sampling. In some sampling weeks, landings were 

 made during 3 or 4 days while in other weeks there 

 were no landings. Landings were sampled as fish 

 moved from the vessel to the plant over a conveyor 

 system. Approximately 200 specimens were 

 collected for each sample. Specimens were first 

 dissected to determine the sex, then were 

 measured (to the nearest centimeter) from the 

 snout to the fork of the tail. An otolith was 

 removed for age determination and, when time 

 allowed, whole specimens were weighed to the 

 nearest decagram. 



To simplify the collection of otoliths, most 

 otolith samples were stratified by 1-cm body length 

 intervals. Otoliths were taken from five males and 

 five females in each length interval until the 

 sample was exhausted. Although average length- 

 at-age information can be taken directly from 

 such stratified samples, randomization was neces- 

 sary to obtain unbiased estimates of age composi- 

 tion. 



Research vessel samples were collected from 

 1964 through 1969 and were both stratified and 

 random. Most research vessel samples were 

 processed at sea for the same biological data as 

 those taken from commercial samples. When 

 weights were taken a small hand-held steelyard 

 was used which provided more consistent readings 

 than did spring scales. 



The research vessel samples used herein are 

 geographically and temporally restricted simply 

 because it was beyond the capability of a single 

 vessel to conduct more extensive sampling. The 

 samples taken at the reduction plant in 1966-67 

 provided the best temporal coverage over a season, 

 but their areal distribution was restricted mainly 

 to fishing grounds off Grays and Willapa Harbors, 

 Wash. 



Sample Representativeness 



Whereas the nature of available samples places 

 constraints on some aspects of the following study, 

 the large number and size of samples collected 

 over several years and over a large part of the 

 species' geographical range render them valuable 

 in examining the reliability of earlier estimates of 

 growth (Best 1963; Tillman 1968). Also, some of the 

 more conspicuous variations in both growth rates 

 and age composition can be examined. 



Figure 1 gives a general representation of the 

 distribution of sampling effort in 1964-69. The 

 adult portion of the population (4- to 13-yr-olds) 



338 



FISHERY BULLETIN: VOL. 73, NO. 2 



occurring off the coast of Washington during the 

 summer was the most intensively sampled, 

 especially in 1966-67 when commercial samples 

 were taken. Research vessels sampled adults off 

 Washington, Oregon, and California, mostly dur- 

 ing the summers of 1965-67. Juveniles (1- to 3-yr- 

 olds) were sampled only sporadically and much less 

 intensively. In the winters of 1965 and 1968, 

 research vessels searching off southern California 

 and northern Mexico for spawning hake obtained 

 some samples of 1- and 2-yr-old specimens. Very 

 few 3-yr-old hake were captured, probably because 

 there was relatively little sampling effort in areas 

 where they were likely to be most abundant. 



Commercial vessels fishing for Pacific hake off 

 Washington used Cobb pelagic and BCF univer- 

 sal trawls (Johnson and High 1970). The stretched 

 mesh size varied from 5.1 to 7.6 cm in the trawl 

 bodies and cod ends. Research vessels used the 



50« 



130° 



45' 



40' 



35' 



Figure 1. -Distribution of sampling effort for Pacific hake, 

 1964-69. (Darker stipling infers more intensive sampling.) 





