Is Ensatina eschscholtzii a ring species?
Herpetologica, 54(2) 1998
Stebbins (1949) regarded the western North American plethodontid salamander Ensatina eschscholtzii as a polytypic species with seven subspecies distributed in a ring around the Great Valley of California. The terminal subspecies overlap in southern California. This complex has been considered a classical example of a circular overlap or "ring species." In a series of studies on the molecular evolution of this complex, David B. Wake and his associates concluded that the hypothesis that Ensatina eschscholtzii is a single ring species either could not be rejected or is supported by their molecular data. I reexamined their data and conclude that the data do not support the ring species hypothesis. Instead, the data suggest an alternative hypothesis: the E. eschscholtzii complex is a superspecies, with at least 11 groups that are most likely parapatric and allopatric species and semispecies. Four of these are cryptic forms. Evidence favoring this hypothesis is as follows. (1) Allozyme data indicate extraordinarily large genetic distances between most groups (Nei D values as high as 0.77), indicating that some groups probably have been differentiating independently of each other since the late Miocene or Pliocene epochs. Thus they would be regarded as different species by those systematists favoring the evolutionary or phylogenetic species concepts. (2) There is considerable concordance between geographic patterns of variation in coloration, allozymes, and cytochrome b sequences. This often is observed between species but rarely between freely interbreeding subspecies. (3) Patterns of geographic allozyme variation indicate a small amount of genetic differentiation within most groups (Nei D < 0.16), typical of that found within other vertebrate species whose populations are interbreeding throughout their ranges. (4) Patterns of geographic differentiation among groups reveal narrow zones where parapatric taxa come into contact. There sometimes is abrupt, coincident change in both coloration and allozyme frequencies in these zones, with a variable amount of hybridization, typical of species' or semispecies' contacts in other animals. Thus they would be regarded as different species by most systematists favoring the biological species concept, except for those that restrict the concept to species that never hybridize with any other taxon. (5) Cytochrome b sequences indicate a low amount of sequence divergence within most groups (usually <4%), typical of within-species comparisons, while sequence differences between groups are larger (5-17%), typical of between-species comparisons in other non-avian vertebrates. The pattern of geographic variation in cytochrome b is mostly concordant with that in allozymes although, in one case, introgression of allozymes appears to be more extensive through the hybrid zone than the maternally inherited cytochrome b haplotype. (6) Considerable reproductive isolation has evolved between the only two pairs that are sympatric (xanthoptica and northern platensis; eschscholtzii and klauberi), and there is yet no molecular evidence of hybridization between central oregonensis and northern platensis at the only other (parapatric) contact between the blotched interior forms and the unblotched coastal forms. (7) The hypothesis that most of the differentiation throughout both the coastal and inland chains of populations is due to gradual differentiation correlated with geography is not supported by the molecular data. I suggest a different hypothesis: that allopatric speciation occurred during long periods of isolation among the highly differentiated groups of this complex.
More information available on Ensatina