Résumé
There are many examples of parasites that manipulate intermediate host behaviour facilitating transmission to the definitive host. We examined this hypothesis as a possible explanation for the observed prevalence of Parelaphostrongylus tenuis, a nematode parasite of white-tailed deer (Odocoileus virginianus) in eastern North America. The prevalence of this parasite in deer populations is typically high compared with its prevalence in its gastropod intermediate hosts. It is unclear how high levels of prevalence are maintained, given that the transmission of this parasite to deer is considered to occur by the accidental ingestion of infected gastropods. We tested the manipulation hypothesis by comparing the vertical climbing behaviour of infected and uninfected individuals of a known intermediate host species, Mesodon sayanus. Snails climbed higher at dawn than during the rest of the day and adults climbed higher than juveniles, but no differences were found between infected and uninfected snails. These results do not explain the observed levels of prevalence, but they improve our understanding of the transmission of P. tenuis, specifically when examining the relative importance of different gastropod intermediate hosts. We review competing hypotheses that might account for the apparent discrepancy in levels of prevalence. Parelaphostrongylus tenuis is a common metastrongyloid nematode of white-tailed deer (Odocoileus virginianus) in eastern North America (Anderson and Prestwood 1981). Like D. dendriticum, it requires the accidental ingestion of infected intermediate hosts for transmission to the final host. In this case, the intermediate hosts are terrestrial gastropods. First-stage larvae (LI) are released with fecal material of the definitive host and terrestrial gastropods are most likely infected during random encounters with Ll that have been washed off the fecal material and that are living in water droplets in the soil (Forrester and Lankester 1997). The natural intensity (number of parasite individuals per infected host) of P. tenuis is typically low (one to five individuals; Lankester and Anderson 1968) and, as there is likely a low probability of gastropods encountering larvae in the environment, most larvae within an intermediate host are probably from the same infection source. Once a gastropod host is infected, at least 27 days are required for Li to develop into infective third-stage larvae (L3) (Anderson 1963); this period can be prolonged depending on temperature and activity (Lankester and Anderson 1968; Halvorsen and Skorping 1982). The parasite is transmitted to a definitive host when the host ingests a gastropod containing L3 (Lankester and Anderson 1968). At least three hypotheses have been proposed to explain the prevalence of P. tenuis as reported in the literature. Of these, at least two, the increasing probability of acquiring infection with age (Anderson and Prestwood 1981) and the manipulation of climbing height of the intermediate host, can be totally or partially discounted; the last hypothesis, that the probability of juvenile deer ingesting infected gastropods is high enough to account for prevalence levels in deer, requires observational and experimental examination. There are at least two more hypotheses that might explain the discrepancy in the prevalence of this parasite in deer and in gastropods. One similar to the manipulation hypothesis also assumes that deer do not accidentally ingest gastropods. If deer select gastropods, there would be a higher probability of transmission occurring. This could be the case if gastropods represent a potential source of a limiting nutrient, such as calcium or sodium, for a foraging deer. If deer actively ingest snails, current prevalence estimates of gastropods could maintain high prevalence in deer. However, we were not able to find any recorded observation of deer eating snails. The other possible explanation is that there is a bias against infected gastropods in samples resulting from common collection techniques (McCoy 1999), so that the prevalence of P. tenuis in gastropod populations has been underestimated. Presently, with the information currently available on the transmission of P. tenuis between intermediate and final hosts, we are unable to distinguish between competing hypotheses. Future work should focus on providing strong falsifications of some of the accumulating alternative hypotheses.