SEXUALLY TRANSMITTED DISEASE AND PARASITE-MEDIATED SEXUAL SELECTION

Few studies have investigated the consequences of parasite-mediated sexual selection on the parasites involved. In some cases parasite-mediated sexual selection could lead to increased virulence, but I develop a simple model that shows that, if a parasite is sexually transmitted (i.e., is a sexually transmitted disease, or STD) and if mating success of the host is adversely affected by the parasite, then less virulent STDs will be selected for because transmission of the STD depends on the mating success of the host. This selection for reduced virulence could have important consequences for the role of STDs in sexual selection.     

Sexually transmitted disease in a promiscuous insect, Adalia bipunctata

Abstract.

• 1 Sexually transmitted diseases (STDs) have rarely been reported in insects and other invertebrates. The majority of those reported involve organisms where sexual transmission is augmented by either vertical (i.e. inherited) transmission, or horizontal transmission, independent of host sexual activity.
• 2 We here demonstrate the existence of an STD in the coccinellid beetle Adalia bipunctata. This species bears a parasitic mite of the genus Coccipolipus. We show that, like many other podapolipid mites, this mite is transmitted between host individuals at a high rate during copulation. It also appears to be transmitted at a low rate between non-copulating individuals.
• 3 We show that infected female A.bipunctata produced eggs at a reduced rate, and that the eggs produced by infected females have highly decreased viability. However, no effect of infection upon host longevity was observed.
• 4 The results are discussed in relation to the incidences of sexually transmitted disease in invertebrates in general, the causes of disease symptoms, and the importance of this disease in the evolution of A.bipunctata.

     

The role of host phenology in determining the incidence of an insect sexually transmitted infection

Changes in the timing of life history events within the year alter the degree to which the activity patterns of different species coincide, making the dynamics of interspecific interactions sensitive to the phenology of the interacting parties. For parasites, the availability of suitable hosts to infect represents a crucial determinant of dynamics, and changes in the host (and parasite) phenology may thus alter disease epidemiology and the conditions for disease maintenance. We tested the hypothesis that the incidence of a sexually transmitted mite infection, Coccipolipus hippodamiae, in Adalia bipunctata ladybird beetles in Sweden was determined by host phenology, namely presence/absence of sexual contact between cohorts of the host. We observed that the pattern of mite presence/absence across Swedish A. bipunctata populations was highly reproducible between years, implying a persistent biological/ecological basis underlying the incidence. Further, ladybirds from populations where the mite was absent were able to acquire mites during copulation, develop a mite infection, and transmit infection onward, indicating an ecological (rather than biological) driver of mite incidence. Observations of ladybird phenology in natural populations provided evidence of sexual contact between overwintered and new cohort adults in populations where the mite was present. In contrast, new cohort ladybirds in the two northern Swedish populations where the mite was not present had not had sexual contact with the overwintered generation, creating a ‘hard stop’ to mite transmission. We conclude that variation in host phenology may be an important driver of the incidence of sexually transmitted infections (STIs) by determining the presence/absence of sexual contact between generations. More generally, we hypothesize that sensitivity to variation in host phenology will be highest for parasites like STIs that infect one host species, one host life stage and are directly transmitted on contact between host individuals.     

The effect of aggregative overwintering on an insect sexually transmitted parasite system

In contrast to the extensively studied sexually transmitted diseases (STDs) of humans, little is known of the ecology or evolutionary biology of sexually transmitted parasites in natural systems. This study of a sexually transmitted parasite on an insect host augments our understanding of both the parasite's population dynamics and virulence effects. The impact of overwintering was assessed on the prevalence of the parasitic mite Coccipolipus hippodamiae on the two-spot ladybird, Adalia bipunctata. First, the effect of infection on host survival was examined during the stressful overwintering period. Box experiments in the field revealed that the infected ladybirds, especially males, are less likely to survive overwintering. The study provides the first evidence that the parasite harms males and suggests revisions of theories on the adaptive virulence of sexually transmitted parasites. It also indicates the importance of using a range of experimental conditions because virulence can be dependent on host condition and sex. Box experiments were also used to examine whether transmission of the parasite occurs within overwintering aggregations. These revealed that substantial transmission does not occur in aggregations and that transmission is predominantly sexual. Overall, the virulence effects and the lack of transmission mean that the overwintering period acts to diminish parasite prevalence and will retard the spring epidemic associated with host reproductive activity.     

Sexually transmitted diseases in polygynous mating systems: prevalence and impact on reproductive success

Studies of disease in relation to animal mating systems have focused on sexual selection and the evolution of sexual reproduction. Relatively little work has examined other aspects of ecological and evolutionary relationships between host social and sexual behaviour, and dynamics and prevalence of infectious diseases; this is particularly evident with respect to sexually transmitted diseases (STDs). Here, we use a simulation approach to investigate rates of STD spread in host mating systems ranging from permanent monogamy to serial polygyny or polyandry and complete promiscuity. The model assumes that one sex (female) is differentially attracted to the other, such that groups of varying size are formed within which mating and disease transmission occur. The results show that equilibrium disease levels are generally higher in females than males and are a function of variance in male mating success and the likelihood of a female switching groups between mating seasons. Moreover, initial rates of disease spread (determining whether an STD establishes in a population) depend on patterns of host movement between groups, variance in male mating success and host life history (e.g. mortality rates). Male reproductive success can be reduced substantially by a sterilizing STD and this reduction is greater in males that are more 'attractive' to females. In contrast, females that associate with more attractive males have lower absolute fitness than females associating with less attractive males. Thus, the potential for STDs to act as a constraint on directional selection processes leading to polygyny (or polyandry) is likely to depend on the details of mate choice and group dynamics.     

Sexually transmitted disease epidemics in a natural insect population

1. The epidemiology of sexually transmitted diseases (STDs) in human and domesticated populations is well documented. However, there has been less study of STDs in natural populations. 2. We investigated STD dynamics in the model system involving a host from the most speciose group of animals: the insects. We investigated temporal variation in the prevalence of the sexually transmitted mite Coccipolipus hippodamiae on its ladybird host, Adalia bipunctata. 3. Field surveys over two seasons showed a repeated pattern of a profound epidemic in the overwintered cohort and a later prevalence decline. 4. In order to understand the key factors in the dynamics of this system we studied the phenology of the host and simulated parasite dynamics in the overwintered cohort using a model with within-sex homogeneity in mating rate and field-measured parameter values. The similarity of natural and simulation prevalence levels allowed us to carry out sensitivity analysis and hence to identify the key determinants of the dynamics. 5. The observed pattern of periodic extreme prevalence combined with system persistence probably results from time lags in host recruitment and widespread promiscuity. 6. Our findings improve our understanding of STDs in natural populations and illustrate the importance of examining seasonality and time delays in population dynamics in order to fully understand the characteristics of natural populations and their parasites.     

Constraints on parasite fecundity and transmission in an insect-STD system

Parasites that are sexually transmitted (causing sexually transmitted diseases, or STDs) can have important effects on host population dynamics, but we know almost nothing for such parasites about constraints on fecundity and transmission. In this study, we have examined the effect of two potentially important constraints in one of the few empirically well-studied animal-STD systems, the ladybird, Adalia bipunctata , and its sexually transmitted mite, Coccipolipus hippodamiae . Using a factorial design, we manipulated: (i) within-host competition, by varying infection intensity; and (ii) host condition, by introducing dietary stress. Infection with C . hippodamiae significantly reduced ladybird survival whether or not diet was restricted, and restricting diet led to reduced survival regardless of infection status. Increased infection intensity and reduced host condition (dietary stress) both independently constrained per capita rates of parasite egg production and the development of infective larvae. Furthermore, when host condition was compromised, significantly fewer larvae were transmitted per adult mite during copulation. The effect of infection intensity on per mite transmission was more complex: there was no significant effect when hosts were fed normally, but when the ladybirds were nutritionally stressed higher infection intensity was associated with a slight increase in the numbers that were transmitted (despite the fact that mite fecundity was reduced under these conditions). These results indicate that host condition and within-host competition may both play an important role in shaping the epidemiology of the A . bipunctata – C . hippodamiae system, by influencing the parasite's basic reproductive ratio (R₀) and the rate of epidemic spread. Our data also extend general insights into STD ecology, by highlighting the importance of constraints on disease dynamics that are likely to be widespread.     

Parasite infection has little effect on sexual signals and reproductive behaviour in strawberry poison frogs

Acoustic and visual signals like mating calls or colour patterns, which are involved in courtship displays, can be affected by the health status of an individual. For example, infection with parasites can result in sexual signals that are less intense or conspicuous than in healthy individuals; thus it is generally assumed that sexual signals indicate the quality of the holder. We use strawberry poison frogs from Costa Rica to find out (1) whether there is variation in the parasite load among individuals, (2) whether the parasite load varies over time, (3) whether the parasite load has an effect on acoustic and visual signals, which are supposed to be important for female mate choice and (4) whether the parasite load influences the reproductive behaviour. We found that 60 % of our focal males were infested with an intestinal nematode of the genus Cosmocerca while the parasite load changed considerably over time. We found slight associations between the intensity of acoustic signals (advertisement calls), visual signals (colour and brightness contrasts) and parasite load. Non-infected males produced significantly lower pulse rates than parasitised males. As pulse rate has been shown to decrease with age, we suggest that males without parasites become older than infected males. Moreover, the visual contrasts of the frogs varied across natural backgrounds. An interaction between parasite infection and the background of the calling place affected the ventral colour contrast. This finding suggests that at least some males without parasites defend calling places that render them more conspicuousness to conspecifics than other potential positions would. There were no differences in several elements of reproductive behaviour between parasitised and healthy males. Further studies are necessary to find out whether parasite infection varies among populations, and whether at higher levels it would have a stronger influence on sexual signals, behaviour and lifetime fitness.     

Sexually transmitted diseases of insects: distribution, evolution, ecology and host behaviour

Sexually transmitted diseases (STDs) of insects are known from the mites, nematodes, fungi, protists and viruses. In total 73 species of parasite and pathogen from approximately 182 species of host have been reported. Whereas nearly all vertebrate STDs are viruses or bacteria, the majority of insect STDs are multicellular ectoparasites, protistans or fungi. Insect STDs display a range of transmission modes, with 'pure' sexual transmission only described from ectoparasites, all of which are mites, fungi or nematodes, whereas the microparasitic endo-parasites tend to show vertical as well as sexual transmission. The distribution of STDs within taxa of insect hosts appears to be related to the life histories of the hosts. In particular, STDs will not be able to persist if host adult generations do not overlap unless they are also transmitted by some alternative route. This explains the observation that the Coleoptera seem to suffer from more STDs than other insect orders, since they tend to diapause as adults and are therefore more likely to have overlapping generations of adults in temperate regions. STDs of insects are often highly pathogenic, and are frequently responsible for sterilizing their hosts, a feature which is also found in mammalian STDs. This, combined with high prevalences indicates that STDs can be important in the evolution and ecology of their hosts. Although attempts to demonstrate mate choice for uninfected partners have so far failed it is likely that STDs have other effects on host mating behaviour, and there is evidence from a few systems that they might manipulate their hosts to cause them to mate more frequently. STDs may also play a part in sexual conflict, with males in some systems possibly gaining a selective advantage from transmitting certain STDs to females. STDs may well be important factors in host population dynamics, and some have the potential to be useful biological control agents, but empirical studies on these subjects are lacking.     

Sexually transmitted disease and the evolution of mating systems

Sexually transmitted diseases (STDs) have been shown to increase the costs of multiple mating and therefore favor relatively monogamous mating strategies. We examine another way in which STDs can influence mating systems in species in which female choice is important. Because more popular males are more likely to become infected, STDs can counteract any selective pressure that generates strong mating skews. We build two models to investigate female mate choice when the sexual behavior of females determines the prevalence of infection in the population. The first model has no explicit social structure. The second model considers the spatial distribution of matings under social monogamy, when females mated to unattractive males seek extrapair fertilizations from attractive males. In both cases, the STD has the potential to drastically reduce the mating skew. However, this reduction does not always happen. If the per contact transmission probability is low, the disease dies out and is of no consequence. In contrast, if the transmission probability is very high, males are likely to be infected regardless of their attractiveness, and mating with the most attractive males imposes again no extra cost for the female. We also show that optimal female responses to the risk of STDs can buffer the prevalence of infection to remain constant, or even decrease, with increasing per contact transmission probabilities. In all cases considered, the feedback between mate choice strategies and STD prevalence creates frequency-dependent fitness benefits for the two alternative female phenotypes considered (choosy vs. randomly mating females or faithful vs. unfaithful females). This maintains mixed evolutionarily stable strategies or polymorphisms in female behavior. In this way, a sexually transmitted disease can stabilize the populationwide proportion of females that mate with the most attractive males or that seek extrapair copulations.     

Extreme Costs of Mating for Male Two-Spot Ladybird Beetles

Male costs of mating are now thought to be widespread. The two-spot ladybird beetle (Adalia bipunctata) has been the focus of many studies of mating and sexual selection, yet the costs of mating for males are unknown. The mating system of A. bipunctata involves a spermatophore nuptial gift ingested by females after copulation. In this study, we investigate the cost to males of mating and of transferring spermatophores in terms of lifespan, ejaculate production and depletion of nutritional reserves. We found that males faced a strong trade-off between mating and survival, with males that were randomly assigned to mate a single time experiencing a 53% reduction in post-mating lifespan compared to non-mating males. This is among the most severe survival costs of a single mating yet reported. However, spermatophore transfer did not impact male survival. Instead, the costs associated with spermatophores appeared as a reduced ability to transfer spermatophores in successive matings. Furthermore, males ingested more food following spermatophore transfer than after matings without spermatophores, suggesting that spermatophore transfer depletes male nutritional reserves. This is to our knowledge the first report of an effect of variation in copulatory behaviour on male foraging behaviour. Overall, our study highlights the advantages of assessing mating costs using multiple currencies, and suggests that male A. bipunctata should exhibit mate choice.     

Female preference for multi-modal courtship: multiple signals are important for male mating success in peacock spiders

A long-standing goal for biologists has been to understand how female preferences operate in systems where males have evolved numerous sexually selected traits. Jumping spiders of the Maratus genus are exceptionally sexually dimorphic in appearance and signalling behaviour. Presumably, strong sexual selection by females has played an important role in the evolution of complex signals displayed by males of this group; however, this has not yet been demonstrated. In fact, despite apparent widespread examples of sexual selection in nature, empirical evidence is relatively sparse, especially for species employing multiple modalities for intersexual communication. In order to elucidate whether female preference can explain the evolution of multi-modal signalling traits, we ran a series of mating trials using Maratus volans. We used video recordings and laser vibrometry to characterize, quantify and examine which male courtship traits predict various metrics of mating success. We found evidence for strong sexual selection on males in this system, with success contingent upon a combination of visual and vibratory displays. Additionally, independently produced, yet correlated suites of multi-modal male signals are linked to other aspects of female peacock spider behaviour. Lastly, our data provide some support for both the redundant signal and multiple messages hypotheses for the evolution of multi-modal signalling.     

Male hosts are responsible for the transmission of a trophically transmitted parasite, Pterygodermatites peromysci, to the intermediate host in the absence of sex-biased infection

Field studies have identified that male-biased infection can lead to increased rates of transmission, so we examined the relative importance of host sex on the transmission of a trophically transmitted parasite (Pterygodermatites peromysci) where there is no sex-biased infection. We experimentally reduced infection levels in either male or female white-footed mice (Peromyscus leucopus) on independent trapping grids with an anthelmintic and recorded subsequent infection levels in the intermediate host, the camel cricket (Ceuthophilus pallidipes). We found that anthelmintic treatment significantly reduced the prevalence of infection among crickets in both treatment groups compared with the control, and at a rate proportional to the number of mice de-wormed, indicating prevalence was not affected by the sex of the shedding definitive host. In contrast, parasite abundance in crickets was higher on the grids where females were treated compared with the grids where males were treated. These findings indicate that male hosts contribute disproportionately more infective stages to the environment and may therefore be responsible for the majority of parasite transmission even when there is no discernable sex-biased infection. We also investigated whether variation in nematode length between male and female hosts could account for this male-biased infectivity, but found no evidence to support that hypothesis.     

Testing for links between face color and age,dominance status,parity, weight,and intestinal nematode infection in a sample of female Japanese macaques

Studies of the role of secondary sexual ornaments in mate choice tend to focus on colorful traits in males, but females of many animal species express colorful ornamentation too. Among non-human primates, investigations into the role of female secondary sexual traits as indicators of life history characteristics, reproductive success, and health status have mostly focused on sexual swellings, whereas only few studies have been conducted on the role of facial color. Recent studies on rhesus macaques and mandrills suggested that female ornamentation might provide information about female life history characteristics, but not on disease resistance factors and parasite infection, which have been shown to affect male ornamentation in some non-primate species. In Japanese macaques (Macaca fuscata), females have brightly colored faces that are indicative of their reproductive status. Here, we aimed to determine whether female facial color might also convey information about age, dominance rank, parity, weight, and intestinal nematode infection in free-ranging individuals. We analyzed whether female facial parameters (luminance and redness) were linked to these individual characteristics, using digital photography and data on intestinal parasite infection collected systematically during 1 month for each of seven free-ranging females. We found no evidence to suggest that female facial color is an indicator of any of these measures in Japanese macaques. Considering our small data set, it is still preliminary to draft any clear conclusions. Future studies combining digital, hormonal, parasitological and behavioral data are needed to assess the possible role of female face color on male preferences and mating choice in Japanese macaques.     

Effect of sexual segregation on host-parasite interaction: Model simulation for abomasal parasite dynamics in alpine ibex (Capraibex)

We investigated whether sexual segregation might affect parasite transmission and host dynamics, hypothesising that if males are the more heavily infected sex and more responsible for the transmission of parasite infections, female avoidance of males and the space they occupy could reduce infection rates. A mathematical model, simulating the interaction between abomasal parasites and a hypothetical alpine ibex (Capraibex) host population composed of its two sexes, was developed to predict the effect of different degrees of sexual segregation on parasite intensity and on host abundance. The results showed that when females tended to be segregated from males, and males were distributed randomly across space, the impact of parasites was the lowest, resulting in the highest host abundance, with each sex having the lowest parasite intensity. The predicted condition that minimises the impact of parasites in our model was the one closest to that observed in nature where females actively seek out the more segregated sites while males are less selective in their ranging behaviour. The overlapping of field observation with the predicted optimal strategy lends support to our idea that there might be a connection between parasite transmission and sexual segregation. Our simulations provide the biological boundaries of host-parasite interaction needed to determine a parasite-mediated effect on sexual segregation and a formalised null hypothesis against which to test future field experiments.     

Cloacal bacterial diversity increases with multiple mates: evidence of sexual transmission in female common lizards

Sexually transmitted diseases have often been suggested as a potential cost of multiple mating and as playing a major role in the evolution of mating systems. Yet there is little empirical data relating mating strategies to sexually transmitted microorganisms in wild populations. We investigated whether mating behaviour influences the diversity and composition of cloacal assemblages by comparing bacterial communities in the cloaca of monandrous and polyandrous female common lizards Zootoca vivipara sampled after the mating period. We found that polyandrous females harboured more diverse communities and differed more in community composition than did monandrous females. Furthermore, cloacal diversity and variability were found to decrease with age in polyandrous females. Our results suggest that the higher bacterial diversity found in polyandrous females is due to the sexual transmission of bacteria by multiple mates. The impact of mating behaviour on the cloacal microbiota may have fitness consequences for females and may comprise a selective pressure shaping the evolution of mating systems.     

Mating Competition,Promiscuity, and Life History Traits as Predictors of Sexually Transmitted Disease Risk in Primates

Competition among males influences the distribution of copulations and should therefore influence the spread of sexually transmitted diseases (STDs). We developed a model to investigate STDs in the mating and social systems found in primates, and we tested predictions using comparative methods. In the model, groups were distributed on a square lattice in which males or females disperse and males undergo characteristic dominance trajectories at maturity (challenge vs. queuing). We investigated the impact of mating rate, mating skew, migration rate of males or females, and group size on disease spread and prevalence. The model generated several predictions: 1) STD prevalence is higher in females than males; 2) STD risk increases with copulation rate; 3) high skew is negatively associated with STD risk; 4) STD risk is higher for all individuals when females disperse and 5) when mortality rates are lower; and 6) reproductive skew and later age of male dominance (queuing) produce more strongly female-biased STD prevalence. In comparative tests, we quantified STD risk as prevalence and richness of sexually transmitted organisms at the host species level. We found positive associations between host longevity and higher STD richness, and only (nonsignificant) weak trends for females to have higher STD prevalence. Mating skew showed a weakly positive association with STD richness, contrary to predictions of our model but consistent with predictions from a previous model. In some tests, we also found that female dispersal resulted in greater STD infection risk. Collectively, these results demonstrate that mating competition and demography influence patterns of STD infection, with mortality rates having the strongest effects in comparative tests.     

Host intrinsic determinants and potential consequences of parasite infection in free‐ranging red‐fronted lemurs (Eulemur fulvus rufus)

Parasites and infectious diseases represent ecological forces shaping animal social evolution. Although empirical studies supporting this link abound in various vertebrate orders, both the study of the dynamics and impact of parasite infections and infectious diseases in strepsirrhine primates have received little empirical attention. We conducted a longitudinal parasitological study on four groups of wild red‐fronted lemurs (Eulemur fulvus rufus) at Kirindy Forest, Madagascar, during two field seasons in consecutive years to investigate i) the degree of gastrointestinal parasite infection on population and individual levels and ii) factors potentially determining individual infection risk. Using a comprehensive dataset with multiple individually assignable parasite samples as well as information on age, sex, group size, social rank, and endocrine status (fecal androgen and glucocorticoid), we examined parasite infection patterns and host traits that may affect individual infection risk. In addition, we examined whether parasite infection affects mating and reproductive success. Our results indicated high variability in parasite infection on individual and population levels. Time of year and group size was important determinants of variability in parasite infection. Variation in hormone levels was also associated with parasite species richness and parasite infection intensity. Differences in parasite infection between years indicate a potential immune‐enhancing function of steroid hormones on nematode infections, which has not been reported before from other vertebrates studied under natural conditions. Male mating and reproductive success were not correlated to any measure of parasite infection, which suggests a nonfunctional role of the parasites we examined in primate sexual selection. Am J Phys Anthropol, 2010. © 2010 Wiley‐Liss, Inc.     

Sexual Behavior and Condom Use among Gay Men,Female Sex Workers,and Their Customers: Evidence from South Korea

Background

Despite the significance for sexually transmitted diseases (STD) control in East Asia, few studies have examined the relationship between high-risk sexual behavior and condom use. We investigated how three sexually vulnerable groups for STDs show differences in condom use behaviors (CUBs) depending on their STD infection.

Methods

The source of data came from the National Survey for STD Prevalence Rate and Sexual Behavior of the High-Risk Sexual Community. The effects of behavioral determinants on CUBs were estimated by using path analysis models. An 11-item questionnaire assessing subjects’ health risk behaviors, sexual beliefs, sexual risk behaviors, and condom use.

Results

Condom use was higher for men who have sex with men (MSM; n = 108) when they were bisexuals and had high self-efficacy, for Johns (Johns; n = 118) when they had experience of STD infection, and for female sex workers (FSWs; n = 1,083) when they had high self-efficacy, did not engage in drunken sex, and were anxious about infection. Regardless of whether they were infected with STDs, FSWs always used condom when they had high sexual beliefs. On the contrary, Johns exhibited a negative relationship between sexual risk behavior and condom use when they had experience of STD infection. The variable commonly significant to all three groups was the number of sex partners; but it exhibited a positive relationship with MSMs and Johns, and a negative one with FSWs.

Conclusions

CUBs were related to sexual beliefs as well as sexual risk behavior. At the same time, the experience of STD infection mediated the relationship between the two. Therefore, we need to draw social attention to promote safer sex among STD-vulnerable groups.     

Female reproductive decisions and parasite burden in a calopterygid damselfly (Insecta: Odonata)

There is currently a gap in sexual selection theory about how much the environment drives female mating decisions. We present field data that suggest that female sexual behaviour in the damselfly Calopteryx haemorrhoidalis is influenced by parasite burden. Male wing pigmentation in Calopteryx is a sexually selected trait that signals a male's ability to cope with eugregarine parasites (an intestinal parasite that feeds on the adult's ingested food). Because adult C. haemorrhoidalis females also show wing pigmentation, we examined whether this trait is similarly influenced by parasite burden and whether it may signal the female's reproductive value. MaleC. haemorrhoidalis defend riverine substrates that females use for oviposition. After copulation and during oviposition, females are guarded by the copulating male against intruder males. Alternatively, females may avoid mating and ‘steal’ an oviposition site within a male's territory. In the present study, we found that the amount of female wing pigmentation was negatively correlated with the number of eugregarines present. Females with more parasites produced fewer eggs, survived fewer days, spent less time during courtship, ‘inspected’ fewer males before mating, had a lower mating success, were guarded for less time during oviposition and engaged in fewer ‘stealing’ events during oviposition. The reduced egg production and survival of heavily infected females may result from eugregarine depletion of the females' consumed food reserves. Thus, to offset reduced longevity, heavily infected females may accept a mating more rapidly and mate with fewer males. ‘Stealing’ behaviour may be related to the female's differential use of sperm from some males, particularly high-quality males. Interestingly, males that mated with low-pigmented females showed greater variance in wing pigmentation than did males that mated with high-pigmented females. Possibly, female wing pigmentation may signal a female's reproductive value, which provides females with longer mate-guarding episodes and reduced interference from intruder males. This study points out one possible constraint, intestine parasites, that females may face during mating decisions. Because females in bad condition mate with males in both good and bad condition, this constraint may be pervasive enough to weaken the intensity of selection for a male sexually selected trait, wing pigmentation, and help to maintain its variation in phenotypic expression. Copyright 2003 Published by Elsevier Science Ltd on behalf of The Association for the Study of Animal Behaviour.