and six measures of flower size. There was significant genetic variation for all seven floral traits in both populations. Flower number was negatively genetically correlated with four measures of flower size in one population and three measures in the other. When the genetic variance-covariance matrices were combined with field estimates of phenotypic selection gradients, the predicted multivariate evolutionary response was less than or opposite in sign to the selection gradient for flower number and five of six measures of flower size, suggesting genetic constraints on the evolution of these traits. More generally, my results indicate that the adaptive evolution of floral displays can be constrained by trade- offs between flower size and number, as has been assumed by many theoretical models of floral evolution.

Evolutionary consequences of thermally varying environments were studied in the ciliated protozoan Tetrahymena thermophila. Replicated lines were propagated for 60 days, a maximum of 500 generations, in stable, slowly fluctuating (red spectrum), and rapidly fluctuating (blue spectrum) temperatures. The red and blue fluctuations had a dominant period length of 15 days and two hours, respectively. The mean temperature of all time series was 25°C and the fluctuating temperatures had the same minimum (10°C), maximum (40°C), and variance. During the experiment, population sizes and biomasses were monitored at three-day intervals. After the experiment, carrying capacity and maximum growth rate were measured at low (15°C), intermediate (25°C), and high (35°C) temperatures for each experimental line. Physiological changes in the lines were assessed by measuring the expression of stress-induced heat shock protein Hsp90 at 25°C, 35°C, and 39°C. Population sizes and biomasses showed no differences between stable, blue, or red temperature treatments during the experiment. Also, after the experiment, mean carrying capacities and maximum growth rates were comparable in the stable, blue, and red temperature treatments. The expression of Hsp90 was higher in lines from the blue environment than in lines from the stable environment. Lines from the red environment had an intermediate level of Hsp90 expression. This supports the hypothesis that inducible thermotolerance and expression of canalizing genes can evolve in response to rapidly varying environments. Furthermore, we found correlative evidence of benefits and disadvantages of high Hsp90 expression. Lines with high expression of Hsp90 had an increased growth rate at the highest temperature when food resources were not limiting growth. At low and intermediate temperatures the same lines had the lowest carrying capacities.

Population genetic theory predicts that in small populations, random genetic drift will fix and accumulate slightly deleterious mutations, resulting in reduced reproductive output. This genetic load due to random drift (i.e., drift load) can increase the extinction risk of small populations. We studied the relationship between genetic variability (indicator of past population size) and reproductive output in eight isolated, natural populations of the hermaphroditic snail Lymnaea stagnalis. In a common laboratory environment, snails from populations with the lowest genetic variability mature slower and have lower fecundity than snails from genetically more variable populations. This result suggests that past small population size has resulted in increased drift load, as predicted. The relationship between genetic variability and reproductive output is independent of the amount of nonrandom mating within populations. However, reproductive output and the current density of snails in the populations were not correlated. Instead, data from the natural populations suggest that trematode parasites may determine, at least in part, population densities of the snails.

Learning is thought to be adaptive in variable environments, whereas constant, predictable environments are supposed to favor unconditional, genetically fixed responses. A dichotomous view of behavior as either learned or innate ignores a potential evolutionary interaction between the learned and innate components of a behavioral response. We addressed this interaction in the context of oviposition substrate choice in Drosophila melanogaster, asking two main questions. First, will learning also evolve in a constant environment in which it always pays to show the same choice? Second, how does an opportunity to learn affect the evolution of the innate (genetic) component of oviposition substrate choice? We exposed experimental populations to four selection regimes, involving selection on oviposition substrate preference (an orange versus a pineapple medium). In two selection regimes the flies were selected for preference either for the orange medium, or for the pineapple medium. In the remaining two selection regimes the flies were also selected for preference for either orange or pineapple, but additionally could use past experience (aversion learning) to decide which medium it paid to avoid. Lines exposed to the latter selection regimes evolved improved learning ability, indicating that learning may be advantageous even if the same behavioral response is favored every generation. Furthermore, of the two selection regimes that favored oviposition on the pineapple medium, the regime that allowed for learning led to the evolution of a stronger innate preference for pineapple, than the regime that did not allow for learning. In contrast, of the two regimes that selected for oviposition on the orange medium, the one that allowed for learning led to a smaller evolutionary change of the innate preference. Thus, an opportunity to learn facilitated the evolution of innate preference under selection for preference for pineapple, but hindered it under selection for preference for orange. We discuss possible mechanisms for this effect.

Drosophila subobscura is geographically widespread in the Old World. Around the late 1970s, it was accidentally introduced into both South and North America, where it spread rapidly over broad latitudinal ranges. This invading species offers opportunities to study the speed and predictability of trait evolution on a geographic scale. One trait of special interest is body size, which shows a strong and positive latitudinal cline in many Drosophila species, including Old World D. subobscura. Surveys made about a decade after the invasion found no evidence of a size cline in either North or South America. However, a survey made in North America about two decades after the invasion showed that a conspicuous size cline had evolved and (for females) was coincident with that for Old World flies. We have now conducted parallel studies on 10 populations (13° of latitude) of flies, collected in Chile in spring 1999. After rearing flies in the laboratory for several generations, we measured wing sizes and compared geographic patterns (versus latitude or temperature) for flies on all three continents. South American females have now evolved a significant latitudinal size cline that is similar in slope to that of Old World and of North American flies. Rates of evolution (haldanes) for females are among the highest ever measured for quantitative traits. In contrast, the size cline is positive but not significant for South or North American males. At any given latitude, South American flies of both sexes are relatively large; this in part reflects the relatively cool climate of coastal Chile. Interestingly, the sections of the wing that generate the size cline for females differ among all three continents. Thus, although the evolution of overall wing size is predictable on a geographic scale (at least for females), the evolution of size of particular wing components is decidedly not.

Barriers to gene flow that act after mating but before fertilization are often overlooked in studies of reproductive isolation. Where species are sympatric, such “cryptic” isolating barriers may be important in maintaining species as distinct entities. Drosophila yakuba and its sister species D. santomea have overlapping ranges on the island of São Tomé, off the coast of West Africa. Previous studies have shown that the two species are strongly sexually isolated. However, the degree of sexual isolation observed in the laboratory cannot explain the low frequency (∼1%) of hybrids observed in nature. This study identifies two “cryptic” isolating barriers that may further reduce gene flow between D. yakuba and D. santomea where they are sympatric. First, noncompetitive gametic isolation has evolved between D. yakuba and D. santomea: heterospecific matings between the two species produce significantly fewer offspring than do conspecific matings. Second, conspecific sperm precedence (CSP) occurs when D. yakuba females mate with conspecific and heterospecific males. However, CSP is asymmetrical: D. santomea females do not show patterns of sperm usage consistent with CSP. Drosophila yakuba and D. santomea females also differ with respect to remating propensity after first mating with conspecific males. These results suggest that noncompetitive and competitive gametic isolating barriers may contribute to reproductive isolation between D. yakuba and D. santomea.

Haplodiploidy (encompassing both arrhenotoky and paternal genome elimination) could have originated from coevolution between male-killing endosymbiotic bacteria and their hosts. In insects, haplodiploidy tends to arise in lineages that rely on maternally transmitted bacteria for nutrition and that have gregarious broods in which competition between siblings may occur. When siblings compete, there is strong selection on maternally transmitted elements to kill males. I consider a hypothetical bacterial phenotype that renders male zygotes effectively haploid by preventing chromosome decondensation in male-determining sperm nuclei. By causing high male mortality, such a phenotype can be advantageous to the bacterial lineage. By eliminating paternal genes, it can also be advantageous to the host female. A simple model shows that the host female will benefit under a wide range of values for the efficiency of resource re-allocation, the efficiency of transmission, and the viability of haploid males. This hypothesis helps to explain the ecological correlates of the origins of haplodiploidy, as well as such otherwise puzzling phenomena as obligate cannibalism by male Micromalthus beetles, reversion to diploidy by aposymbiotic male stictococcid scale insects, and the bizarre genomic constitution of scale insect bacteriomes.

The origin of and evolutionary transitions among the extraordinary diverse forms of parental care in teleost fish remain largely unknown. The “safe harbor” hypothesis predicts that the evolution from a “guarding” to a “brooding” form of care in teleost fish is associated with shifts in reproductive and life-history features such as reduced fecundity, and increased egg volume with higher parental investment. Robust phylogenetic hypotheses may help to identify evolutionary changes in key traits associated with differences in the form of parental care. Here, we used reconstruction of ancestral character states to study the evolution of the two forms of parental care, bubble nesting and mouthbrooding in the fighting fish genus Betta. We also applied a comparative analysis using the phylogenetic generalized least-squares method to test the “safe harbor” hypothesis by evaluating differences between the two forms of parental care in standard length, life-history traits, and three habitat variables. Evolutionary hypotheses were derived from the first molecular phylogeny (nuclear and mitochondrial DNA sequence data; 4448 bp) of this speciose group. Ancestral character state reconstructions of the evolution of the form of parental care in the genus Betta, using the methods of unweighted parsimony and maximum likelihood, are uncertain and further indicate a high rate of evolutionary transitions. Applying different weights for the suspected directionality of changes, based on the consistent phenotypic and behavioral differences found between bubble nesters and mouthbrooders, recurrent origin of mouthbrooding in the genus Betta is favored using parsimony. Our comparative analyses further demonstrate that bubble nesters and mouthbrooders do not have a consistent set of life-history correlates. The form of parental care in Betta is correlated only with offspring size, with mouthbrooders having significantly bigger offspring than bubble nesters, but is not correlated with egg volume, clutch size, and broodcare duration, nor with any of the three habitat variables tested. Our results thus challenge the general predictions of the “safe harbor” hypothesis for the evolution of alternative brood care forms in the fighting fish genus Betta.

Loberg Lake, Alaska was colonized by sea-run Gasterosteus aculeatus between 1983 and 1988, after the original stickleback population was exterminated. Annual samples from 1990 to 2001 reveal substantial evolution of lateral plate (armor) phenotypes. The 1990 sample was nearly monomorphic for the complete plate morph, which is monomorphic in local sea-run populations; the low plate morph, which is usually monomorphic in local freshwater populations, was absent. By 2001, the frequency of completes had declined to 11%, and lows had increased to 75%. The partial plate morph and two unusual intermediate plate phenotypes were generally rare, but occurrence of the intermediates was unexpected. These intermediate phenotypes rarely occur in other, presumably older, polymorphic populations. When low morphs first appeared, they averaged 6.8 plates per side, indicating that the ancestral plate number of low morphs is high, and their mean has subsequently declined. Contemporary evolution in this population indicates that threespine stickleback adapt to freshwater habitats within decades after invasion from the ocean, and thus phenotypes in most populations are adapted to current conditions.

Estimating species phylogeny from a single gene tree can be especially problematic for studies of species flocks in which diversification has been rapid. Here we compare a phylogenetic hypothesis derived from cytochrome b (cyt b) sequences with another based on amplified fragment length polymorphisms (AFLP) for 60 specimens of a monophyletic riverine species flock of mormyrid electric fishes collected in Gabon, west-central Africa. We analyze the aligned cyt b sequences by Wagner parsimony and AFLP data generated from 10 primer combinations using neighbor-joining from a Nei-Li distance matrix, Wagner parsimony, and Dollo parsimony. The different analysis methods yield AFLP tree topologies with few conflicting nodes. Recovered basal relationships in the group are similar between cyt b and AFLP analyses, but differ substantially at many of the more derived nodes. More of the clades recovered with the AFLP characters are consistent with the morphological characters used to designate operational taxonomic units in this group. These results support our hypothesis that the mitochondrial gene tree differs from the overall species phylogeny due at least in part to mitochondrial introgession among lineages. Mapping the two forms of electric organ found in this group onto the AFLP tree suggests that posteriorly innervated electrocytes with nonpenetrating stalks have independently evolved from anteriorly innervated, penetrating-stalk electrocytes at least three times.

Although mitochondrial DNA markers have several properties that make them suitable for phylogeographic studies, they are not free of difficulties. Phylogeographic inferences within and between closely related species can be mislead by introgression and retention of ancestral polymorphism. Here we combine different phylogenetic, phylogeographic, and population genetic methods to extract the maximum information from the Liolaemus darwinii complex. We estimate the phylogeographic structure of L. darwinii across most of its distributional range, and we then estimate relationships between L. darwinii and the syntopic species L. laurenti and L. grosseorum. Our results suggest that range expansion of these lineages brought them into secondary contact in areas where they are presently in syntopy. Here we present the first evidence for introgression in lizards from temperate South America (of L. darwinii mitochondrial DNA into L. laurenti and L. grosseorum), and for incomplete lineage sorting (between L. darwinii and L. laurenti). We show that a combination of methods can provide additional support for inferences derived from any single method and thus provide more robust interpretations and narrow the range of plausible hypotheses about mechanisms and processes of divergence. Additional studies are needed in this group of lizards and in other codistributed groups to determine if Pleistocene climatic changes could be a general factor influencing the evolutionary history of a regional biota.

The simultaneous effects of selective agents acting on somatic growth rates, their interactions, and their interactions with local environmental conditions that vary across a species' geographic range are potentially complex and poorly known. This is particularly true of viviparous ectotherms whose offspring may be adapted to the gestation environment provided by their mothers. We studied multiple sources of growth rate variation in a widespread, viviparous reptile, including the effect of the maternal environment on growth following parturition. Females in early pregnancy were collected from replicate populations close to the tropical and temperate margins of this species' range. These females completed gestation in either of two different, common environments designed to simulate the thermal and photoperiod environments at the sampling locations. Our experiments revealed complex growth rate evolution between the northern and southern extremes of Eulamprus quoyii's geographic range and local adaptation of growth rates to maternal environments. Unique to this study was the manifestation of these growth rate differences, entrained in utero, but expressed following parturition and maintained through to maturity despite the presence of compensatory growth. In addition to providing the most complete picture to date of the evolution of somatic growth in a viviparous ectotherm, our study suggests that understanding local adaptation to maternal gestation environments, in terms of both mean growth rates and growth rate reaction norms, could change our understanding of how growth rates have evolved in other viviparous ectotherms. Indeed, such local adaptation may provide a selective advantage in the evolution of viviparity.

Heterozygosity-fitness correlations (HFCs) at noncoding genetic markers are commonly assumed to reflect fitness effects of heterozygosity at genomewide distributed genes in partially inbred populations. However, in populations with much linkage disequilibrium (LD), HFCs may arise also as a consequence of selection on fitness loci in the local chromosomal vicinity of the markers. Recent data suggest that relatively high levels of LD may prevail in many ecological situations. Consequently, LD may be an important factor, together with partial inbreeding, in causing HFCs in natural populations. In the present study, we evaluate whether LD can generate HFCs in a small and newly founded population of great reed warblers (Acrocephalus arundinaceus). For this purpose dyads of full siblings of which only one individual survived to adult age (i.e., returned to breed at the study area) were scored at 19 microsatellite loci, and at a gene region of hypothesized importance for survival, the major histocompatibility complex (MHC). By examining siblings, we controlled for variation in the inbreeding coefficient and thus excluded genome-wide fitness effects in our analyses. We found that recruited individuals had significantly higher multilocus heterozygosity (MLH), and mean d² (a microsatellite-specific variable), than their nonrecruited siblings. There was a tendency for the survivors to have a more diverse MHC than the nonsurvivors. Single-locus analyses showed that the strength of the genotype-survival association was especially pronounced at four microsatellite loci. By using genotype data from the entire breeding population, we detected significant LD between five of 162 pairs of microsatellite loci after accounting for multiple tests. Our present finding of a significant within-family multilocus heterozygosity-survival association in a nonequilibrium population supports the view that LD generates HFCs in natural populations.

Age-specific reproductive success has been demonstrated in many species. Three hypotheses have been raised to explain this general phenomenon: the experience hypothesis based on age-specific reproductive experience, the effort hypothesis based on age-specific reproductive effort, and the selection hypothesis based on progressive disappearance of phenotypes due to variation in individual productivity and survival. We used data from a long-term study of Leach's storm-petrels (Oceanodroma leucorhoa) to present a single test of mutually exclusive predictions about the relationship between early breeding success and longevity. There should be no correlation between early breeding success and longevity under the experience hypothesis, a negative correlation under the effort hypothesis, and a positive correlation under the selection hypothesis. We found a significant (P < 0.0001) positive relationship between success in the first two breeding attempts and longevity in this population of long-lived seabirds, strongly suggesting that low-productivity parents were also less likely to survive early breeding. These data provide some of the strongest support to date for the selection hypothesis.

A recent study suggests that sex-specific dispersal rates can be quantitatively estimated on the basis of sex- and state-specific (pre- vs. postdispersal) F-statistics. In the present paper, we extend this approach to account for the hierarchical structure of natural populations, and we validate it through individual-based simulations. The model is applied to an empirical data set consisting of 536 individuals (males, females, and predispersal juveniles) of greater white-toothed shrews (Crocidura russula), sampled according to a hierarchical design and typed for seven autosomal microsatellite loci. From this dataset, dispersal is significantly female biased at the local scale (breeding- group level), but not at the larger scale (among local populations). We argue that selective pressures on dispersal are likely to depend on the spatial scale considered, and that short-distance dispersal should mainly respond to kin interactions (inbreeding or kin competition avoidance), which exert differential pressure on males and females.

While it now appears likely that sympatric speciation is possible, its generality remains contentious. If it really is rare, then most natural populations must not fit the assumptions of sympatric speciation theory. A better understanding of these assumptions may help identify when sympatric speciation is or is not likely. This paper investigates two such assumptions: that genetic variation for stringent assortative mating is not limiting and that females are not penalized for mating assortatively. Simulations demonstrate that the speed of sympatric speciation is very sensitive to the population's capacity for stringent assortative mating and is potentially extremely slow. The rapid divergence often thought to be a hallmark of sympatric speciation may only occur in a restricted area of parameter space.

RNA viruses are widely used to study evolution experimentally. Many standard protocols of virus propagation and competition are done at nominally low multiplicity of infection (m.o.i.), but lead during one passage to two or more rounds of infection, of which the later ones are at high m.o.i. Here, we develop a model of the competition between wild type (wt) and a mutant under a regime of alternating m.o.i. We assume that the mutant is deleterious when it infects cells on its own, but derives a selective advantage when rare and coinfecting with wt, because it can profit from superior protein products created by the wt. We find that, under these assumptions, replication at alternating low and high m.o.i. may lead to the stable coexistence of wt and mutant for a wide range of parameter settings. The predictions of our model are consistent with earlier observations of frequency-dependent selection in vesicular stomatitis virus and human immunodeficiency virus type 1. Our results suggest that frequency-dependent selection may be common in typical evolution experiments with viruses.

We demonstrate that extending copulation enhances probability of paternity in sand lizards and that determinants of copulation duration depend on a males' mating order (first or second). First males, with no information on presence of rivals, extend copulation when mating with a more fecund female. Second males, however, adjust copula duration in relation to a first male's relatedness with his female, which there is reason to believe can be deduced from the MHC-related odor of the copulatory plug. Male-female relatedness negatively influences a male's probability of paternity, and when second males are in a favored role (i.e., the first male is the one more closely related to the female), second males transfer larger ejaculates, resulting in higher probability of paternity. This result corroborates predictions from recent theoretical models on sperm expenditure theory incorporating cryptic female choice and sexual conflict. More specifically, the results conform to a “random roles” model, which depicts males as being favored by some females and disfavored by others, but not to a “constant-type” model, in which a male is either favored or disfavored uniformly by all females in a population.