By A.J. Gharrett, W.W. Smoker
Fisheries Research, Volume 18, Issues 1-2, Biological Interactions of Natural and Enhanced Stocks of Salmon, October 1993, Pages 45-58, ISSN 0165-7836, DOI: 10.1016/0165-7836(93)90039-A.

Salmon hatcheries, like other resource management practices and tools, potentially have genetic effects on wild-spawning populations of salmon. These effects, which theory predicts will erode vital genetic diversity, would be expected to operate through straying and gene introgression, and through other processes. Genetic diversity among populations is well known from analysis of neutral biochemical traits and is less well known from observation of polygenic, ecologically adaptive, phenotype differences. We note increasing evidence of adaptively important polygenic genetic diversity within populations and evidence that this variability is partitioned temporally or spatially among distinct segments of salmon populations, a partitioning we call infrastructure. For example, in one well-studied small population, Auke Creek pink salmon, there is evidence of genetically based variability of timing of anadromous migration and simultaneous evidence of the importance to survival of that timing. We believe that the adaptedness and productivity of salmon stocks are dependent on genetic infrastructure; fisheries management practices, including enhancement of harvests by hatcheries and ocean ranching, potentially reduce genetic infrastructure as they may reduce other levels of genetic diversity. Rational resource management should seek to conserve genetic diversity at all levels. Treatment of salmon stocks as homogeneous units, neglecting within-stock diversity or infrastructure, will not be adequate to conserve fitness and productivity of these commercially valuable resources.

More details...