7831 - Laserfiche WebLink

Home Browse Search

- Dou.fj IIg et al. acronym code describing the pattern for each enzyme (f~ble 1). Statistical Analysis Levels of genetic diversity were assessed by direct count of haplotypes (alleles) and estimates of haplotypic diver- sity. Number of haplotypes was corrected for sample size. Haplotype diversity (iJ) was defined by Nei and Tajima (1981) as 2 11 (l - Ix. ) , n - 1 where n is sample size and Xi is frequency of the ith hap- lotype in the sample. This value is frequently used to as- sess levels of mtDNA diversity within a sample of individu- als (Avise 1992) and provides a measure of the relative contribution of each haplotype. High values (maximum = 1.0) are found when each genotype is equally repre- sented in the sample; low values (minimum = 0.0) occur when specific genotypes are over-represented. The significance of differences in h values among source and hatchery samples was tested by random re- sampling (modification by T. E. Dowling of the com- puter program HAPLOID written by Weir [1990]). A dis- tribution of diversity values was generated by drawing 1000 samples of 10 individuals from the source popula- tion represented by our sample of 25 individuals from Lake Mohave and calculating diversity for each. Sam- pling of an individual was followed by its immediate re- placement, so an individual could be selected several times or not at all in each replication. Mean diversity val- ues for each set of 1000 resamples and its 95% confi- dence interval were obtained from the distribution of di- versity values (Weir 1990). The effect of broodstock size on 'J and the number of alleles (A) was also examined by computer simulation. The program was modified to sample a highly diverse source population (from 2 to 100 individuals, each pos- sessing a different haplotype) to generate 10,000 prog- eny. Progeny haplotypes were obtained by randomly sampling parental haplotypes (with replacement as de- scribed above). From this model hatchery stock 1000 samples of 10 individuals were removed; haplotype fre- quencies and h were calculated each time. Mean diver- sity values and number of alleles and 95% confidence limits were calculated as described above. h == Results and Discussion mtDNA Variation in the Source and Hatchery Populations Restriction endonuclease analysis of mtDNA from razor- back suckers with eight enzymes produced rvI60 frag- ments. Analysis of all 56 individuals (25 wild and 31 -- mtDNA Diversity ill Razorback Sucker 123 hatchery fish) yielded 27 haplotypes (fable 1). All changes in fragment pattern resulted from restriction site gains and losses, with no evident variation attributable to length differences. The 25 wild-caught individuals from Lake Mohave ex- hibited 17 haplotypes (fable 1). The two most common were in three individuals each. Most (11 of 17) were unique to an individual. Diversity was high (h = 0.97) compared with that of other vertebrates (A vise et al. 1989) and consistent with the high fecundity (rv2000 ova/cm standard length; Minckley et al. 1991) and large population size (Marsh 1994). Analysis of subsamples from three DNFH year classes yielded results in Tables 1 and 2. Unique haplotypes (not found in the Lake Mohave sample) were present in all year classes but were more common in 1987 and 1989 (four and six unique haplotypes, respectively) than in 1990 (one unique haplotype). The 1987 and 1989 year classes possessed levels of variation-relative number of haplotypes and diversity estimates-comparable to those of the source (Lake Mohave), whereas the 1990 year class possessed fewer haplotypes and' a consider- ably lower diversity value. Statistical comparison of h values from hatchery year classes with the source population was provided by re- sampling the Lake Mohave sample (n = 25). The mean diversity value of the 1000 resamples was 0.91, with the 95% confidence interval ranging from 0.80 to 0.98. Only the 1990 year class (h = 0.71) was significantly different from the source population, with only one of the 1000 resamples exhibiting a lower estimate (Fig. 1). This means of testing for significant differences in di- versity between the source and hatchery populations is conservative because our representative sample from the source was relatively small (n = 25). This reflects the fact that even if the source population is maximally diverse, the finite number of parents forces an upper limit on h. An increased source size decreases the prob- Table 2. Characteristics of hatchery samples of razorback suckers and their source (Lake Mohave). Lake HatclJelJ'lislJ ClJaracteristic MolJave 1987 1989 1900 Sample size 25 10 11 10 Number of haplotypes 17 6 7 4 Haplotypes per individual 0.68 0.60 0.64 0.40 Diversity 0.97 0.89 0.91 0.71 Number of females 11 ,650a 55 14 17 Nfeb ? 23.8 32.2 6.9 Fecundityc 0.84d 0.84 1.00 0.94 Survivorship (%) 0 49 46 11 a Assumes 1:1 sex ratio. b Effectil'e /lumber of females. c Number of eggs per female (X 10'). d Based 011 20 ripe females collected ill 1983 -1984 (Millckley et al. 1991). Consc:rvation Biology Volume 10, No. I, Februal)' 1996 ----.-:::::.