(1) To describe the species-area relationships among communities of Plasmodium and Haemoproteus parasites in different island populations of the same host genus (Aves: Zosterops). (2) To compare distance-decay relationships (turnover) between parasite communities and those with potential avian and dipteran hosts, which differ with respect to their movement and potential to disperse parasite species over large distances.
Two archipelagos in the south-west Pacific, Vanuatu and New Caledonia (c. 250 km west of Vanuatu) and its Loyalty Islands, with samples collected from a total of 16 islands of varying sizes (328-16,648 km2).
We characterized parasite diversity and distribution via polymerase chain reaction (PCR) from avian (Zosterops) blood samples. Bayesian methods were used to reconstruct the parasite phylogeny. In accordance with recent molecular evidence, we treat distinct mitochondrial DNA lineages as equivalent to species in this study. Path analysis and parasite lineage accumulation curves were used to assess the confounding effect of inadequate sampling on the estimation of parasite richness. Species-area and species-distance relationships were assessed using linear regression: distance-decay relationships were assessed using Mantel tests.
Birds and mosquito species and Plasmodium lineages exhibited significant species-area relationships. However, Plasmodium lineages showed the weakest 'species-area' relationship; no relationship was found for Haemoproteus lineages. Avian species richness influenced parasite lineage richness more than mosquito species richness did. Within individual avian host species, the species-area relationship of parasites showed differing patterns. Path analysis indicated that sampling effort was unlikely to have a confounding effect on parasite richness. Distance from mainland (isolation effect) showed no effect on parasite richness. Community similarity decayed significantly with distance for avifauna, mosquito fauna and Plasmodium lineages but not for Haemoproteus lineages.
Plasmodium lineages and mosquito species fit the power-law model with steeper slopes than found for the avian hosts. The lack of species-distance relationship in parasites suggests that other factors, such as the competence of specific vectors and habitat features, may be more important than distance. The decay in similarity with distance suggests that the sampled Plasmodium lineages and their potential hosts were not randomly distributed, but rather exhibited spatially predictable patterns. We discuss these results in the context of the effects that parasite generality may have on distribution patterns.
- Distance-decay relationship
- island biogeography
- New Caledonia
- species-area relationship
- SPECIES-AREA RELATIONSHIP
- GEOGRAPHICAL DISTANCE
- MALARIA PARASITES
- LESSER ANTILLES
- SAMPLING EFFORT