Abstract:
The distribution of lizard parasites, on ten offshore islands and on the mainland in the Wellington region, is used as a baseline, from which to evaluate certain hypotheses of island biogeography. Two endemic hosts, Hoplodactylus maculatus and Leiolopisma nigriplantare maccanni, formed the bulk of 515 lizards examined. Twenty four parasite species were found, of which 14 were undescribed. The distribution of these parasite species on the island localities is represented by the power function, S = CA z, where S is the number of species, A is island area, C and z are constants. However, the semilog model (S = logA + constant) gives a better fit. The power function is used for comparison with other island biogeographic studies, which use the S = CA z model, and biological meanings of the C and z coefficients are reviewed.
Possible biological factors influencing parasite distribution are discussed, and their involvement in the species-area relationship is considered. Lizard diversity (i.e., the number of lizard species per locality) is used as an index of habitat heterogeneity, and this explains more of the variation in the number of parasite species than does area, corroborating Williams' (1964) habitat heterogeneity hypothesis. The number of parasite species is not correlated with isolation indices (i.e., distance to nearest larger landmass or distance to mainland), disagreeing with the predictions of MacArthur and Wilson's (1967) equilibrium theory. Combined island localities harbour more parasite species than does the mainland, emphasising the difficulties in identifying a source area and in choosing isolation indices.
Three theories of island biogeography, MacArthur and Wilson (1967), Williams (1964) and Connor and McCoy (1979), are evaluated in the light of the above findings. Connor and McCoy's passive sampling theory is considered a null hypothesis against which the others are examined. Strong evidence is given against MacArthur and Wilson's theory, and although Connor and McCoy's theory can not be ruled out, the study most strongly supports Williams' hypothesis.
