The Origin and Speciation of Oncorhynchus Revisited

Abstract

Thirty-seven years have passed since Ferris Neave published his classic paper on the origin and speciation of the salmonid genus, Oncorhynchus. Since then, new data on fossils, chromosomes, molecules, and morphology have accumulated, and new analytical techniques for quantifying phylogenetic data have been developed. These new data are reviewed, summarized, and used to reexamine the evolutionary history of the genus. Apparently, Salmo and Oncorhynchus diverged sometime in the early Miocene (∼20 million years ago [mya]) and most of the early speciation (lineage splitting) in the genus occurred during the Miocene. By the late Miocene, or early Pliocene (∼6 mya), members of the chum (O. keta), pink (O. gorbuscha) and sockeye (O. nerka) salmon lineages were present in Idaho and Oregon. Geographically, most of the living members of the earliest divergences (the Pacific trout) are concentrated in western North America near the southern margin of the distribution of Oncorhynchus. Later divergences (the Pacific salmon) probably occurred in the Pacific Northwest and in Asia. Although processes that produced lineage splitting in the past can never be identified with certainty, one can examine genetic divergence in modern species. Thus, I use the sockeye-kokanee salmon divergence as an example of lineage splitting. I conclude that speciation in modern Oncorhynchus usually involves geographic isolation, followed by local adaptation, genetic divergence and, where divergent forms come into secondary contact, competitive interactions among forms. There is no reason to suppose that past divergences were driven by different processes. The evolutionary history of the genus suggests that most local adaptation is ephemeral and that clusters of populations (metapopulations) may be appropriate management units.