Stark: Temporal and spatial variations in maturation and growth of female Gadus macrocephalus 



399 



Lj. and age data (Venables, 1997). Between- 

 sex and between-area differences in growth 

 were tested by first fitting the von Berta- 

 lanffy model with a term distinguishing sex 

 and area and again tested for the combined 

 sexes and areas. The likelihood ratio of the 

 two models was then determined for each 

 category (Kimura, 1980). Significance of the 

 likelihood ratio was tested to determine if 

 growth differed by sex or area. 



The gonadosomatic index (7^) was calcu- 

 lated from the specimens sampled for ma- 

 turity as the ratio of gonad weight (Wq) to 

 body weight (W) with the gonads removed 

 (/G = 100 WJW). The body weight portion of 

 the data included food contents for the Gulf 

 of Alaska specimens, whereas stomachs were 

 emptied for all 250 Bering Sea specimens 

 that were weighed. However, weight of the 

 stomach contents represented less than \'7c 

 of the total body weight for the Bering Sea 

 specimens. A z-test (Sokal, 1969) was per- 

 formed with S-Plus software to determine if 

 there were between-area and between-month 

 differences in 7^. 



Table 2 



Length-at-maturity results based on samples of ovaries (n) collected 

 between the late prespawning and initial spawning period for female 

 Pacific cod iGadus macrocephalus) in the Gulf of Alaska (GOAl and 

 Bering Sea (BS) areas, by date of collection. The parameters of the 

 logistic equation that were used to fit the data are given: B (slope of the 

 line) and A (F intercept), variance (the square of the standard devia- 

 tion of B and A), covariance (the product of the standard deviations of 

 B and A and the coefficient of correlation between them), length (mm) 

 at which 50Vc of females were expected to reach sexual maturity (L,p^^), 

 and variance of Lj. . 



Sampling 

 statistics 



n 



B 



A 



Variance (B) 



Variance (A) 



Covariance (B,A) 



TT 



Variance (LT^f^) 



Gulf of Alaska 



January 1999 

 and 2004 



157 

 0.0222 



-11.1425 



2.0649 



6.5528 



-0.0002 



502.5543 



212.5733 



Bering Sea 



February 

 2003 



133 



0.0132 



-7.6248 



1.3807 



3.2433 



0.0001 



579.9599 



309.3380 



Results 



The results provided information on the rela- 

 tionship between ovary growth and matura- 

 tion for Pacific cod in the Gulf of Alaska and 

 Bering Sea. The Iq data were available for 

 95% of the Gulf of Alaska specimens and 

 99% of the Bering Sea specimens sampled for 

 maturity. For both areas, the 7g cycled accord- 

 ing to the season (Fig. 2). The observed Iq was 

 highest during the winter months (Jan=4.24 

 Iq and February=8.48 IqJ, declined through 

 the spring (March=6.37 Iq and April=3.50 

 IG), and reached the lowest levels in early 

 summer (June = 1.02 Iq). Maturation of the 

 population of oocytes to be spawned the fol- 

 lowing year occurred at a slow rate through 



October (1.74 7g). 



Similarly linked to the Pacific cod spawn- 

 ing cycle, was the development of oocytes 

 within the ovaries (Fig. 3), which was consis- 

 tent in all quadrants of each ovary. By Janu- 

 ary, over 70% of the sampled ovaries were 

 in the vitellogenesis stage of development in 

 preparation for spawning, in both the Gulf 

 of Alaska and Bering Sea Pacific cod. Fe- 

 males that had vitellogenesis-stage oocytes 

 during the spring were expected to develop 

 and spawn those oocytes that year. By June, all stocks 

 of vitellogenesis-stage oocytes had been exhausted. In 

 the Gulf of Alaska, over half of the female Pacific cod 

 spawned during April and June. Pacific cod spawning 

 began during February in the Bering Sea area, and 



Month of sample collections 



Figure 2 



Mean gonadosomatic index of female Pacific cod (Gadus macrocepha- 

 lus). Sample sites were located in the Gulf of Alaska (GOA) during 

 January 1999 and 2004, April and June 1999, and October 1998 

 (total n = 327). and the Bering Sea (BS) during January, February, 

 and March 2003 (total ;!=261). The 95'7f confidence intervals are 

 represented by bars located adjacent to each data point. 



approximately 10% of the fish population participated. 

 During October, 5% of female Pacific cod had vitello- 

 genesis-stage ovaries, whereas 95% of Gulf of Alaska 

 cod were in earlier stages of maturation. The likeli- 

 hood of female Pacific cod spawning was significantly 



