Thesis (Ph. D.)--University of Texas at Austin, 2005. / Supervisor: Walt Wilczynski. Vita. Includes bibliographical references.

Multicomponent displays are composed of traits, such as coloration, structural ornaments, and behavior, that become integrated and signal information to conspecifics. Estimation of multicomponent displays in fishes often involves measurement of color traits. Fish color measurements are often obtained following immobilization via chemical anesthesia; however, the anesthetics may alter the resulting measurements, for example by darkening the skin. Girardinus metallicus, a poeciliid fish endemic to Cuba, has a multicomponent courtship and aggressive display. Black morph males exhibit black ventral coloration including the gonopodium (copulatory organ) and yellow in the non-black areas of their bodies. I investigated the effects of common anesthetics on coloration measurements of G. metallicus. I measured the hue, saturation, and brightness of the anterior dorsal, posterior dorsal, posterior ventral, and caudal body regions, from digital images of the same males obtained without using anesthetic and anesthetized using tricaine methane sulfonate (MS222) and eugenol (clove oil). Because multicomponent displays are intriguing with respect to sexual selection, I investigated the importance of size and coloration traits in sexual selection via female choice and male-male competition in G. metallicus. I found that saturation and hue did not differ significantly across treatments (anesthetization using MS222, anesthetization using clove oil, and without anesthetic in a small glass chamber containing water). However, brightness was greater under the anesthetics, possibly due to photographing the fish behind water and glass in the Non-anesthetic treatment or due to reflectivity differences of the iridophores. The body regions varied in hue, saturation, and brightness. Most importantly, I found differences in the responses of different body regions to the anesthetic treatments, suggesting that anesthetics may affect coloration in unpredictable ways, and that multiple regions of fish should be measured when assessing overall coloration. My results suggest that photographing fish in a glass chamber without anesthetic may be an effective way to obtain digital images for color analysis without using anesthetics that may influence coloration. Having determined a good method for color measurement, I then investigated the role of the multicomponent display in sexual selection. Through direct interaction tests, I found that dominant males had brighter and more saturated yellow coloration than subordinate males, and that dominant males courted more than subordinate males. Within high yellow males, dominant males attempted more copulations than subordinate males. Interestingly, low yellow, subordinate males attempted more copulations than low yellow, dominant males, suggesting that subordinate males invested time into attempting copulations rather than engaging in potentially risky aggressive behavior. I observed a greater difference in body size between the males in pairs to which I could assign dominance status than pairs to which I could not assign dominance status, suggesting the importance of standard length in aggression in this species. I found that yellow saturation may serve to signal status without the males resorting to aggressive interactions due to only half the pairs exhibiting aggression. Because aggression is key to mating success in G. metallicus, my findings that yellow coloration is correlated with aggression, in concert with previous studies showing the importance of ventral black area and body size for aggression, reinforce the idea that these males exhibit a multicomponent signal to conspecifics in the context of sexual selection.

Reproductive interference occurs when individuals direct sexual behaviours towards other species and this causes a reduction in the fitness of one or both of the participants. While initially considered to be little more than an aberration, reproductive interference is increasingly recognised, not just as a factor in determining species co-existence and community dynamics, but also as an opportunity to further our understanding of sexual selection. In this thesis I investigate the causes and consequences of reproductive interference between several species of true bugs (Order: Hemiptera) in the family Lygaeidae. These species have a polgynandrous mating system, in which both males and females mate multiple times. I found that the interspecific mating attempts often witnessed in these bugs is likely a consequence of this mating system, as potential cues for species discrimination, in the form of cuticular hydrocarbons, are available and can be utilised by the bugs in some contexts. This is further supported by the finding that pre-copulatory selection on factors such as diet and chemical protection are weak in these insects. Furthermore, my work highlights the context-dependant nature of both the fitness costs associated with reproductive interference, and also of the bugs' intra-specific behaviour. When housed in groups, female Lygaeus equestris did not show consistent fitness losses in response to harassment by either conspecific or heterospecific males, despite previous evidence showing that such males inflict costs on both egg production and longevity when interacting one-to-one. This, combined with the finding that male L. equestris alter their mate-guarding behaviour in the presence of other males, highlights the potential of behavioural flexibility to influence the outcome of inter-species interactions, and the importance of context when attempting to measure phenomena such as reproductive interference.

No description available.

The determinants of sexual dimorphism in a family of false scorpions (Pseudoscorpionida, Chernetidae) were investigated experimentally and with a literature analysis of comparative morphometric and habitat data. Species vary in the extent to which males and females differ in size of the conspicuous, prehensile pedipalps. Patterns within the Chernetidae suggest that dimorphism is a highly variable condition, relatively unconstrained by phylogenetic influences. The evolution of species with enlarged male pedipalps appears to be associated with a change from nonpairing to pairing sperm transfer behavior, and aggressive mate acquisition by males. Experiments with Dinocheirus arizonensis demonstrate a high correlation between male combat ability and chela size. Comparison of male and female life histories show prolonged development in males, and morphological comparisons implicate pedipalp dimorphism as a causative factor in this developmental rate difference. Prolonged development may be particularly costly to males, given the pattern of female sexual receptivity in this species. Females were found to become unreceptive soon after mating and remain so throughout a protected period of brood development. Experimental manipulations suggest that the male developmental rate cost is only outweighed under high density conditions when superior combat ability results in increased mating success. Repeated measures experiments failed to show any correlation between male pedipalp size and number of spermatophores accepted by a female. Parent-offspring regressions suggest the existence of additive genetic variance for male chela size and indicate a strong genetic correlation between this trait and cephalothorax length. Full-sib phenotypic correlations suggest that in D. arizonensis sexual divergence through sexual selection may be constrained by a high genetic correlation between males and females. Finally, the role of phoresy in the colonization of ephemeral, patchy habitats is investigated. Results support the hypothesis that attachment of pseudoscorpions to larger, more mobile arthropods represents a behavior functioning specifically for dispersal.

No description available.

No description available.

No description available.

How parasites influence the population dynamics of their hosts depends on 1) theproportion of individuals that carry the infection in the population, 2) what type of individuals aremost susceptible to infection and 3) the fitness effects of infection. In this study I first investigate thefrequency of malaria strains transmitted in the African winter quarters or at the European breedinggrounds in collared flycatchers (Ficedula albicollis). I then zoom in on the relationship between avianmalaria infection status and condition, expression of sexually selected ornament and reproductiveperformance of male collared flycatchers. I found that female flycatchers are more likely to beinfected than males and that both sexes have a large bias towards infection with European strains ofmalaria. Infected male flycatchers have higher reproductive success and tend to have largerornaments but there was no detected relationship between malaria infection and male condition.This is the first example, that I am aware of, of a positive relationship between malaria infection andreproductive success.

This thesis is an evaluation of the hypothesis that the spennathecae of spiders affects the sperm precedence patterns in a predictable way (Austad 1984). Spermathecae come in two varieties: cul-de-sac and conduit. Cul-de-sac spennathecae, according to the hypothesis, are supposed to lead to second male sperm priority and conduit to first male sperm priority . The hypothesis was evaluated both directly and indirectly. Direct measurements were made of paternity in two species, Pholcus phalangioides and Tetragnatha montana, both of which are cul-de-sac species. It was found that P. phalangioides complies with the predicted precedence pattern and thus does not disprove the hypothesis. This second male priority pattern was despite a much shorter mating time by second mating males. In T. montana no precedence pattern was found, with equal likelihood of first or second mating males of gaining paternity. There was in T. montana a possible influence of the duration of mating affecting the precedence pattern, with longer mating males gaining a higher paternity no matter what order they mated in. It is discussed whether or not this is due to sperm loading or genitalic stimulation (Eberhard 1985). Indirect evaluation of the hypothesis included an analysis of mating behaviour in Zygiella x-notata which is a conduit species and was chosen as a comparison to the two cul-de-sac species. In Z. x-notata it was found that there was no difference between mating duration in first and second mating males. Mating persistence is thus the same in first and second mating males, suggesting that the males cannot detect that the female is a denuded resource to second mating males. Hence first male priority may not be a factor in this species. Other indirect methods of evaluating the hypothesis involved charting the incidence of mate-guarding and mating-plugs. The expected pattern of mate-guarding was for conduit species to pre-mate guard and for cul-de-sac species to post-mate guard, because of the predicted sperm precedence patterns associated with the spermathecae. The predicted pattern was not found. In the case of mating-plugs it was predicted that these should be deployed by cul-de-sac species because it is in these species that second males are able to usurp paternity to a large extent. The opposite pattern was found with mating-plugs of various design being utilized by conduit species. It is postulated that mating-plugs are the mechanism by which first male priorities are established in conduit species, where this pattern is found. The absence of plugs in cul-de-sac species is possibly the reason that second males can cuckold. The additional data collected since 1984 reveal that patterns of paternity found in spiders seem to be more complex than was originally assumed by Austad (1984). Spermathecae are species-specific in character and this may reflect a species specificity in sperm precedence patterns. Thus the conduit I cul-de-sac dichotomy may not reflect a useful prediction of paternity patterns.