128
T.F. Cross et al.
any deliberate selection and to minimise the length of the period in culture (Cross
et al. 2007). However, domestication selection, an incompletely understood
phenomenon (see Bekkevold et al. 2006), still takes place even in very short periods
of culture and Reisenbichler et al. (2004) argue that it is impossible to totally avoid
genetic modification of animals being reared for stocking.
Breeding of strains for ranching (the preparation of juveniles for deliberate
release into the sea, for growth to harvest size in the wild then subsequent recapture,
usually in a commercial exercise) is a contentious area. Since recapture efforts are
rarely totally successful, some individuals will be free to interact with animals from
wild populations, and thus it might be wise to avoid or minimise anthropogenic
selection. However, economic forces may dictate otherwise. It has been shown with
Atlantic salmon ranching in Iceland that breeding for favourable traits (such as high
return-rate) is feasible, at least in the commercial context (Jonasson et al. 1997).
4.2.2 Between Population Intra-Specific Comparisons
Wild Atlantic salmon exhibit a highly defined population structure (Verspoor et al.
2007), probably resulting from disjunct geographical distribution of freshwater
spawning habitats, propensity for accurate natal homing and typically small
population size relative to many marine species (the latter meaning that genetic drift
has a much more profound effect in promoting structure). There are three major
population groupings (rivers in eastern North America, western Europe and around
the Baltic Sea respectively), but also a high degree of population structure at
regional, and between- and within-river catchment levels among these groupings
(Verspoor et al. 2007). Since there are also strong indications of local adaptation
even between nearby rivers (McGinnity et al. 2004), then the provenance and
particularly, the domestication history, of reared strains will be of major importance
in considering the potential implications of interactions.
Other native marine species that are currently farmed in Europe (cod, halibut,
turbot, sea bass, sea bream, lobster, scallop, mussels and flat oysters) differ from
Atlantic salmon in that genetic population (stock) structure is much less defined
and the extent of local adaptation has not been fully established (Waples 1998;
deWoody and Avise 2000; Conover et al. 2006). It is not clear whether the results
concerning population structure result from lower philopatry in these species or
from exponentially lower genetic drift, or from both-see Bekkevold et al. (2005).
Even though there are much smaller, though often statistically significant, genetic
differences between most groupings than in salmon, there is often evidence of
genetically different population groupings in large geographical areas and of different spawning populations in more local areas, e.g., Atlantic cod (see review by
Imsland and Jonsdottir 2002) appear to differentiate into at least three major
groupings corresponding to the western North Atlantic, eastern North Atlantic and
Baltic respectively (O’Leary et al. 2007). In addition, there is also evidence of
differences between Arctic and coastal cod off Norway (Fevolden and Pogson
T.F. Cross et al.
any deliberate selection and to minimise the length of the period in culture (Cross
et al. 2007). However, domestication selection, an incompletely understood
phenomenon (see Bekkevold et al. 2006), still takes place even in very short periods
of culture and Reisenbichler et al. (2004) argue that it is impossible to totally avoid
genetic modification of animals being reared for stocking.
Breeding of strains for ranching (the preparation of juveniles for deliberate
release into the sea, for growth to harvest size in the wild then subsequent recapture,
usually in a commercial exercise) is a contentious area. Since recapture efforts are
rarely totally successful, some individuals will be free to interact with animals from
wild populations, and thus it might be wise to avoid or minimise anthropogenic
selection. However, economic forces may dictate otherwise. It has been shown with
Atlantic salmon ranching in Iceland that breeding for favourable traits (such as high
return-rate) is feasible, at least in the commercial context (Jonasson et al. 1997).
4.2.2 Between Population Intra-Specific Comparisons
Wild Atlantic salmon exhibit a highly defined population structure (Verspoor et al.
2007), probably resulting from disjunct geographical distribution of freshwater
spawning habitats, propensity for accurate natal homing and typically small
population size relative to many marine species (the latter meaning that genetic drift
has a much more profound effect in promoting structure). There are three major
population groupings (rivers in eastern North America, western Europe and around
the Baltic Sea respectively), but also a high degree of population structure at
regional, and between- and within-river catchment levels among these groupings
(Verspoor et al. 2007). Since there are also strong indications of local adaptation
even between nearby rivers (McGinnity et al. 2004), then the provenance and
particularly, the domestication history, of reared strains will be of major importance
in considering the potential implications of interactions.
Other native marine species that are currently farmed in Europe (cod, halibut,
turbot, sea bass, sea bream, lobster, scallop, mussels and flat oysters) differ from
Atlantic salmon in that genetic population (stock) structure is much less defined
and the extent of local adaptation has not been fully established (Waples 1998;
deWoody and Avise 2000; Conover et al. 2006). It is not clear whether the results
concerning population structure result from lower philopatry in these species or
from exponentially lower genetic drift, or from both-see Bekkevold et al. (2005).
Even though there are much smaller, though often statistically significant, genetic
differences between most groupings than in salmon, there is often evidence of
genetically different population groupings in large geographical areas and of different spawning populations in more local areas, e.g., Atlantic cod (see review by
Imsland and Jonsdottir 2002) appear to differentiate into at least three major
groupings corresponding to the western North Atlantic, eastern North Atlantic and
Baltic respectively (O’Leary et al. 2007). In addition, there is also evidence of
differences between Arctic and coastal cod off Norway (Fevolden and Pogson