Analysis of Host-Parasite Cospeciation: Effects of Spatial and Temporal Scale.

Abstract

All species of pocket gophers (Geomyidae) are parasitized by at least one species of chewing louse (Trichodectidae), and recent genetic studies have demonstrated that a pattern of cospeciation exists between these hosts and parasites (Hafner and Nadler, 1988; Hafner et al., 1994). Little work has been done on this host-parasite system at lower taxonomic (intrageneric) levels. Although cospeciation is evident in a study of pocket gophers within the genus Geomys and their chewing lice (Geomydoecus), reticulate evolution and retention of ancestral lineages obscure the pattern (Chapter 2). One key to understanding how these patterns of cospeciation are produced and how they are affected by spatial and temporal scale, is the mode of parasite transmission. The long-standing dogma has been that pocket gophers transmit chewing lice along genealogical, primarily matriarchal, lines. However, mtDNA evidence disproves the hypothesis of strict-maternal transmission of parasites (Chapter 3). Decreasing the scale of the study even further, I examine parasite transmission on a microspatial scale using nuclear-DNA fingerprint data of gophers from a contact zone between two species of chewing lice (Chapter 4). The fingerprint data indicate little or no relationship between genetic relatedness among gophers and the species composition of their respective louse populations. Instead, the species composition of louse populations on individual gophers exhibits significant spatial autocorrelation. Therefore, louse transmission within this zone depends more on spatial proximity (i.e., louse composition on nearest neighbors) than on mating regimes of the hosts. These nearest-neighbor effects have caused louse populations to be distributed in patches of like-species groups. This distribution pattern is fractal-like, and simple models using nearest-neighbor effects and basins of attraction account for the maintenance of the narrow zone of contact between the lice and, by extension, mosaic contact zones in general (Chapter 5).