Abstract. Understanding the phylogeography of a species requires not only elucidating patterns of genetic structure among populations, but also identifying the possible evolutionary events creating that structure. The use of a single phylogeographic test or analysis, however, usually provides a picture of genetic structure without revealing the possible underlying evolutionary causes. We used current analytical techniques in a sequential approach to examine genetic structure and its underlying causes in the bogus yucca moth Prodoxus decipiens (Lepidoptera: Prodoxidae). Both historical biogeography and recent human transplantations of the moth's host plants provided a priori expectations of the pattern of genetic structure and its underlying causes. We evaluated these expectations by using a progression of phylogenetic, demographic, and population genetic analyses of mtDNA sequence data from 476 individuals distributed across 25 populations that encompassed the range of P. decipiens. The combination of these analyses revealed that much of the genetic structure has evolved more recently than suggested by historical biogeography, has been influenced by changes in demography, and can be best explained by long distance dispersal and isolation by distance. We suggest that performing a suite of analyses that focus on different temporal scales may be an effective approach to investigating the patterns and causes of genetic structure within species.Key words. Biogeography, isolation by distance, mismatch, nested clade, population structure, yucca moth.Received February 19, 2002. Accepted May 23, 2002 Avise (2000) defines phylogeography as ''. . . a field of study concerned with the principles and processes governing the geographic distributions of genealogical lineages' ' (p. 35). Phylogeography is the melding of biogeography and the evolutionary history of a lineage or groups of lineages. Early investigations used the methodology of phylogenetics to relate evolutionary history to the geographic distribution of populations within a lineage. The resulting phylogenetic tree of haplotypes was overlain on the current geographic structure of populations to make inferences about both historical and current patterns of population subdivision. This approach has and continues to work well for lineages that have been separated for long periods of time. In such lineages, the accumulation of genetic divergence translates into a phylogenetic signal that may correspond well with geographic separation.As an increasing number of studies have shown, the methodology of phylogenetics often lacks resolving power and may even obscure the evolutionary relationships of lineages that are relatively recent or have experienced demographic variation such as population bottlenecks or expansions (Crandall et al. 1994;Crandall and Templeton 1996;Smouse 1998). For these lineages the use of a bifurcating tree, whether parsimony or distance-based, may be misleading, especially when the ancestral haplotypes are extant. The use of haplotype networks will...