The parasitic nematode Strongyloides ratti has a complex life cycle. The progeny of the parasitic females can develop into three distinct morphs, namely directly developing infective third-stage larvae (iL3s), free-living adult males and free-living adult females. We have analysed of the e¡ect of host immune status (an intra-host factor), environmental temperature (an extra-host factor) and their interaction on the proportion of larvae that develop into these three morphs. The results are consistent with the developmental decision of larvae being controlled by at least two discrete developmental switches. One is a sexdetermination event that is a¡ected by host immune status and the other is a switch between alternative female morphs that is a¡ected by both host immune status and environmental temperature. These ¢nd-ings clarify the basis of the life cycle of S. ratti and demonstrate how such complex life cycles can result from a combination of simple developmental switches.

Factors constraining the evolution of host-specificity were investigated using a gastrointestinal parasitic nematode, Strongyloides ratti. S. ratti is a natural parasite of rats which can also reproduce, with decreased success, in laboratory mice. Observed host-specificity arose from lower establishment, reduced per capita fecundity and more rapid expulsion of parasites from mice relative to rats. Variation in the efficacy of thymus-dependent immunity between host species (rats and mice) was insufficient to explain the majority of the observed differences in parasite establishment and reproductive success. The role of natural selection in determining host-specificity was addressed using experimental selection followed by reciprocal fitness assays in both host species. Experimental selection failed to modify the host-specificity of S. ratti to any measurable degree, suggesting either a lack of genetic variation for this trait or the involvement of as yet unidentified factors underlying the differences in S. ratti fitness in rats and mice respectively. These results are discussed in relation to competing theoretical models of ecological specialization, host immunology and previous attempts to experimentally alter the host-specificity of parasitic nematodes.