508 
Fishery Bulletin 96(3), 1998 
When the four samples were pooled and tested as one 
population, none of the 30 tests of polymorphic loci de- 
viated from expected Hardy- Weinberg proportions. 
Average heterozygosity ranged from 0.135 to 0.142 
(unweighted mean=0.136, Table 7) Percentage of 
polymorphic loci at the P Q 95 level ranged from 32.4% 
to 35.1%, and the mean number of alleles per locus 
ranged from 2.2 to 2.4. Seventeen unique, rare (fre- 
quency <0.01) alleles were identified; two each in 
15 20 25 30 35 40 45 50 
Length (cm) 
Figure 5 
Estimated weight-length curves for Atka mackerel by fishery 
management area (541=eastern Aleutian, 542=central Aleutian, 
543=western Aleutian Islands, and 610=western Gulf of Alaska) 
based on data collected from resource assessment surveys of 
the Aleutian Islands and Gulf of Alaska during 1993 and 1994, 
respectively. 
Umnak Island and Attu Island, four in Kiska Island, 
and nine in Seguam Pass. Correlation of these alle- 
les with sample size was 0.99. 
Allele frequencies were similar for all samples for 
all polymorphic loci. The largest pair-wise allele fre- 
quency difference was 11.7% between Umnak Island 
(n- 59) and Seguam Pass (n= 96) for ADA *100 (Table 
3). 5 Only one of the 14 chi-square contingency tests 
for the P Q 95 loci was statistically significant ( AH-2 *, 
P=0.004). The total chi-square was not statisti- 
cally significant (P=0.08). Nei’s unbiased genetic 
distance values ranged from 0.0000 to 0.00017 
for the four samples. The mean F gT for the 30 
polymorphic loci was 0.004. 
Discussion 
Growth and electrophoretic analyses 
Analysis of variance showed that length-at-age 
and weight-at-age data were statistically dif- 
ferent among the four management areas 
tested. In order to illustrate these differences, 
the age-length data were fitted to the commonly 
5 The power of this single locus pair-wise comparison is 
fairly high. For example, the probability of detecting a 
frequency difference of 0.15 between P 2 =0.40 and P 0 = 
0.55 is 80% for samples consisting of 93 individuals each — 
similar to these samples sizes (Sokal and Rohlf, 1981). 
To be 90% certain of detecting such a difference, a sample 
of 122 individuals is needed. 
Table 3 
Parameter estimates from the von Bertalanffy growth equation (L m , K, t Q ), and length-weight relationship (a, b), standard errors 
of the parameter estimates, mean square errors (MSE), and sample sizes (n) for the western Gulf of Alaska, eastern Aleutian, 
Central Aleutian, and Western Aleutian Islands. 
Eastern Central Western 
Gulf of Alaska Aleutian Islands Aleutian Islands Aleutian Islands 
Standard Standard Standard Standard 
Estimate 
error 
Estimate 
error 
Estimate 
error 
Estimate 
Error 
von Bertalanffy model 
L 
46.63 
0.911 
47.25 
1.219 
44.02 
0.735 
40.93 
0.618 
K 
0.869 
0.173 
0.39 
0.064 
0.404 
0.037 
0.461 
0.053 
^0 
0.965 
0.182 
-0.041 
0.389 
-0.094 
0.183 
-0.05 
0.247 
MSE 
5.98 
5.36 
3.44 
6.52 
n 
95 
94 
163 
158 
Length-weight model 
a 
0.859E-05 
0.452E-05 
0.237E-03 
1.103E-04 
0.878E-04 
0.316E-04 
0.652E-04 
0.190E-04 
b 
3.132 
0.137 
2.212 
0.123 
2.46 
0.098 
2.55 
0.08 
MSE 
0.01 
0.01 
0.01 
0.01 
n 
95 
94 
163 
158 
