Problems with species: patterns and processes of species formation in salamanders1

How many species one recognizes within a given taxon remains a difficult question, especially when morphology is relatively stable or when clinal variation is present, thus complicating diagnosis. I accept the general lineage concept of species, and my goal is to recognize historically distinct evolutionary lineages that are likely to remain distinct. Here I analyze this task with respect to patterns of species formation in two genera of plethodontid salamanders in California. Ensatina is a ring species complex surrounding the Central Valley of California. At present it is a single species with seven subspecies that are linked by apparent clinal variation in intergrade zones, but there are also some narrow hybrid zones where morphologically and ecologically differentiated forms interact. In contrast, Batrachoseps, which has much the same distribution, has about 20 species in California, most occurring in sympatry with Ensatina. Divergence in the two taxa is based on two fundamentally different phenomena, and yet there are some common themes. Adaptive divergence in coloration is the dominant theme in Ensatina, whereas differentiation is largely perceived at the molecular level in Batrachoseps. Yet both have evolved in the same region and have been affected by many of the same climatic and earth historical phenomena. Within the Ensatina complex, different adaptations related to predator avoidance have evolved. Coloration has diverged in different directions in coastal and inland populations, even though genetic interactions continue to take place. Where coastal populations meet other coastal populations, ecologically and morphologically similar populations merge genetically, even if well differentiated in molecular traits. In contrast, where the ring is crossed and where ecologically and morphologically differentiated populations meet, they hybridize narrowly or are sympatric and behave as if they are species. Within the ring-like distribution, clinal patterns of variation occur. The current polytypic taxonomy is retained, even though it is problematic, because alternatives are even less appropriate. In contrast, where genetically differentiated populations of Batrachoseps meet they typically do not merge. Instead, they replace one another spatially, in part because they are so similar ecologically. Apparently the periods of isolation were sufficiently long that even in the absence of adaptive divergence there has been divergence of isolating mechanisms. Analysis of patterns of genetic differentiation in allozymes and mtDNA in relation to the geological history of California is used to generate biogeographic scenarios to help explain the contrast between Batrachoseps and Ensatina.