Recapture of male and female dragonflies in relation to parasitism by mites, time of season, wing length and wing cell symmetry.

For aquatic mites parasitic on dragonflies, completion of their life cycle depends on their being returned to appropriate water bodies by their hosts, after completion of engorgement. We examined whether differences among hosts in timing of emergence or phenotypic attributes might affect their probability of return to an emergence pond, and hence success of mites. Parasitized males and females of the dragonfly Sympetrum obtrusum (Hagen) did not differ in overall recapture rates. Females that had wing cell symmetry and emerged early were more likely to be recaptured than females that emerged later or had wing cell asymmetry, but there were no consistent relations between these variables and parasitism by mites. No such relations between wing cell asymmetry, emergence date, and recapture likelihood were found for males. Using randomization tests, we found that mean intensities of Arrenurus planus (Marshall) mites at host emergence were the same for recaptured females and females not recaptured; however, males that were recaptured had lower mean intensities of mites at emergence than males not recaptured. Further, mature females carried more mites than mature males, and the latter had fewer mites than newlyemerged males not recaptured. Biases in detachment of engorging mites do not explain the differences in parasitism between mature males and females, nor the differences in mite numbers between mature males and newly emerged males that were not recaptured. Rather, heavily parasitized males appear to disperse or die and are not recaptured, which should have implications for dispersal of mites and fitness of male hosts.

Communication via sex pheromones within and among Arrenurus spp. mites (Acari: Hydrachnida; Arrenuridae).

We present direct experimental evidence of pheromone use in six species of Arrenurus and indirect evidence for four species, including members of the subgenera Megaluracarus, Truncaturus, and Arrenurus. Water in which females were housed elicited arrestant behaviour in males, males oriented to the source, and at least some individuals in each species assumed the male readiness posture, a precursor to coupling. Most species responded to water treated with conspecific females, but there was also interspecific sex pheromone responsiveness. Arrenurus manubriator and A. megalurus demonstrated reciprocal pheromone cross-attractancy. Males of A. major, A. marshallae, and A. birgei responded to water from females of related species from within their subgenera. Arrenurus apetiolatus males failed to respond to conspecific female-treated water, but the same water elicited arrestant behaviour and orientation in A. manubriator. Heterospecific reactions to female-conditioned water were limited to cases involving members of the same species group and were not seen between species representing different species groups or different subgenera. The species for which cross-attractancy has been demonstrated commonly co-occur in nature, so apparently these pheromones are of limited value for species recognition. Shared reaction to sex pheromones provides additional evidence for inferring close phylogenetic relationship among species, and thus far, corresponds with morphological evidence based on adult males and larvae.

Experimental evidence for a female sex pheromone in Arrenurus manubriator (Acari: Hydrachnida; Arrenuridae).

We present experimental evidence for a water-borne female-produced sex pheromone in aquatic parasitengonine mites. Water that has contained adult female Arrenurus manubriator Marshall will elicit arrestant behaviour in conspecific adult males, and if the cue is sufficiently strong, the males will assume a readiness posture (with 4th pair of legs held over the back, bent anteromedially at the genuotibial joint) that is typically a precursor to coupling. Water that has not been exposed to female mites does not induce any behavioural response from male mites. Female-conditioned water that has been passed through a C-18 column does not elicit any response from male A. manubriator, while the rehydrated residue from the column does induce arrestant behaviour and may result in the readiness posture. The results from the C-18 extraction indicate that the pheromone is nonpolar in nature.

Conflict between antipredator and antiparasite behaviour in larval damselflies.

Larval damselflies resist infestation by parasitic larval mites by exhibiting behaviours such as grooming, crawling, swimming, and striking at host-seeking mites. Larval damselflies are known to increase time spent in these behaviours in the presence of mites but reduce time spent in these behaviours in the presence of fish predators. The presence of both fish and larval mites presents an obvious conflict: a larval damselfly may actively avoid parasitism by mites, thus increasing its risk of predation, or it may reduce its activity when fish are present, thus increasing its risk of parasitism. We analysed the behaviour of larval Ischnura verticalis in an experiment where we crossed presence and absence of fish with presence and absence of larval mites. Presence of mites induced a large increase in activity of larval I. verticalis but fish had no effect and there were no interpretable interactions between effects of mites and fish. Subsequent experiments indicated that larval I. verticalis in the presence of both mites and fish were more likely to be attacked and killed by fish than those exposed only to fish. The high activity level of I. verticalis larvae in the presence of both fish and mites may suggest that costs of parasitism are high, or that under field conditions it is rare for larvae to be in the immediate presence of both fish predators and potentially parasitic mites.

Allocation of reproductive effort in female Arrenurus spp. water mites (Acari: Hydrachnidia; Arrenuridae).

Twenty-seven species of water mites of the genus Arrenurus were compared with respect to allocation of reproductive effort. Clutch volume was positively correlated to female volume, female volume was positively correlated with clutch size and with larval volume, while clutch size and larval volume were negatively correlated. In threespace, corresponding to female volume, clutch size and larval volume, species were arranged along two trajectories representing separate reproductive strategies. The strategy characterized by small larvae, large clutches and large females corresponded to species of the subgenus Arrenurus, known to be larval parasites of adult Odonata. Members of three subgenera, known to be larval parasites of adult Diptera, followed the second strategy of small clutch size and alrge larvae. Of nine species of mites tested for intraspecific relationships, only one significant relationship was found, between female volume and clutch size for A. (Megaluracarus) bartonensis.