This paper addresses the genetic consequences of aquaculture on natural fish populations. The study is motivated by rapidly increasing numbers of intentionally and accidentally released fish and is based on empirical observations reported in the literature. A wide variety of outcomes, ranging from no detectable effect to complete introgression or displacement, has been observed following releases of cultured fish into natural settings. Where genetic effects on performance traits have been documented, they always appear to be negative in comparison with the unaffected native populations. These findings are consistent with theoretical considerations of the implications of elevated levels of gene flow between cultured and locally adapted natural populations; they raise concerns over the genetic future of many natural populations in the light of increasing numbers of released fish. Strategies for the genetic protection of native populations from the effects of aquaculture are outlined including more secure containment, the use of sterilized fish, and modifying the points of rearing and release. We recommend strong restrictions on gene flow from cultured to wild populations and effective monitoring of such gene flow.

Various population parameters and physiological, behavioral, morphometric, meristic, calcareous, biochemical, and cytogenetic characters have been used to identify fish stocks. We define a stock as an intraspecific group of randomly mating individuals with temporal or spatial integrity. Each character set and the associated methodology relates to specific aspects of the stock definition. Population parameters are useful primarily for the recognition of putative stocks at the practical fisheries management level. Physiological and, to some degree, behavioral characters are used primarily to study differences in the adaptation of stocks to different environments. Behavioral characters are also important for the recognition of stocks and the study of their spatial and temporal discreteness. Morphological characters, including morphometric measurements, meristic counts, and the shape, size, and type of zonation in calcareous structures provide data that are useful for the precise description of and differentiation among stocks. Although the genetic control of this type of variation is poorly understood, multivariate methods coupled with shape analyses provide techniques that describe intraspecific subdivisions that have been found to correspond to genetic stock structure as determined by other methods. Intraspecific chromosomal variation has, on occasion, been employed for stock identification. This variation has had only limited application to the study of stocks because of complications arising from intraindividual variation and artifactual variation introduced by the methodology. Electrophoresis provides an important method for measuring the genetic discreteness of stocks and for the study of genetic relationships among stocks. Electrophoretic data have recently attained a primary position among the methods used for stock identification.Key words: stock identification, genotype, phenotype, population parameters, marking, physiological, behavioral, morphometric, meristic, calcareous, cytogenetic, and biochemical characters

No abstract

The amount of genetic differentiation between stocks of Atlantic cod (Gadus morhua) was estimated from electrophoretically detectable protein loci expressed in skeletal muscle and liver. Variant alleles at 13 of these loci were detected among nine samples covering most of the species range: North America, Greenland, Iceland, Barents Sea, Norwegian coastal waters, the North Sea, and the Baltic Sea. A very low amount of genetic differentiation was observed among stocks. Only two loci (LDH-3 and PGI-1) showed large statistically significant heterogeneity of allele frequencies between samples. Standard genetic distances (Nei) between these samples based on 10 variable loci ranged from 0.00015 to 0.01072 with Baltic cod as the genetically most divergent sample. There was a highly significant correlation between genetic and geographic distance for the samples from the Atlantic suggesting that a substantial gene flow has occurred between these stocks with geographic distance being an inhibiting factor. The relatively larger genetic distance between the Baltic cod and other stocks may reflect isolation resulting from geographic and perhaps ecological barriers. However, the absolute amount of genetic differentiation in the Atlantic cod appears to be very low throughout its range.

Compatible data from 11 polymorphic allozyme loci were assembled from 14 published studies and additional unpublished information for 232 native collections of brown trout sampled over a broad geographical range. Most of the genetic variation was explained by the distribution of allelic variation at the LDH-C* and CK-A1* loci. Patterns of geographical distribution coupled with postglacial geological events supported a model of colonization from three preglacial lineages, each fixed for different sets of alleles at these two loci. During glacial retreat, recolonization is proposed to have occurred mainly into adjacent areas through (i) a northwestern migration from an eastern Mediterranean-Caspian refuge, (ii) a northern expansion from a refuge in Atlantic drainages of Iberia and southern France, and (iii) a northern and eastern radiation from a refuge centred near the English Channel. Extant populations in deglaciated areas are suggested to represent mixed or pure descendants of these migrating groups. Repopulation from a fourth Mediterranean refuge distinguished by the presence of the LDH-A2*100QL allele was excluded based on the absence of this allele in repopulated areas.

A genetic investigation of anadromous trout populations in the Puget Sound area revealed numerous juvenile individuals from two streams with electrophoretic phenotypes consistent with those expected for hybrid descendents of steelhead trout (Salmo gairdneri) and coastal cutthroat trout (S. clarki clarki). The likelihood of hybridization was evaluated with a hybrid index measuring the relative probability that the combined genotype for a particular fish at several diagnostic loci could have arisen by random mating within each of the two Salmo species. The distribution of hybrid index scores among fish from the two creeks clearly demonstrated the genetic distinctness of the two species and the intermediate genotypic composition of the unknown fish. We concluded that these electrophoretically intermediate fish were natural steelhead–cutthroat hybrids based on their restricted occurrence at specific sample sites in only 2 of 23 streams surveyed, the linear distributions of juveniles from the two parental species within each stream, and the distribution of hybrid index values for a hatchery population of known mixed ancestry. Further, from estimates of gametic disequilibria and the absence of a consistent excess of heterozygotes we suggest that backcrossing may have occurred. The existence of these natural hybrids raises many questions concerning the biological bases for maintaining species integrities in regions of sympatry and indicates the need to fully understand the biological consequences of present and future management practices.

This paper traces the development of applications of biochemical genetic methods to problems of fishery management over a period of four decades. In the 1950s, details of presumed genetic structuring of fish species appeareddestined for revelation through Mendelian characters identified by immunogenetic procedures. In the 1960s, immunogenetic methods were displaced by protein electrophoresis, with a proliferation ofreports of genotypic and allelic data for protein-coding loci. In the 1970s, disagreement about the biological significance of protein polymorphisms delayed acceptance of management applications of this variation. In the 1980s, management applications included identification of relationships among populations, analyses of mixed stock fisheries, and uses in fish culture, conservation biology and forensics. The complementary relationship between proteinelectrophoresisandnucleicacid technologiesisstressed, with aplea torecognize the unique attributes of properly applied protein electrophoresis in fishery management.

Abstract– We used six polymorphic microsatellite loci to investigate a potential metapopulation system in bull trout (Salvelinus confluentus) from five spawning localities in the Lightning Creek drainage, a tributary to Lake Pend Oreille, Idaho. The number of spawners as estimated by redd counts is low in all populations sampled. Analytic viability models indicate that local isolated populations of these sizes are unlikely to persist. We tested two hypotheses: (1) these are remnant populations that are vulnerable to local extinction, or (2) these populations are interconnected by migration and are being maintained at larger effective sizes than indicated by the redd counts (i. e. metapopulation dynamics). All populations within the Lightning Creek basin are significantly differentiated (P > 0.005), and the allele frequencies appear to be stable among temporally separated subsamples within locations. It is therefore unlikely that extensive dispersal has linked tributaries in a manner consistent with a metapopulation structure. The low number of spawning individuals combined with the degree of isolation indicated by the genetic data suggest that extinction of the tributary populations is probable if temporal variability and small size is prolonged. However, these populations contain amounts of genetic variation similar to populations throughout the range of bull trout. Therefore, if the apparent demographic trends are reversed, these populations may recover without suffering the detrimental effects of a severe bottleneck.