534 
Fishery Bulletin 96(3), 1998 
The importance of tissue quality for allozyme stud- 
ies is well documented (e.g. Utter et al., 1987) but is 
particularly critical for crustacean tissue, which de- 
grades more rapidly than finfish tissue. Laboratory 
analysis revealed obvious differences in enzyme reso- 
lution for several loci that resulted in conservative 
allele pooling and loss of some population data. Op- 
timal sample handling may have increased detect- 
able variation. 
Population diversity of C. bairdi 
We detected differences among C. bairdi populations 
of the three major geographic regions: Bering Sea, 
Gulf of Alaska, and Southeast Alaska. Most notably, 
we detected population subdivision within the Bering 
Sea. Samples collected in Bristol Bay (east of approxi- 
mately 162°45'W long.) were statistically different 
from samples collected near the Pribilof Islands (west 
of approximately 167°00'W long.). 
Other biological data support our inference that 
subpopulations of C. bairdi occur within the Bering 
Sea. Using trawl survey data from 1974 onward, Otto 
(1982) reported size-frequency distributions of C. 
bairdi in the Pribilof Islands that were different from 
those in the area north of the Alaska Peninsula, in- 
cluding Bristol Bay. Further, as National Marine 
Fisheries Service (NMFS) trawl assessment surveys 
expanded westward, it became evident that crabs in 
the Pribilof area were larger in size compared with 
those along the continental slope north and west of 
the Pribilof Islands. Somerton (1981) found size dif- 
ferences among large females from the eastern 
Bering Sea seemed to partition into two subareas 
along 167°15'W long. The mean size of adult females 
was quite constant in the eastern subarea (east of 
167°15'W long.) but decreased steadily in the west- 
ern subarea. Although both investigators speculated 
upon the possibility of genetic or environmental fac- 
tors causing these differences, the species has been 
considered a single Bering Sea stock for management 
purposes (Otto, 1982). In our study, we lacked C. 
bairdi specimens from west of approximately 173°W 
long, and thus were unable to test whether crabs 
collected from near the Pribilof Islands differed from 
crabs collected near the continental slope. Although 
there does not appear to be a precise correlation be- 
tween our findings and the observations of Otto 
(1982) and Somerton (1981), all data seem to sug- 
gest that C. bairdi in the Bering Sea may not be com- 
posed of a single panmictic population. 
However, any conclusions we may draw from bio- 
chemical data should be confirmed by other types of 
data. The American lobster ( Homarus americanus ) 
is a case in point. Tracey et al. (1975) noted genetic 
and morphological differences between inshore and 
offshore populations. In tagging studies, Fogarty et 
al. ( 1980) demonstrated limited movement of lobsters 
released at inshore stations, in contrast with exten- 
sive seasonal migrations by lobsters tagged at off- 
shore locations. These studies indicated that despite 
the potential for genetic exchange during seasonal 
mixing, the inshore and offshore populations retained 
their genetic identity. 
The planktotrophic larvae of Chionoecetes may 
spend up to two months in surface waters (Slizkin, 
1990), raising the possibility that larvae released in 
one area may become recruits in another. Further, most 
of what is known or hypothesized about migration of 
Bering Sea C. bairdi is based upon abundance and dis- 
tribution estimates from annual NMFS trawl surveys 
and subsequent fishery captures. 2 Migration studies 
of Chionoecetes in Alaska have been hindered by tag 
losses after molting. Development of a permanent tag 
for Chionoecetes would provide a valuable tool for ex- 
amining population migration and further for evaluat- 
ing the null hypothesis of panmixia. Our findings of 
regional and within-Bering Sea heterogeneity among 
C. bairdi populations merit further investigation. 
Population diversity of C. opilio 
We detected only small differences between Bering Sea 
and North Atlantic C. opilio, primarily in PGM-1* al- 
lele frequencies. The low-frequency alleles of other loci 
observed in Alaskan C. opilio did not contribute sig- 
nificantly to overall differences. Although sampling 
error may have been a factor (Gregorius, 1980) in de- 
tection of rare alleles, no private alleles were found in 
the North Atlantic. The minimal genetic differentia- 
tion among this circumpolar species contrasts with find- 
ings in halibut (Hippoglossus), herring ( Clupea ) and 
cod (Gadus) (Grant, 1987) where significant differen- 
tiation has been described across a similar geographic 
range. The close genetic affinity of the Bering Sea C. 
opilio collections with the collection from the North 
Atlantic and the detection of low-frequency Bering Sea 
Chionoecetes alleles in the North Atlantic suggest re- 
cent or ongoing gene flow. This possibility is supported 
by Garth ( 1958) whose range description includes a dis- 
tribution of C. opilio northward (from the Bering Strait) 
and eastward through the Beaufort Sea and Davis 
Strait to the North Atlantic. This species tolerates very 
low temperatures; the optimum temperature for im- 
mature individuals has negative values even in the 
summer (Slizkin, 1990). Thus, it is plausible to hypoth- 
2 Morrison, R. 1997. Alaska Dep. Fish and Game, RO. Box 
920587, Dutch Harbor, AK 99692. Personal commun. 
