Salmon-farming impacts on wild salmon.

C oncerns have been raised about the impact of farmed fish on wild fish through a range of mechanisms including disease and the impact of fish that escape from farms (1). Although Alaska has banned salmon farming and Washington has developed only a very small industry, British Columbia, with an extensive network of protected inlets, has developed a very large salmon-farming industry, now considerably bigger in terms of tons produced than the wild salmon production there. The salmonfarming industry is a significant and growing part of the local economy with the potential to grow to the scale that has occurred in Norway and Chile. In this issue of PNAS, Krkošek et al. (2) present data to show that migrating juvenile salmon are infected by sea lice as they pass salmon farms and that such sea lice infestations cause significant mortality. Their article provides support for those who oppose net-pen aquaculture of salmon within the range of wild salmon, and the results will undoubtedly intensify the debate on the impact of salmon farms on wild fish. To understand the significance of the new results on the public-policy issue, we must consider two questions. First, are there possible flaws in the experimental design that could cause the effect estimated in the article to be exaggerated? Second, what evidence is there for population-level effects of sea lice from salmon farms on wild populations?