Below are the full abstracts of some of my more recent publications. They are listed in cronological order, based on year of publication (updated January, 2002). See my complete list of publications.

BACK TO my main home page...

Abstract - In the water strider Gerris odontogaster, the sexes engage in a precopulative struggle. Females are reluctant to mate and try to dislodge males attempting copulation by repeatedly performing backward somersaults. This paper addresses the ultimate significance of the precopulative struggle. Females may be reluctant (1) in order to avoid costly and superfluous copulations (intersexual conflict), or (2) females may be assessing the males' quality as mates (mate assessment). In a laboratory experiment the effects of sex ratio and population density on the mating behaviour of G. odontogaster was studied. As the density of males increased, females were less reluctant to mate and mated more frequently. The results of the experiment are consistent with predictions generated by a model of a sexual conflict situation, but not with what may be expected if females were assessing the quality of the males. It is concluded that the primary function of female reluctance in this species is simply to avoid copulations. Further, it has been demonstrated in previous studies that females bias matings towards certain male phenotypes, and that the mechanism of selection is female reluctance behaviour. The results of this study are discussed in terms of 'good genes' and 'nonadaptive' views of female choice (sensu Kirkpatrick 1987a). It is suggested that intersexual selection in G. odontogaster represents a case of 'nonadaptive' female choice, since mate selection is a side-effect rather than an ultimate aim of female reluctance behaviour.

Abstract - Studies of phenotypic selection in natural populations are often concerned with simply detecting selection. In adopting a more mechanistic approach, this study compares the sexual selection regimes in natural populations of the water strider Gerris odontogaster with a priori predictions of selection, based on a number of previous field and laboratory studies of the behavioral mechanisms of selection. In this species, a general reluctance of females to mate allows for intersexual selection for ability to subdue reluctant females in males. Female reluctance to mate has been shown to decrease with increasing population density, suggesting that sexual selection should be weaker in high density populations. Three different populations with large differences in population density were studied. A number of traits including parasite load, body mass, body size and male abdominal process length were found to experience significant sexual selection. The investigated populations differed considerably with regard to the total strength of selection on the measured traits and the form of selection on single traits. In general, males in the population with the highest density experienced the weakest selection for grasping ability. This pattern is ascribed to density-related alterations of female mating behavior. Selection for male grasping ability, as reflected by selection on male abdominal process length, is reduced in high-density populations where reluctant females are more easily subdued. Further, the studied populations differed significantly in mean phenotype and phenotypic variance for male abdominal process length. It is suggested that interpopulational differences in selective regimes may generate local adaptations with respect to male abdominal process length, and that gene flow may contribute to the maintenance of the high genetic variation in this trait. It is further suggested that more empirical effort should be made in quantifying and understanding spatial and temporal variation in selection in natural populations, since this may provide information e.g. on the prevalence of local adaptations in metric traits and on the mechanisms of selection.

Abstract - The courtship behavior of the semi-aquatic Pisaurid fishing spider Dolomedes fimbriatus was examined in the laboratory. Male courtship was triggered by the presence of female drag-lines, presumably by a female sex pheromone since males did not respond with courtship to male drag-lines. Male courtship behavior included vibratory signaling (water surface waves), leg-waving, and following female drag-lines. Vibratory signaling was a major courtship component, and signals were produced at a regular rate (mean rate: 8.33 ± 1.53 sec., n = 97). Irrespective of whether females were mated or unmated, females were very aggressive towards males, and sexual cannibalism prior to copulation occurred in 6.6 % of the female attacks on males (n = 76). The capture success rate of females depended on whether the male was attacked from a distance or from immediate proximity. The occurrence of sexual cannibalism of courting males by virgin Dolomedes females is discussed, and it is suggested that this behavior of fishing spiders may represent an adaptive female strategy rather than mistaken identity.

Abstract - Natural selection is generally thought to constrain the effects of sexual selection on secondary sexual traits. In hemimetabolous arthropods, the evolution of complex secondary sexual morphologies may be constrained by interference during the molting process. This study assesses the effects of phenotypic variation in secondary sexual traits during ontogeny in the water strider Gerris odontogaster. In this species, males are provided with a grasping apparatus consisting of two abdominal, ventral processes. It is demonstrated that the length of the processes is positively related to the duration of the ultimate larval molt, presumably because the processes constitute a mechanical hindrance during molting. It is further shown that larvae experience a high risk of mortality from cannibalism during their ultimate molt, and that this risk is density dependent. Based on the assumption that cannibalistic events occur at random during molting, males with long processes will suffer higher risk of mortality during molting than will males with short processes. It is concluded that this form of developmental constraint can constrain the evolution of male abdominal processes, and quantitative cost curves for the length of male abdominal processes are presented. Simulations of natural selection due to larval mortality show positive density dependence, in contrast to sexual selection in this species. It is suggested that the net effect of selection on secondary sexual traits in this species varies between negative and positive values, and that net selection is zero only within a narrow range of environmental conditions. An interpopulational comparison between natural populations showed that mean trait value covaries negatively with environmental variables related to the cost of the trait in accordance with this suggestion.

Abstract - Conflict between the sexes over control of copulation may drive the coevolution of elaborate genitalia and other secondary sexual structures. Support for this hypothesis is limited to male adaptations that function to enhance male control over females in copulation (e.g. structures that function to clasp females). Evidence for morphological adaptation in females is critical to the hypothesis, yet lacking. Here, we present the first experimental evidence, demonstrating that female abdominal spines in water striders function to increase female control over copulation. By experimentally extending the phenotypic size range of these spines, we show that this morphological adaptation, specific to females, allows them to thwart harassing males, and as a result reduce the frequency of costly matings to females. This demonstrates the coevolutionary nature of sexual conflict and that females are indeed active participants in the evolutionary conflict over control of reproduction.

Abstract - Assortative mating by size is a common mating pattern that can be generated by several different behavioral mechanisms, with different evolutionary implications. Assortative mating is typically associated with sexual selection, and has been regarded as an attribute of populations, species, mating systems or even higher order taxa. In most animal groups, however, appropriate analyses of assortative mating at these different levels are lacking, and the causes and forms of assortative mating are poorly understood. Here, we analyze 45 different population level estimates of assortative mating and non-random mating by size in seven confamiliar species of water striders that share a common mating system. A hierarchical comparative analysis shows that virtually all the variance within the clade occurs among samples within species. We then employ meta-analysis to estimate the overall strength of assortative mating, to determine the form of assortative mating, and to further assess potential differences among species as well as the probable causes of assortative mating in this group of insects. We found overall weak but highly significant positive assortative mating. We show that analyses of the degree of heteroscedasticity in plots of male versus female size are critical, since the evolutionary implications of "true" and "apparent" assortative mating differ widely, and conclude that the positive assortative mating observed in water striders was of the "true" rather than the "apparent" form. Further, within samples, mating individuals were significantly larger than non-mating individuals in both males and females. All of these non-random mating patterns were consistent among species, and we conclude that weak positive assortative mating by size is a general characteristic of those water strider species that share this mating system. We use our results to illustrate the importance of distinguishing between different forms of assortative mating, to discriminate between various behavioral causes of assortative mating, and to assess potential sources of inter-populational variance in estimates of assortative mating. Finally, we discuss the value of using meta-analytic techniques for detecting overall patterns in multiple studies of non-random mating.

Abstract - Field studies demonstrate that natural populations of water striders (Heteroptera: Gerridae), sharing a common mating system, are characterized by weak assortative mating by size and by large sizes of mating males and females relative to single individuals. This study presents an experimental assessment of the causes of these mating patterns. The effects of male and female body size on each of three components of mating were studied in three water strider species in the laboratory. Large females of all three species mated more frequently, and were copulated and guarded longer than small females. Large males mated more frequently than small males in all three species, and also guarded females for longer durations in the two species where the average duration of mate guarding was prominent. However, we found an antagonistic effect of male size on copulation duration: small males copulated longer than large males in all three species. We show that the combined effects of these size biases mimic the mating patterns found in the wild, e.g. weak and variable assortative mating, and stronger and less variable size ratios of mating versus non-mating individuals in females relative to males. We suggest that the antagonistic effects of male size on copulation and guarding duration may be a key source of interpopulational variation in assortative mating and sexual selection on male size. Further, the results of the current study show that neither spatial or temporal covariation in size, nor mechanical constraints, cause assortative mating in water striders. It is concluded that assortative mating in this group of insects is the combined result of several interacting mechanisms. Male and female choice (either active or passive forms) of large mates and male-male exploitation competition for mates play potentially important roles in producing population level assortative mating.

Abstract - Water striders (Heteroptera: Gerridae) exhibit two different types of mating behavior. The most common mating system (type I) is characterized by strong apparent conflicts of interest between the sexes, and conspicuous pre- and post-copulatory struggles. Some species exhibit a mating system which involves much less apparent conflict (type II), and lack the intense copulatory struggles. I argue that the predominant mating system in water striders is a direct consequence of sexual conflicts over mating decisions. Matings involve high costs to females (increased predation risk and energetic expenditure) but few, if any, balancing direct benefits. Sperm displacement rates are high, and males thus gain from rematings. Mating frequencies are high, and females mate multiply for reasons of convenience. In these species, males are considered to have "won" the evolutionary conflict over the mating decision in the sense that they have made acceptance of superfluous matings "the best of a bad job" for females, by evolving behavioral and morphological traits that make it costly for females to reject males attempting copulations. Females, however, have apparently evolved a variety of counteradaptations to male harassment, to gain control over mating. Further, I suggest that sexual conflict may have played a crucial role in the evolution of type II matings from type I matings. Water strider mating systems are very plastic. Females assess the rate of male harassment, and make adaptive mating decisions based on this assessment. As a consequence of variation in female mating behavior, the characteristics of the mating system vary with a number of environmental factors. Knowledge of the behavioral dynamics of water strider mating systems makes mechanistic hypotheses of sexual selection possible and provides a framework in which variations between populations in non-random mating and sexual selection can be understood and even accurately predicted. Recent insights gained from the study of water strider mating systems highlight two fruitful research avenues; (1) intraspecific variability of mating systems deserves more attention and (2) observational and experimental studies at the population level concerned with mating patterns should be linked with experimental studies of behavioral processes at the individual level.

Abstract - Several hypotheses have been proposed for the evolution of sexual cannibalism by females. Newman and Elgar (1991; Am. Nat. 138, 1372-1395) suggested that sexual cannibalism prior to mating by virgin female spiders may have evolved as a result of female foraging considerations. According to this model, an adult female's decision to mate or cannibalize a courting male should be based on an assessment of the male's value as a meal versus his value as a mate. The current study provides an empirical test of the assumptions and predictions of this model in the sexually cannibalistic fishing spider. Adult females were subjected to different food treatments, and exposed to adult males in the laboratory. However, only one of the assumptions of the model and none of its five predictions were upheld. We failed to find any effects of female foraging, female mating status, female size, male size or time of the season on females' behavior towards courting males. Females behaved stereotypically, and many females were left unmated despite numerous mating opportunities. We also demonstrate costs of sexual cannibalism in a natural population. We propose that the act of sexual cannibalism in the fishing spider is non-adaptive, and develop a model for the evolution of premating sexual cannibalism in spiders based on genetic constraints. According to this hypothesis, sexual cannibalism by adult females may have evolved as an indirect result of selection for high and non-discriminate aggression during previous ontogenetic stages. Genetic covariance between different components of aggressive behavior may constrain the degree to which (1) juvenile and adult aggression, and/or (2) aggression towards con- and heterospecifics, can vary independently. We briefly review the support for our model, and suggest several critical tests that may be employed to assess the assumptions and predictions of the model.

Abstract - Rapid evolution of genitalia is one of the most general patterns of morphological diversification in animals. Despite its generality, the causes of this evolutionary trend remain obscure. Several alternative hypotheses have been suggested to account for the evolution of genitalia (notably the lock-and-key, pleiotropism, and sexual selection hypotheses). Here, I argue that thorough intraspecific studies are key to gain insight into the patterns and processes of genitalic evolution. Critical assumptions and predictions that may be used to distinguish between the different hypotheses are identified and discussed. However, current knowledge of selection on genitalia, or even of the degree of phenotypic and genotypic variability of genital morphology, is highly limited, allowing only a very tentative assessment of the various hypotheses. In-depth single species studies of current patterns and processes of selection on genitalia are badly needed, and a single species research program is briefly outlined.

Abstract - Rapid divergence of male genitalia is one of the most general evolutionary trends in animals with internal fertilization, but the mechanisms of genital evolution are poorly understood. The current study represents the first comprehensive attempt to test the main hypotheses that have been suggested to account for genital evolution (the lock-and-key, sexual selection and pleiotropy hypotheses) with intraspecific data. We measure multivariate phenotypic selection in a water strider species, by relating five different components of fitness (mating frequency, fecundity, egg hatching rate, offspring survival rate and offspring growth rate) to a suite of genital and non-genital morphological traits (in total 48). Body size had a series of direct effects in both sexes. Large size in females was positively related to both fecundity and offspring survival. There was positive sexual selection for large size in males (mating frequency), which to some extent was offset by a reduced number of eggs laid by females mated to large males. Male genitalic morphology influenced male mating frequency, but the detected directional selection on genitalia was due to indirect selection on phenotypically correlated non-intromittent traits. Further, we found no assortative mating between male intromittent genitalia and female morphology. Neither did we find any indications of male genitalia conveying information of male genetic quality. Several new insights can be gained from our study. Most importantly, our results are in stark disagreement with the long standing lock-and-key hypothesis of genital evolution, as well as with certain models of sexual selection. Our results are, however, in agreement with other models of sexual selection as well as with the pleiotropy hypothesis of genital evolution. Fluctuating asymmetry of bilaterally symmetrical traits, genital as well as non-genital, had few effects on fitness. Females with low fluctuating asymmetry in leg length produced offspring with a higher survival rate, a pattern most probably caused by direct phenotypic maternal effects. We also discuss the relevance of our results to sexual conflict over mating, and the evolution of sexual traits by coevolutionary arms races between the sexes.

Abstract - Understanding the costs of sexual interaction is fundamental to any analysis of the evolution of sexual behavior, including intersexual conflicts of interest over courtship, mating or mate-guarding. We used flow-through CO2 respirometry to compare the energetic costs to female water striders of locomotion and rest on still and moving water surfaces and to estimate costs to females of resisting male mating and guarding attempts. Water striders were placed solitarily or as mating or non-mating pairs in a respirometry chamber containing a water surface adequate to fully float an active strider. In the main set of experiments, repeated measures of energetic expenditure were performed on individual females while (1) mating, (2) bearing a small solder weight approximating the weight of a male, (3) bearing a recently euthanized male and, (4) alone; each male's standard (resting) metabolic rate was also determined. Behavior was closely monitored throughout all recordings. For water striders, the estimated gross cost of self-transport represents an exceptionally low cost of pedestrian locomotion and the first record of the energetics of skating across a water surface. However, energetic expenditure was highly context dependent. First, mate carrying involved increased energetic expenditure for females. The average individual standard (resting) metabolic rates (SMR) for females (76.3uW) increased up to 41% when there was a male in the mating position, although a portion of this increase may have been due to increases above male SMR associated with grasping the female. Solitary females engaged in routine locomotory behavior (i.e., "cruising", positional maintainance on an aquatic treadmill) consumed an average of 334.6 uW of energy. Cruising females carrying males or solder weights consumed 19.9% and 26.8% more energy, respectively, than unburdened cruising females. Females engaged in more intensive evasive locomotion consumed 400.2 uW when solitary and 572.2 uW when carrying a male in the mating position, a 43% increase between the solitary and mating contexts (double the increase observed in cruising females betwen the unloaded and loaded states). Second, water striders engage in premating struggles, during which females attemp to reject males attempting copulation. Our study shows that mate rejection, and mate choice, is energetically very costly for females. Females consumed an averageof 936.6 uW during struggling activity, a dramatic increase in energetic expenditure compared to non-struggling mating females. We also develop a quantitative model of the female convenience polyandry hypothesis, and show that optimal female mating behaviour (energetically) will depend critically on male harassment rate. At a certain harassment rate threshold, accepting superfluous matings becomes the "best of a bad job" for females.

Abstract - Rapid and divergent evolution of male genitalia represents one of the most general evolutionary patterns in animals with internal fertilization, but the causes of this evolutionary trend are poorly understood. Several hypotheses have been proposed to account for genitalic evolution, most prominent of which are the lock-and-key, sexual selection and pleiotropy hypotheses. However, insights into the evolutionary mechanisms of genitalic evolution are hindered by a lack of relevant in-depth studies of genital morphology. The current study represents the first comprehensive intraspecific study of the patterns of inheritance and plasticity of genital morphology. We used a biparental progenies breeding design to study the effects of food stress during ontogeny on phenotypic expression of a suite of genital and non-genital morphological traits, both linear traits and multivariate shape indicies, in a natural population of the water strider Gerris incognitus. In general, genitalic traits were as variable as non-genital traits, both phenotypically and genotypically. Average narrow sense heritability of genital traits was 0.47 (SE = 0.05). Further, while food stress during development had a large impact on adult morphology, and expression of genitalic traits exhibited significant levels of condition dependence, different genotypes did not significantly differ in their ability to cope with food stress. Genitalic conformation was also both phenotypically and genetically correlated with general morphological traits. Our results provide a novel framework with which to test hypotheses of genitalic evolution. The documented patterns of the genetic architecture of genital morphology are in stark disagreement with predictions generated by the long-standing lock-and-key hypothesis. However, our results can be accomodated by the pleiotropy hypothesis, as well as by several variants of the sexual selection hypothesis. We failed to find any additive genetic components in fluctuating asymmetry of any bilaterally symmetrical traits. Further, the effects on fluctuating asymmetry of food stress during development were, surprisingly enough, very low and insignificant. Some methodological implications of our study are discussed, such as the bias introduced by the non-negativity constraint in restricted maxumum likelihood estimation of variance components.

Abstract - The study of phenotypic plasticity, one of the most important mechanisms of phenotypic adaptation, is by tradition focused on differences in ontogenetically static phenotypic expression in different environments. Ontogenetic reaction norms, in contrast, describe how phenotypes unfold during growth in different environments. We studied the ontogenetic reaction norms of the shape of a series of defensive spines in dragonfly larvae, both in the laboratory and in a number of natural populations. In a laboratory rearing experiment, we demonstrated that these spines grew more solid and elongated when water borne environmental cues of fish predators were present: a graded phenotypic plasticity of defensive spine morphology. The ontogenetic reaction norms of defensive spines were also found to differ in natural populations with and without fish. A detailed analysis of the growth trajectories showed that this differentiation was primarily due to ontogenetic acceleration in environments with fish, leading to relative allometric peramorphosis in these environments. However, while the ontogenetic trajectories of shape in some spines diverged at the onset of ontogeny in the two environments, those of others remained parallel until a given phase of ontogeny. Hence, the timing of the developmental divergence of these phenotypically integrated traits differed, suggesting differences in the underlying regulatory mechanisms. Our results illustrate that an integration of environmental and ontogenetic approaches to the study of phenotypic differentiation can significantly promote our understanding of the ecology and evolution of adaptive phenotypic plasticity.

Abstract - Random measurement error is ubiquitous in morphometric data, and it often causes serious problems since it reduces the power of all biological and statistical inferences. We stress that measurement error is a potential problem primarily when true phenotypic variation in shape is relatively small, such as in studies of intraspecific or intrapopulational variation in shape. A model for the partitioning of measurement error in landmark based morphometrics is presented. We recommend that two steps are taken to reduce the inferential problems caused by measurement error: minimization and quantification. The impact of measurement error can be reduced in a number of ways, depending on the methods used to collect, process and analyse data, and we give some practical advice. We recommend that morphometric studies routinely report the impact of measurement error, at least in studies dealing with relatively small amounts of variation in shape, by calculating and reporting the repeatabilities of the shape measures used. Using shape variation in young perch as an empirical example, we demonstrate (a) that shape space is more seriously affected by measurement error than is size space, (b) that the estimates of repeatability will critically depend on the error components that are actually repeated in each repeated measure, (c) that the impact of different components of measurement error, as well as effect factors, can be partitioned and assessed by planned hierarchical repeated measurements followed by nested analyses of variance, (d) that measurement error is unevenly distributed among different components of shape space and (e) that total shape space is increasingly affected by measurement error as the number of landmarks used in the analysis increases.

Abstract - Rapid divergent evolution of male genitalia is one of the most general evolutionary trends in animals with internal fertilisation - the shape of genital traits often provide the only reliable characters for species identification. Yet, the evolutionary processes responsible for this remarkable pattern remain obscure. The longstanding lock-and-key hypothesis, still deeply rooted among taxonomists, suggests that genitalia evolve by pre-insemination hybridisation avoidance. The recently proposed sexual selection hypothesis, in contrast, suggests that divergent evolution of genitalia is the result of sexual selection, brought about by variation in post-insemination paternity success among males. By contrasting pairs of related clades of insects differing in mating system, I was able to assess how the opportunity for post-mating sexual selection affects the rate of divergent evolution of male genitalia. Genital evolution was found to have been more than twice as divergent in groups where females mate multiply compared to groups where females mate only once, a pattern which was not found for other morphological traits. These findings provide strong empirical evidence in favour of the post-mating sexual selection hypothesis of genital evolution. See supplementary information.

Arnqvist, G. and Danielsson, I. (1999) Copulatory behavior, genital morphology and male fertilization success in water striders. Evolution 53:147-156.

Abstract - Recent theoretical and empirical interest in post-mating processes have generated a need for increasing our understanding of the sources of variance in fertilization success among males. Of particular importance is whether such post-mating sexual selection merely reinforces the effects of pre-mating sexual selection, or if other types of male traits are involved. In the current study, we document large intraspecific variation in last male sperm precedence in the water strider Gerris lateralis. Further, male relative paternity success was repeatable across replicate females, showing that males differ consistently in their ability to achieve fertilizations. By analyzing shape variation in male genital morphology, we were able to demonstrate that the shape of male intromittent genitalia was related to relative paternity success. This is the first direct experimental support for the suggestion that male genitalia evolve by post-mating sexual selection. A detailed analysis revealed that different components of male genitalia had different effects, some affecting male ability to achieve sperm precedence and others affecting male ability to avoid sperm precedence by subsequent males. Further, the effects of the shape of the male genitalia on paternity success was in part dependent on female morphology, suggesting that selection on male genitalia will depend on the frequency distribution of female phenotypes. We failed to find any effects of other morphological traits, such as male body size or the degree of asymmetry in leg length, on fertilization success. Although males differed consistently in their copulatory behavior, copulation duration was the only behavioral trait that had any significant effect on paternity.

BACK TO my main home page...