140
T.F. Cross et al.
equal sex ratio and fitness) than nuclear DNA because of its haploid nature, making
the mitochondrial genome more likely to be affected by genetic drift when population
size is reduced.
The greater the genetic difference between wild and reared morphs, the easier it
will be to establish good levels of discrimination. With Atlantic salmon, because of
the high level of population structure and also because of relatively low N e , it is
usually easy to find discriminatory markers, even when the native population is used
to provide the progenitors of the reared strain (due to the use of limited broodstock
numbers which results in a reduction of genetic variation and alteration in genetic
composition). With other species cultured in Europe, e.g., cod (Hutchinson et al.
2001), lobsters (Triantaphyllidis et al. 2005), the relatively low levels of population
structure observed suggests that it may be much more difficult to find suitable
discriminatory markers. Genes coding for functional proteins (e.g., MHC-Box 4.2)
may be more useful as markers of short term effects than neutral genes, since frequencies may be rapidly changed by the different selection regime experienced in
culture. Identification of hybrids in opportunist situations in species where the level
of discrimination between wild and reared morphs is very low, is not likely to be
easy or even feasible and some type of experimentation may be necessary (see
below). Modelling studies are urgently needed to determine the most appropriate
approach.
4.6.2 Experimental Situations
One of the simplest experimental approaches to investigating the fate of reared
individuals accidentally or deliberately introduced to the wild is by tagging.
Physical tagging has previously been used (e.g., Hansen (2006) investigating the
behaviour of farmed salmon in simulated escapes in Norway) but this method
cannot track offspring, which requires genetic tagging. In the latter approach, a rare
allele at a specific locus is chosen, and two heterozygotes (likely the only genotype
available containing the rare allele, assuming Mendelian autosomal inheritance) are
crossed to produce rare homozygotes (~25%) for release. “Rare” homozygotes
recovered from the wild will most likely be reared individuals, and heterozygotes
(at a frequency above “background”) will be F 1 offspring of wild X reared matings.
However, there are certain limitations with this method;
1. Unless many crosses of heterozygotes are undertaken ab initio, the marked
individuals may show extremely limited genetic variation, and detecting many
heterozygotes to use as parents will be difficult if the allele is rare.
2. There may be functional differences in the fitness of genetically marked and
unmarked individuals. To establish whether there is equivalent fitness in marked
and other reared individuals, tank experiments are often undertaken, where
growth and survival are compared. However, differences that might become
apparent in the much harsher conditions in the wild are unlikely to be observed
T.F. Cross et al.
equal sex ratio and fitness) than nuclear DNA because of its haploid nature, making
the mitochondrial genome more likely to be affected by genetic drift when population
size is reduced.
The greater the genetic difference between wild and reared morphs, the easier it
will be to establish good levels of discrimination. With Atlantic salmon, because of
the high level of population structure and also because of relatively low N e , it is
usually easy to find discriminatory markers, even when the native population is used
to provide the progenitors of the reared strain (due to the use of limited broodstock
numbers which results in a reduction of genetic variation and alteration in genetic
composition). With other species cultured in Europe, e.g., cod (Hutchinson et al.
2001), lobsters (Triantaphyllidis et al. 2005), the relatively low levels of population
structure observed suggests that it may be much more difficult to find suitable
discriminatory markers. Genes coding for functional proteins (e.g., MHC-Box 4.2)
may be more useful as markers of short term effects than neutral genes, since frequencies may be rapidly changed by the different selection regime experienced in
culture. Identification of hybrids in opportunist situations in species where the level
of discrimination between wild and reared morphs is very low, is not likely to be
easy or even feasible and some type of experimentation may be necessary (see
below). Modelling studies are urgently needed to determine the most appropriate
approach.
4.6.2 Experimental Situations
One of the simplest experimental approaches to investigating the fate of reared
individuals accidentally or deliberately introduced to the wild is by tagging.
Physical tagging has previously been used (e.g., Hansen (2006) investigating the
behaviour of farmed salmon in simulated escapes in Norway) but this method
cannot track offspring, which requires genetic tagging. In the latter approach, a rare
allele at a specific locus is chosen, and two heterozygotes (likely the only genotype
available containing the rare allele, assuming Mendelian autosomal inheritance) are
crossed to produce rare homozygotes (~25%) for release. “Rare” homozygotes
recovered from the wild will most likely be reared individuals, and heterozygotes
(at a frequency above “background”) will be F 1 offspring of wild X reared matings.
However, there are certain limitations with this method;
1. Unless many crosses of heterozygotes are undertaken ab initio, the marked
individuals may show extremely limited genetic variation, and detecting many
heterozygotes to use as parents will be difficult if the allele is rare.
2. There may be functional differences in the fitness of genetically marked and
unmarked individuals. To establish whether there is equivalent fitness in marked
and other reared individuals, tank experiments are often undertaken, where
growth and survival are compared. However, differences that might become
apparent in the much harsher conditions in the wild are unlikely to be observed