7827 - Laserfiche WebLink

Home Browse Search

134 MINCKLEY ET AL. ficial habitats (Marsh, in press), bonytails stocked in Lake Mohave in 1981 could have attained sizes by 1986 comparable to smaller wild individuals (about 40 cm TL), and by 1988 they could have grown well into the size range of most wild fish (> 40 to about 60 cm TL). Six of 45 loci were polymorphic in naturally produced FZ specimens available for study. The distribution of electromorphs at each polymor- phic locus agreed with Hardy-Weinberg expecta- tions (Table 2). Allozymic variation was com- bined maximally as shown by high probabilities of random associations of allelic products from the sample (e.g., exact probability of 1.0 at five of six variable loci). Of greater importance, however, was the magnitude of this variation. Wild popu- lations of bonytails are reduced to such low num- bers that an accurate measure of genetic diversity in nature may not be possible now. Comparisons with other species thus provide our only perspec- tive, and these have limitations. Comparative data of genetic variation in natural populations exist for many fish species (Nevo 1978) and include many for those species designated as threatened or endangered (Echelle, in press). However, com- parisons should be made among closely related taxa (e.g., cyptinids) and ideally among congeners. Furthermore, meaningful interspecific compari- sons of heterozygosity should be based on ho- mologous loci (Selander 1976), a factor often overlooked. We examined data for congeners whose homologous loci have been scored and found that only arroyo chub Gila orcutti were ap- propriate for comparison (Crabtree and Buth i 981). Bonytails and arroyo chub have 29 loci scored in common. The two species share comparable levels of variation except mean heterozygosity was TAet~ 3. - Estimates of genetic variability in bonytails (N = 24) and arroyo chub (N = 8; Crabtree and Buth 1981). The direct comparison of these two species is based on the products of 29 loci scored in common. Num- Proportion of heterozygotes ber Proportion (mean ± SE) of of polymor- locia phic locib Observed° Expectedd Bonytail 45 0.133 0.031 ±0.015 0.032±0.016 24 0.138 0.027±0.0[7 0.026±0.017 Arroyo chub 29 0.138 0.065±0.038 0.06 f ±0.030 a Each locus had 1.2 ± 0. I alleles (mean ± SE). b A locus was considered polymorphic i(more than one allele was detected. ~ Direct count of heterozygotes. e Unbiased estimate based on Hardy-Weinberg expectations (Nei 1978). about twice as great for arroyo chub (0.061 ± 0.030) as for bonytails (0.026 ± 0.017; Table 3). Our comparison was limited, however, and the arroyo chub that we used may be atypical. Avise (1977) surveyed heterozygosity among represen- tatives of many North American cyprinids and reported that western genera including Gila ex- hibited amean value of 0.038 ± 0.008. Thus, our sample of bonytails exhibits heterozygosity slight- ly below the mean for related forms. As additional data become available for other species of Gila, more accurate assessments of this variation will be possible. Relatively high genetic variability in hatchery- produced FZ fish is accredited to Hamman's (1982) technique for the production of a brood stock from five pairs of wild-caught bonytails. Eggs of each female were fertilized by sperm of a different male TABLE 2.-Genotypic distributions at six polymorphic loci in a sample of 24 bony[ails from Dexter National Fish Hatchery, New Mexico. Allelic products (genotypes) are lettered in order of increasing anodal electrophoretic mobility. Genotype or test' used sAat-A Ca[-1 Est-2 Locus Gpi-B Iddh-A Pgm-A Distribution as 0 6 0 0 22 II ab I IO 4 5 2 10 bb 23 8 20 l9 0 2 ac 0 0 0 0 0 I Prababilityb Exact test 1.00 0.43 1.00 1.00 1.00 1.00 Chi-square test 1.00 0.37 0.70 0.61 0.88 0.93 a Swofford and Selander (1981). b Likelihood that each genotypic array represents one randomly breeding population