To gain insight into genetic variation for resistance to parasites, this study assayed clonal variation in cyclically parthenogenetic Daphnia magna with respect to parasitic infection. Samples were collected from natural populations, and the allozyme phenotypes of infected hosts were compared to those of uninfected hosts. Differences between the clonal composition of the infected and uninfected class were evident in only two of 16 populations examined. This result stands in contrast to a study of species in the D. pulex and D. longispina species complexes, where clonal variation for infection was found in 12 of 25 populations (Little & Ebert, 1999). Considering all populations from both studies, associations between host genotype and infection were typically evident only in populations that showed low genotypic diversity and evidence of genetic disequilibria, with D. magna showing the least amount of disequilibria. This pattern is compatible with at least two possibly overlapping hypotheses. First, it may be that those populations lacking clonal variation for infection experienced weaker parasite-mediated selection. We can not rule out variation in selection pressure as an explanation, but found no evidence that the prevalence or intensity of parasitism differed either among species, or between those populations which showed clonal variation for infection and those that did not. Second, it could be that some populations, especially those of D. magna, have more frequent sexual recruitment than others. Sexual recombination breaks up gene combinations which are in linkage disequilibrium, and our method to detect clonal variation for resistance relies on linkage between genetic markers (allozymes) and resistance loci. Past work on Daphnia has shown that the level of sexual recruitment (which is in turn mediated by habitat permanency) is indeed commonly linked to the occurrence of genetic disequilibria. Our results may thus underestimate the prevalence of clonal variation for infection (especially for D. magna), because most of the populations analysed appeared to have high levels of sexual recruitment and therefore lacked the linkage disequilibrium that underlies associations between allozymes and susceptibility.