Evolution of sockeye salmon ecotypes.

Abstract

Rapid evolution of reproductive isolation is proposed by A. Hendry and colleagues to be the cause of the genetic differences between two adjacent populations of Washington State sockeye salmon (Reports, 20 Oct., p. [516][1]). These two populations or “ecotypes,” one living in a tributary (Cedar River) and the other along a beach (Pleasure Point) of Lake Washington, are suggested to have evolved after introductions of sockeye into Lake Washington from Baker Lake between 1937 and 1945 (within about 13 generations) and in spite of high estimated migration rates. Hendry et al. 's conclusions, however, depend on several assumptions. First, they assume that the hatchery population used for the introductions was genetically homogeneous. However, the hatchery population, in fact, included both beach and river ecotypes (as well as other gene pools) in its ancestry in various proportions in the five brood years used for the introductions ([1][2]). Although some mixing of the ecotypes undoubtedly occurred in the hatchery, natural segregation based on different spawn times could have maintained much of the diversity between them until the time of the introductions. Therefore, the current diversity of ecotypes in Lake Washington, and any reproductive isolation, could reflect divergence much older than 13 generations. Another assumption the authors make is that the “beach immigrants” are derived from the Cedar River population, on the basis of temperature-sensitive otolith (a calcareous ear stone) banding patterns. However, some or all of these immigrants could have come from other beach-spawning sites in Lake Washington that have temperature profiles similar to that of Cedar River ([2][3]). 1. [↵][4]1. R. Rathbun , Rep. U.S. Comm. Fish Fisheries 25, 251 (1900) J. Kemmerich, “A review of the artificial propagation and transplantation of the sockeye salmon of the Puget Sound area in the State of Washington conducted by the federal government from 1896 to 1945” (U.S. Fish and Wildlife Service, Leavenworth, WA, 1945);R. G. Gustafson et al., NOAA Tech. Memo NMFS-NWFSC-33 (1997) (available at ). [OpenUrl][5] 2. [↵][6]1. J. G. Woodey , (1996) thesis, University of Washington(. # Response {#article-title-2} We reported that two new salmon populations within Lake Washington have diverged in response to natural selection and are now partially reproductively isolated. Gustafson and colleagues suggest some alternative scenarios. First, they point out that both beach and river ecotypes contributed to the initial Baker Lake hatchery stock used for introductions into Lake Washington. This is true, but, as we have discussed elsewhere ([1][7]), the hatchery propagated all (or nearly all) of the fish as a single mixed stock for up to eight generations. As a result, any preexisting genetic differentiation would have been homogenized before the introductions. Furthermore, salmon did not colonize Pleasure Point Beach until after the introductions from Baker Lake had ceased ([2][8]), suggesting that the Pleasure Point population was secondarily derived from the population of Cedar River. Reproductive isolation, therefore arose de novo and in situ. A mix of genes from beach and river ecotypes in the introduced group would simply have provided genetic variation to facilitate rapid divergence. Second, Gustafson et al. suggest that some of the Pleasure Point immigrants could have come from other beach populations. This is unlikely because the immigrants were genetically similar to Cedar River fish, and because most of the beach-spawners at the time of our study were at Pleasure Point. In shoreline surveys by boat, we found only five spawning pairs elsewhere. Third, they propose that the observed genetic differences could have resulted from founder effects. Although possible, this is unlikely, given the observed high rates of adult movement (any founder effect would be rapidly destroyed by gene flow unless reproductive isolation had evolved). The alternatives offered by Gustafson et al. are contradictory: they suggest that genetic differences were present before the introduction and then suggest that genetic differences arose in situ. In conclusion, our explanations remain the most consistent with the data. Although our study was the first to document the evolution of reproductive isolation on such a short time scale, we expect that other studies will reveal analogous situations. 1. [↵][9]1. A. P. Hendry, 2. T. P. Quinn , Can. J. Fish. Aquat. Sci 54, 75 (1997). [OpenUrl][10][CrossRef][11][Web of Science][12] 2. [↵][13]1. J. G. Woodey , (1966) thesis, University of Washington(. [1]: /lookup/doi/10.1126/science.290.5491.516 [2]: #ref-1 [3]: #ref-2 [4]: #xref-ref-1-1 View reference 1 in text [5]: {openurl}?query=rft.jtitle%253DRep.%2BU.S.%2BComm.%2BFish%2BFisheries%26rft.volume%253D25%26rft.spage%253D251%26rft.atitle%253DREP%2BUS%2BCOMM%2BFISH%2BFISHERIES%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [6]: #xref-ref-2-1 View reference 2 in text [7]: #ref-3 [8]: #ref-4 [9]: #xref-ref-3-1 View reference 1 in text [10]: {openurl}?query=rft.jtitle%253DCan.%2BJ.%2BFish.%2BAquat.%2BSci%26rft.volume%253D54%26rft.spage%253D75%26rft_id%253Dinfo%253Adoi%252F10.1139%252Fcjfas-54-1-75%26rft.genre%253Darticle%26rft_val_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Ajournal%26ctx_ver%253DZ39.88-2004%26url_ver%253DZ39.88-2004%26url_ctx_fmt%253Dinfo%253Aofi%252Ffmt%253Akev%253Amtx%253Actx [11]: /lookup/external-ref?access_num=10.1139/cjfas-54-1-75&link_type=DOI [12]: /lookup/external-ref?access_num=A1997XB78600008&link_type=ISI [13]: #xref-ref-4-1 View reference 2 in text