FISHERY BULLETIN: VOL. 85, NO. 3 



batch spawning or rematuration occur in Bering 

 Sea pollock. 



METHODS 



Data and Sample Collection 



Data and specimens of walleye pollock were col- 

 lected from December 1983 to October 1984 by 

 National Marine Fisheries Service (NMFS) ob- 

 servers aboard foreign commercial fishing vessels 

 and by NMFS pesonnel aboard NOAA research 

 vessels. Data collection was divided into three 

 phases. First, fisheries observers logged the time 

 and location of commerical hauls in which spawn- 

 ing walleye pollock (those with running eggs or 

 milt) were observed. A total of 1,538 observations 

 was made between January and October. Second, 

 otoliths were collected from walleye pollock in 

 hauls where spawning was observed. Third, 

 walleye pollock ovaries were collected for histol- 

 ogy and anaylsis of fecundity. Three ovaries per 

 maturity stage (Hinckley 1986) were collected 

 and preserved in 10% neutral buffered formalin 

 for histological analysis. A separate collection of 

 late developing or mature ovaries was made for 

 the fecundity analysis. Five ovaries per 5 cm 

 length interval were collected over the entire 

 length range encountered. These ovaries were 

 preserved in modified Gilson's solution (Ito 

 1977'^). A total of 345 ovaries were collected for 

 histology and 294 for fecundity analysis. 



Data and Sample Processing 



Location of Spawning and Length 

 at Age of Spawners 



Spawning locations of walleye pollock were 

 plotted by month. The distribution of fishing ef- 

 fort by the foreign commercial fleet over the 

 spawning season was examined and compared 

 with locations where spawning was found. Water 

 temperatures and depths of capture at spawning 

 locations were also examined. 



For the length-at-age analysis, five areas were 

 defined within the Bering Sea based on oceano- 

 graphic features (after Lynde et al. fn. 2): the 

 southeast continental shelf, the southeast conti- 



nental slope, the northwest shelf, the northwest 

 slope, and the Aleutian Basin (Fig. 1). Northwest 

 and southeast areas were divided at the Pribilof 

 Islands and buffer zones were defined in order to 

 clearly separate them. Ages were assigned to a 

 maximum of 200 otoliths per area by readers at 

 the Northwest and Alaska Fisheries Center 

 (NWAFC) age and growth laboratory. 



Lynde et al. (fn. 2) found that vv^alleye pollock 

 from the Aleutian Basin and the northwest slope 

 were generally slower growing than pollock from 

 the southeast shelf and slope areas. Based on this 

 observation, R. Francis and A. Hollowed'* classi- 

 fied walleye pollock as "northern" (slow-growing) 

 or "southern" (fast-growing), and developed two 

 corresponding growth curves. In the present 

 study, the growth of walleye pollock from spawn- 

 ing concentrations in different areas was com- 

 pared to the two growth curves described in Fran- 

 cis and Hollowed's unpublished study. The 

 geographical distribution of the two growth types 

 was then examined. 



To derive their "northern" and "southern" 

 growth curves, Francis and Hollowed (unpubl. 

 data) used age data collected by foreign fisheries 

 observers from 1978 to 1983. Age samples were 

 separated into 2° latitude by 1° longitude cells. 

 The mean length at age was estimated for each 

 cell by year, quarter, and sex. 



Estimates of mean length at age for each cell 

 were used to classify the cell as "northern", 

 "southern", or "unknown". The "northern" classi- 

 fication indicated that the distribution of mean 

 length at age in a given cell was similar to that 

 seen by Lynde et al. (fn. 2) in the northwest slope 

 and Aleutian Basin areas (Fig. 1). The "southern" 

 classification indicated that growth was similar 

 to that observed by Lynde et al. (fn. 2) in the 

 southeast slope and shelf areas (Fig. 1). 



For this classification, the von Bertalanffy 

 (1938) growth model was fitted to the weighted 

 mean length-at-age data for the "northern" and 

 "southern" areas for each quarter and sex over a 

 period of 6 years (1978-83) (Lynde et al., fn. 2). 

 The model was fitted using the BMDP PAR 

 derivative-free nonlinear least squares estima- 

 tion procedure (Dixon 1983) to produce predicted 

 mean lengths at age for "northern" and 

 "southern" fish. 



3Ito, D. H. 1977. Fecundity of the copper rockfish, Sefcasfe.s 

 caurinus (Richardson), from Puget Sound, Washing- 

 ton. Unpubl. manuscr. Northwest and Alaska Fisheries 

 Center, National Marine Fisheries Service, NOAA, 7600 Sand 

 Point Way N.E., Seattle, WA 98115. 



^Robert Francis and Anne Hollowed, Northwest and Alaska 

 Fisheries Center, National Marine Fisheries Service NOAA, 

 7600 Sand Point Way N.E, Seattle, WA 98115, pers. commun. 

 January 1985. 



482 



