7986 - Laserfiche WebLink

Home Browse Search

k 418 Adapf3tion to Captive Breeding pected because the allele is assumed to be associated with reduced fitness in the natural environment. Extremely strong values of selection are of interest for two reasons. First, the observed rapid rates of adaptation to captivity suggest that strong selection is involved. Second, we are primarily interested in those changes that occur in the first few generations of captivity. For similar reasons, we are not interested in the case where the allele favored in captivity is recessive, because it takes many generations for even strong natural selection to increase the frequency of a rare recessive allele. Reducing the variability in family size is extremely effective in retarding the effects of natural selection (Fig. 1). For example, in the case of a dominant allele favored in captivity with twice the fitness of the alter- native allele, the expected allele frequency after five generations under the general selection model is 0.20; with equal family sizes, however, the expected allele frequency is only 0.04 (Fig. 1). It is important to consider how sensitive these results are to the assumption of a single major locus affecting survival in captivity. In a randomly mating population, o 1 5 10 15 5 10 15 20 1 Additive .50 .75 1.00 Dominant .50 1 .00 1.00 0.8 0.6 c: 0.4 - >- U 0.2 c: Q) :J C" Q) .... LL Q) 0.8 ~ ~ 0.6 Additive 70 .85 1.00 Dominant .70 1.00 1.00 0.4 0.2 o o 5 10 15 5 10 15 Generations Figure 1. Relative rate of adaptation to captive con- ditions with the general selection model (upper line) and with the equal family size selection model (lower line). The initial frequency of the allele fa- vored in captivity is 0.01. Relative suroivals (a, b, and c) are given on each graPh. Conservation Biology ADendorf half of the additive genetic variance in a polygenic model is within families and half is between families (Crow & Kimura 1970:240). King (1965) has shown that with additive fitness (that is, when the fitness of the heterozygote is the mean of the two homozygotes) and weak selection, the rate of gene frequency change with familial selection is approximately one-half that with ordinary selection. This is in agreement with the polygenic model. This result is accompanied by two important caveats. First, it holds only for weak selection; and, as discussed before, strong selection is expected to accompany ad- aptation to captivity. With strong selection, the rate of increase in advantageous alleles with familial selection may be less than one-half that with ordinary selection. Second, the expected rate of change is approximately one-half per generation Thus, after several generations, the amount of allele frequency change with familial se- lection will be much less than one-half of that with or- dinary selection. 0.8 Implications Genetic management programs for captive populations often recommend increasing effective population size in order to avert the loss of genetic variation through ge- netic drift. 'Equalizing family size, ascoDSideredh~' often. difficult and expensiy",.;lblJS~~' beenreco~endedOnly:Wilai'the'popWafioii.SiZetno 'SOlidI that genetic drift is of major .concem: The current results indicate that equalizing family size may be a valu- able management tool to retard adaptation to captivity even with large captive populations. For example, the Apache trout (Oncorhynchus apache) is currently listed as threatened under the En- dangered Species Act (David 1990). A single captive population, originated from individuals collected from the wild in 1983 and 1984, is the cornerstone of a re- covery program with a goal of establishing 30 discrete populations of this species within its historic range. Ad- vances in culture techniques and the fecundity of these fish have resulted in a hatchery program that now pro- duces hundreds of thousands for fry per year for re- stocking. Hundreds of mature adult fish are spawned each year. . J;>ublished hatchery genetic guidelines (e.g., Allendorf & Ryman 1987; Kapuscinski & Jacobson 1987) suggest that the effective population size of this stock is proba- bly large enough that the additional expense and effort of equalizing family size is not warranted. However, this conclusion is based on the expected effects of genetic drift assuming selective neutrality. In view of the value and role of this captive population and the potential for adaptation to hatchery conditions, a program. to equal- . ize the family sizes of this population would be desir- able. 0.6 0.4 0.2 0.8 0.6 0.4 0.2 ~ .~ .~ ~ i t ~. "f. !} .;7. ~