Kimura et a!.: Stock structure and movement of Anopiopoma fimbria 
477 
Length (cm) 
Figure 9 
(A) Length frequencies from trap index surveys, plotted by areas, showing the dichotomy 
between Alaska (areas 17 and 18) and west coast (areas 20-27) sablefish. The two distribu- 
tions shifted to the right are from Alaska. (B) The same length frequencies but only for sable- 
fish with length greater than 60 cm. Top distributions are from Alaska (areas 17 and 18), the 
middle distributions are from the west coast north (areas 20-24), and the lower distributions 
are from the west coast south (areas 25-27). 
migrating from Alaska to the U.S. west coast appear 
to concentrate in the northern area of the west coast 
(areas 20-22, Table 6). 
Methot 6 recognized from fishery data that west 
coast sablefish varied greatly in size at age. He sug- 
gested that this was due to a commingling of south- 
ern California and Alaska “morphs.” Extending this 
hypothesis, tag recoveries (Table 6) suggest that area 
20 contains a substantial number of fish from Alaska, 
but that areas farther south contain substantially 
fewer migrants. Similarly, length frequencies of large 
fish from the NMFS trap index survey (Fig. 9B) sug- 
gest three populations: Alaska fish from areas 17 and 
18; a mixture of Alaska and smaller southern Cali- 
fornia fish from areas 20-24; and small southern 
California fish from areas 25-27. 
On the basis of size and age-at-depth data, adult 
sablefish are believed to inhabit greater depths as 
they grow older owing to a protracted, ontogenetic 
migration to greater depths (Fujioka et al., 1988; 
Saunders et al., 1997). Norris (1997) disagreeing with 
this view, conjectured that the broad bathymetric 
range was due to radiative evolutionary adaptation 
of enzyme systems to greater depths. Norris (1997) 
regarded the age-at-depth data as an artifact caused 
by varying size-selective fishing mortality with depth. 
He felt that reproducing populations inhabit vari- 
ous depth zones. Sigler et al. ( 1997 ) hypothesized that 
6 Methot, R. D. 1993. Latitudinal and bathymetric patterns 
in sablefish growth and maturity off the U.S. West coast. Paper 
presented at the international symposium on the biology and 
management of sablefish, Alaska Fisheries Science Center, April 
13-15, 1993. 
Table 8 
Parameter estimates and their statistical significance for 
a tag-recovery growth model intended to detect a growth 
effect due to ENSO events. Data were divided between 
Alaska and the west coast and the ENSO parameter was 
included when a particular tagged fish was at liberty dur- 
ing any ENSO event. The model used was S 2 = S j + A t exp 
< A?nso + Pi^i + A>A), where S 1 was the size of fish at the 
time of tagging, S 2 was the size of fish at time of recovery, 
A, was the time elapse between tagging and recovery, and 
Penso represents an effect from any exposure to ENSO 
events. Both release and recovery were in the same region 
for the data used. 
Parameter Estimate SE f-value 
Alaska modeling results. 
Degrees of freedom for t-statistic=6,030. 
P 0.05781 2.2200 x 10" 2 
/3j -0.08307 4.8603 x 10" 4 
P 2 -0.00031 9.9375 x 10“ 6 
West coast modeling results. 
Degrees of freedom for t-statistic=3,666. 
P -0.12918 3.2513 x 10" 2 
P x -0.09636 7.4322 x 10" 4 
P 2 -0.00036 1.4059 x 10“ 5 
2.604 
-170.910 
-31.093 
-3.973 
-129.651 
-25.853 
age-at-depth data could be explained by considering 
the movement of sablefish to greater depths as be- 
ing a random walk (i.e. resulting from a sequence of 
random movements). From this point of view, greater 
age-at-depth could be viewed as an ontogenetic phe- 
nomenon resulting from statistical behavior. 
