Seeb et al : Genetic variation in Panulirus margimtus from the Hawaiian Islands 



717 



the two esterase loci is of particular concern. Is this 

 variability indeed indicative of genetic isolation be- 

 tween the two sampling localities or of changes in the 

 stock structure? 



The data of Shaklee and Samollow (1984) showed 

 that EstD allele frequencies fluctuate in spiny lobster. 

 They detected annual significant fluctuations in allele 

 frequency within localities, but these annual fluctua- 

 tions were parallel among Maro Reef, Necker Island, 

 and an additional locality, Kure Atoll. Thus, the fluc- 

 tuating frequencies did not lead to any significant dif- 

 ferences in allele frequency between localities within 

 years. 



Our data, 7 years later, are somewhat consistent with 

 this pattern for EstD. We detected significant dif- 

 ferences between years, but no significant differences 

 were found between localities sampled the same year. 

 These data support the contention that fluctuations in 

 the northwestern Hawaiian Islands do occur in EstD 

 allelic frequencies. The data are insufficient to warrant 

 invoking selection or drift as an explanation; however, 

 one suitable hypothesis might be that these fluctuations 

 were caused by immigi-ation of a closely related popula- 

 tion differing only at EstD frequencies. 



However, unlike the previous data set, our data show 

 a significant difference between the 1987 Necker Island 

 and Maro Reef collections at EstS. Such change at a 

 single locus could indicate either selection or stochastic 

 processes operating on the EstS allelic frequencies with 

 a lack of gene flow between the two localities. 



No significant differences in levels of heterozygos- 

 ity were detectable. This finding is not surprising. The 

 rate of change in heterozygosity based on random drift 

 is inversely proportional to population size (Hartl and 

 Clark 1989) and is calculated as 



H, = (1 - 1/2 AO' H„ 



where H,, is the original heterozygosity value, H, is the 

 heterozygosity value after / generations, and N is the 

 population size. At most, four generations have passed 

 since the previous sampling. Thus, heterozygosity 

 values should be indistingiiishable with population sizes 

 characteristic of a commercially exploitable species. 

 An analysis of low-frequency or rare alleles is often 

 a more sensitive test to detect changes in variability. 

 We observed a number of low-frequency alleles that 

 were previously detected by Shaklee and Samollow 

 (1984). Gregorius (1980) estimates that a sample size 

 of 754 or greater would be needed to assure a 95% 

 probability of detecting all alleles at a locus with fre- 

 quencies of 0.01 or greater. Sample sizes were large 

 in the earlier study— ranging from 386 to 666— increas- 

 ing the probability of detection of rare alleles. Thus, 

 it was expected that some of the previously detected 



rare alleles would not be observed in the 1987 col- 

 lections, where sample sizes were 200 or less. Our 

 data suggest that there has been no overall reduction 

 in number of rare alleles within the populations; in 

 fact, we detected rare alleles that were previously 

 undetected. 



The unchanged average heterozygosities and the per- 

 sistence of rare alleles indicate that there has been no 

 measurable loss of genetic variability due to fishing 

 pressure. An additional concern in the management of 

 the lobster fishery is whether each bank is primarily 

 a self-recruiting population with minimal larval recruit- 

 ment coming from other banks, or whether larval 

 recruitment at any bank comes from an archipelago- 

 wide gyre so the depletion of the spawning stock at any 

 one bank will not affect recruitment to that bank. The 

 significant difference between the 1987 Necker Island 

 and Maro Reef frequencies at EstS is consistent with 

 the hypothesis that the banks are largely self-recruit- 

 ing, although this locus also appears to have varied 

 significantly over time. Additional genetic studies, 

 possibly including more sensitive DNA markers, are 

 warranted to further elucidate the amount of genetic 

 differentiation between these two important lobster 

 populations. At present, though, a conservative man- 

 agement strategy would be based on the existence of 

 two self-recruiting populations rather than a strategy 

 based on a single panmictic unit. 



Acknowledgments 



We wish to acknowledge the assistance of the Depart- 

 ment of Biological Sciences, University of Idaho, 

 Moscow, Idaho, and University of Washington Friday 

 Harbor Laboratories, Friday Harbor, Washington, 

 where portions of the analysis were conducted. This 

 study was funded by the Honolulu Laboratory, 

 Southwest Fisheries Center, National Maiine Fisheries 

 Service. 



Citations 



Allendorf, F., N. Ryman, and F". Utter 



1987 Genetics and fishery management: Past, present, and 

 future. In Ryman, N., and F. Utter (eds.), Population genetics 

 and fishery management, p. 1-19. Wash. Sea Grant Prog., 

 Ihiiv. Wash. Press. Seattle. 

 Archie. J.W. 



1985 Statistical analysis of heterozygosity data: Independent 

 sample comparisons. Evolution 43:678-68.3. 

 Clayton, J.W.. and D.N. Tretiak 



1972 Amine-citrate buffers for pH control in starch gel elec- 

 trophoresis. J. Fish. Res. Board Can. 29:1169-1172. 

 Gregorius, H-R. 



1980 The probability of losing an allele when diploid genotj'pes 

 are sampled. Biometrics 36:643-6.52. 



