FISHERY BULLETIN: VOL. 82, NO. 4 



allele frequencies, heterozygosity per locus, F gT 

 per locus, and the number of genes successfully 

 scored for each polymorphic locus at each locality 

 are presented in Table 3. Glucosephosphate 

 isomerase-B (Gpi-B), lactate dehydrogenase-C 

 (Ldh-C), and L-iditol dehydrogenase (Iddh) ( = 

 sorbitol dehydrogenase) each exhibited only the 

 two alleles shown in Figure 1 and Table 3. Alcohol 

 dehydrogenase (Adh) exhibited one additional al- 

 lele ("very slow") in one heterozygote. This rare 

 allele was pooled with the "slow" allele in the 

 analysis. Similarly, umbelliferyl esterase (Umb) 

 exhibited one additional allele ("very slow") in one 

 heterozygote. This rare allele was pooled with the 

 "slow" allele in the analysis, x 2 tests of goodness of 

 fit to Hardy- Weinberg expectations of genotype 

 distributions revealed two significant deviations 

 (out of a total of 25 tests). These were a heterozy- 

 gote deficiency for Ldh-C at Necker Island (x 2 ! = 

 8.17; P < 0.005), and a heterozygote excess for 

 Umb at Hawaii (x 2 x = 3.95; P < 0.05). Since one 

 significant outcome in 20 tests is expected by 

 chance given an« = 0.05, this is not surprising. 

 The per locus heterozygosity ranged from 0.293 for 

 the Gpi-B locus to 0.495 for the Umb locus. The 

 average heterozygosity across all loci (H) was 

 0.047. 



A comparison of the allele frequencies at each 

 locus across the five localities (Table 3) revealed 

 that there was little overall variation among loca- 

 tions. x 2 contingency tests of all possible pairwise 

 comparisons of localities (10 comparisons x 5 loci 

 = 50 tests) yielded only one significant value, that 

 for Iddh, Necker Island vs. Molokai (x 2 x = 5.99; P 

 < 0.025). The apparent homogeneity was even 

 more evident in x 2 tests of the two pooled groups 

 (NWHI vs. main) where no significant outcomes 

 occurred in the five tests. Three other measures 

 of genetic characteristics of these samples also 

 failed to reveal subpopulation differentiation. 

 The number of common alleles per polymorphic 

 locus was two at all localities for all enzymes. The 

 average heterozygosity per locus (H) showed al- 

 most no difference (Maro Reef// = 0.045, French 

 Frigate Shoals H = 0.047, Necker Island H = 

 0.048, Molokai H = 0.047, Hawaii H = 0.046) 

 among localities. Finally, the values of Wright's 

 F ST were small for all five polymorphic loci and 

 the meanF ST was only 0.005. Overall, none of the 

 measures used indicated subpopulation or 

 stock heterogeneity in this species in the Hawai- 

 ian Islands. 



It could reasonably be argued that the above 

 analyses of pooled collections consisting of fish of 



TABLE 3. — Allele frequencies, heterozygosity values, and FsT val- 

 ues at five polymorphic loci in pink snapper, Pristipomoid.es filarnen- 

 tosus. Localities as in figure 1 of Shaklee and Samollow (1984). 



'Adh = alcohol dehydrogenase, Gpi = glucosephosphate isomerase, Iddh = 

 L-iditol dehydrogenase ( = sorbitol dehydrogenase), Ldh = lactate dehydrogenase, 

 Umb = umbelliferyl esterase, h = per locus heterozygosity, FsT = Wright's (1965) 

 fixation index. 



2 f = fast, s = slow, n = number of genes successfully scored, W = number of 

 individuals analyzed. 



Consisting of fish from Maro Reef, French Frigate Shoals (FFS), and Necker 

 Island. 



■"Consisting of fish from Molokai, Hawaii, and 25 specimens from Kauai. 



706 



