Chilton: Maturity and growth of Sebastes variabilis in the central Gulf of Alaska 
71 
Table 1 
Summary of seasonal data and collection method for central Gulf of Alaska female dusky rockfish ( Sebastes variabilis) samples 
used to estimate reproductive maturity, gonadosomatic index, and growth parameters. Sample collections: trawl = bottom trawl; 
hook-and-line. 
Port sample collections 
Chartered vessel collections 
Trawl survey collections 
February 2000 
March 2000 
July 2000 
n= 4 
7i = 9 (trawl) 
n = 1 
7i = 8 (hook-and-line) 
January 2001 
April 2000 
January 2001 
77=15 
77 = 3 (hook-and-line) 
77=27 
February 2001 
November 2000 
May 2001 
n = 3 
77 = 36 (trawl) 
77 = 18 
March 2001 
April 2001 
June 2001 
n=20 
77=14 (trawl) 
77 = 5 
77=30 (hook-and-line) 
assessment model to determine the fishing quota for 
this rockfish species. Maturity estimates and growth 
parameters derived from this study could improve the 
estimation of spawning stock biomass and the stock 
recruitment relationship of future dusky rockfish stock 
assessments. 
Materials and methods 
Samples for the maturity estimates were randomly 
collected from three sources: 1) port sampling of com- 
mercial jig and trawl fisheries around Kodiak Island, 
Alaska; 2) hook-and-line and bottom trawl samples from 
chartered vessels around Kodiak Island; and 3) bottom 
trawl sampling during the Gulf of Alaska groundfish 
trawl survey conducted by the National Marine Fish- 
eries Service (NMFS) in 2001. Samples were collected 
from February through April, June, and November of 
2000, and from January through June of 2001 (Table 1). 
The majority of the samples were collected locally from 
nearshore waters on the southeast side of Kodiak Island. 
Specific catch locations were not available for many of 
the samples because of the confidential nature of the 
catch reporting system; however, Alaska Department 
of Fish and Game statistical areas were reported with 
each collection and are used here to show generalized 
collection locations (Fig. 1). 
Sagittal otoliths and ovaries were collected from each 
individual female rockfish, as well as fork length (FL) 
to the nearest mm, total body weight (W) to the near- 
est gram, and individual ovary weight to the nearest 
gram. Otoliths were aged according to standard break 
and burn procedures (Chilton and Beamish, 1982). 
Tissue samples used for histological maturity exami- 
nation were taken from the middle of the right ovary 
and fixed in a 10% neutral buffered formalin solution. 
Tissue was taken from the left ovary if the right ovary 
was torn because it has been determined in previous 
histological studies of rockfish maturity that there is 
no difference in oocyte development between the left 
and right ovary (Shaw, 1999). These tissue samples 
taken at all macroscopic stages, except stage 5 (eyed 
larvae), were embedded in paraffin, thin sectioned to 
7 pm with a rotary microtome, mounted on slides, and 
stained with standard hematoxylin and eosin (Sheehan 
and Hrapchak, 1980). Ovary samples with eyed larvae 
were fragile and not amenable to the paraffin and sec- 
tioning process. 
A compound microscope and ocular micrometer were 
used to measure the diameter of the oocytes. The di- 
ameter of the fifth largest oocyte on a randomly select- 
ed transect along the histological cross section of the 
gonad was used as the decisive factor for evaluating 
the most advanced nonatretic oocyte and to determine 
the developmental stage of the ovary (West, 1990). 
This method eliminated the possibility of a single and 
largest oocyte misrepresenting the development stage 
of the ovary. These ovary stages were evaluated on 
the basis of seven maturity stages and corresponding 
oocyte development previously described for northern 
rockfish ( S . polyspinis) (Chilton, 2006) (Table 2). Ova- 
ries with vitellogenic oocytes developed to the migra- 
tory nucleus stage or the presence of postovulatory fol- 
licles were used as evidence of a mature ovary sample. 
Nichol and Pikitch (1994) found darkblotched rockfish 
(S. crameri) oocytes in the early vitellogenic stage (de- 
scribed as secondary yolk formation) with high levels 
of atresia (resorption of the unfertilized oocyte) and 
classified females with ovaries developed to this stage 
as functionally immature. 
Age and length at 50% maturity (A 0 5 and L 0 5 ) were 
determined by fitting a logistic function to the maturity 
data as a function of age or length with generalized lin- 
ear modeling generated by S-plus statistical software, 
and then evaluating the fitted model at a maturity 
proportion of 0.50 (S-plus, vers. 6.2; Insightful Corp., 
Seattle, WA). The variance of A 0 5 and L 0 5 , as well 
as 95% confidence intervals (Cl), were estimated by 
bootstrapping methods (Efron and Tibshirani, 1993). 
