FISHERY BULLETIN: VOL. 82, NO. 4 



classes. This pooling was necessary because many 

 of the alleles were extremely rare. The distribu- 

 tion of genotypes at each locus in each sample was 

 examined for internal consistency with the 

 Mendelian inheritance model by chi-square tests 

 of observed genotype ratios with those expected for 

 a single random mating population in the absence 

 of differential selection among alleles. The ex- 

 pected ratios were calculated from observed allele 

 numbers using Levene's (1949) unbiased method 

 for small samples. The heterozygosity at each 

 locus (h ) was calculated as h = 1 - IX, where X, is 

 the frequency of the ith allele. Average heter- 

 ozygosity (H) was calculated as the unweight- 

 ed arithmetic mean of A (the individual locus het- 

 erozygosity) over all loci examined. A locus was 

 considered polymorphic if the frequency of the 

 most common allele was <0.99. 



Three levels of analysis were employed to 

 examine genetic differentiation (stock heter- 

 ogeneity) among the population samples. First, 

 for each polymorphic locus, contingency chi- 

 square tests of all possible pairwise combinations 

 of localities were conducted. Second, contingency 

 tests comparing pooled samples representing the 

 main Hawaiian Islands and the Northwestern 

 Hawaiian Islands were conducted for all loci. Fi- 

 nally, large samples of spiny lobsters collected 

 from three localities allowed the analysis of year- 

 to-year variability within and among these 

 localities. 



RESULTS 



Forty-six presumed gene loci encoding 28 dif- 

 ferent enzyme systems were surveyed for genetic 

 variation in Panulirus marginatus . Thirty-nine of 

 these were either monomorphic in the first 100 or 

 more animals screened or exhibited only rare var- 

 iants (P < 0.01) These enzymes which were not 

 studied further were aspartate aminotransferase 

 (2 loci), acid phosphatase (2 loci), adenylate 

 kinase, alkaline phosphatase (3 loci), alanine 

 aminotransferase, arginine kinase, esterase (3 

 loci), glyceraldehyde-3-phosphate dehydrogenase, 

 glycerol-3-phosphate dehydrogenase, glucose-6- 

 phosphate dehydrogenase, hexokinase, isocitrate 

 dehydrogenase (2 loci), lipoamide dehydrogenase 

 (= diaphorase), lactate dehydrogenase (2 loci), 

 malate dehydrogenase (2 loci), malate 

 dehydrogenase-NADP + (= malic enzyme) (2 loci), 

 monophenolmonooxygenase (= tyrosinase), 

 naphthyl amidase (= leucine aminopeptidase), 

 peptidase (3 loci), peroxidase (2 loci) pyruvate 



kinase, superoxide dismutase (2 loci), and 

 triosephosphate isomerase (2 loci). The remaining 

 seven loci were polymorphic (frequency of the most 

 common allele <0.99 in at least one population) 

 and the conditions for their analysis are sum- 

 marized in Table 1. Thus, P. marginatus exhibits a 

 P M = 0.152. 



TABLE 1. — Polymorphic enzymes in the spiny lobster, Panulirus 

 marginatus: Characteristics and conditions for analy- 

 sis. M = muscle, DG = digestive gland, V = vertical starch 

 gel, H = horizontal starch gel. 



'Presumed structure based on enzyme banding pattern in heterozygotes 

 (see text). 



2 EDTA-bonc acid-Tris pH 8 6 butter of Boyer et al. (1963). 



3 Tris-citric acid pH 7.0 buffer; buffer 1 of Shaw and Prasad (1970) 



discontinuous lithium hydroxide pH 8.1 butter (slightly modified buffer 2 of 

 Selander et al 1971). 



Continuous lithium hydroxide pH 8. 1 buffer (modified buffer 2 of Selander et 

 al. 1971 with stock "B" of gel buffer replaced with water). 



Zymogram patterns for 6 of the 7 polymorphic 

 enzymes are shown in Figure 2. The allele fre- 

 quencies, heterozygosity per locus, and number of 

 alleles successfully scored for each polymorphic 

 locus at each locality are presented in Table 2. 

 Although all seven loci in Table 2 are polymorphic 

 at the 0.99 level in at least one sample, only two 

 loci [Mpi (mannosephosphate isomerase) and 

 Pep-1] exhibit a per locus heterozygosity of >0.1. 

 As a result, the heterozygosity (H) averaged over 

 all loci is only 0.021 for this species. 



Several features of the data summarized in 

 Table 2 warrant explanation. First, as mentioned 

 in Materials and Methods, many of the allelic 

 classes summarized in this table are heteroge- 

 neous containing two or more rare alleles. For 

 example, the Est-3 (esterase) "fast" (Est-3 f ) class 

 contains alleles Est-3 105 and Est-3 103 , the 

 "medium" class consists of allele Est-3 100 only, and 

 the "slow" class contains alleles Est-3 97 , Est-3 95 , 

 and Est-3 90 . Similar groupings were carried out 

 for several other loci as indicated in Table 2. The 

 allelic classes for Mpi, Pep-1, and Pep-2 were 

 not detectably heterogeneous. Examples of 

 phenotypes expressing several of the rare alleles 

 contributing to these pooled allelic classes are 

 shown and identified in Figure 2. Second, because 



H9fi 



