Olsen et a\. An examination of spatial and temporal genetic variation in Theiagia chalcogramma 



761 



an(i 6). Second, two allozyme loci [SOD-2*; MPI*) provided 

 the strongest evidence of regional genetic variation among 

 tlie 1998 samples (Table 6). In fact, the locus SOD-2* 

 had exceptionally high values of 6 for most regional com- 

 parisons in contrast to the other nuclear loci and mtDNA. 

 Diversifying selection may be acting on this locus (e.g. 

 Hudson et al., 1997); however, the general concordance 

 between SOD-2*. MPI*, and mtDNA data suggests that 

 the values of d most likely reflect the influence of genetic 

 drift(Table6. Fig. 2). 



Spatial structure in genetic variation 



North America to Asia The signature of a major east- 

 west division between Asian and North American walleye 

 pollock was evident in our allozyme and mtDNA data but 

 not in our microsatellite data. The results of our F^^ analog 

 tests and goodness-of-fit tests confirm the findings of ear- 

 lier studies (e.g. Iwata, 1975a, 1975b; Grant and Utter, 

 1980; Shields and Gust, 1995, Bailey et al., 1999) and add 

 some new insight regarding genetic heterogeneity in wall- 

 eye pollock from different geographic regions. For example, 

 we confirm the significant SOD-2* variation observed in 

 the past but we also reveal that a second allozyme locus, 

 MPI*, exhibits substantial spatial variation at this geo- 

 graphic scale (Table 6, Fig. 2). Nevertheless, only these 

 two loci and the mtDNA reveal genetic variation between 

 these two regional groups although some of the other 25 

 nuclear loci screened were equally or more polymorphic 

 (e.g. ALAT*, PGDN*, TchIO, Tchll, and Tch22). These 

 results suggest that, despite broad spatial separation, 

 Asian and North American walleye pollock populations are 

 remarkably genetically similar. This outcome supports the 

 notion that the effective population size in walleye pollock 

 is large, the rate of genetic drift is very low, and migration 

 despite physical distance is a strong homogenizing force. 



North America-Bering Sea and Gulf of Alaska The allo- 

 zyme and mtDNA provided discordant results with re- 

 spect to inter- and intraregional population structure with- 

 in North American walleye pollock. The most noteworthy 

 instance was the contradictory evidence of genetic varia- 

 tion between populations from the Bering Sea and the 

 Gulf of Alaska. That is, significant genetic differentiation 

 between populations from these two regions was evident 

 from both the allozymes and mtDNA, but in different 

 years (Table 5). We believe that the intraannual differ- 

 ences between marker classes is an artifact due primarily 

 to the lack of SOD-2* data for the Bogoslof Island sample 

 in 1997. The values of and <Pgj were similar in magni- 

 tude for the 1998 Bering Sea and Gulf of Alaska compari- 

 son, but only was statistically significant (Table 5). This 

 result appears to be due to the influence of SOD-2* for 

 which d was 0.043 and highly significant (P<0.002, Table 

 6). The value of over all allozymes for the 1997 Bering 

 Sea and Gulf of Alaska comparison might have been simi- 

 lar to or larger than the (P^^j for mtDNA if the SOD-2* data 

 had been available for the 1997 Bogoslof Island sample. 



The 1998 samples provided the first indication of statis- 

 tically significant genetic variation among populations in 

 the Gulf of Alaska with both allozymes and mtDNA. Popu- 

 lations from Prince William Sound and Middleton Island 

 were genetically distinguishable from the Shelikof Strait 

 population, but they were not distinguishable from each 

 other. These results are important for two reasons. First, 

 this is the first indication of limited gene flow between a 

 fjord spawning population (Prince William Sound) in the 

 Gulf of Alaska and the numerically dominant Shelikof 

 Strait population which spawns over the ocean shelf 

 Obvious factors that may limit gene flow between these 

 two populations are the differences in spawning habitat 

 preference and the counterclockwise flow of the Gulf of 

 Alaska gyre which prevents larval drift from Shelikof 

 Strait to the Prince William Sound (Bailey et al., 1999). 



