SOMERTON and OTTO: DISTRIBUTION AND REPRODUCTION OF GOLDEN KING CRAB 



3.0 -, . 3.0 



54 55 56 57 58 59 60 



700 



0.7 



0.7 



55 



56 57 51 



LATITUDE 



59 60 



100 300 500 700 

 DEPTH 



Figure 5.— Catch per hour, by sex, and the proportion of males of golden king crab are shown as a function 



of latitude (left panels) and depth (right panels). 



and highly significant in both years (P < 0.01), but 

 the depth coefficient for females, although negative 

 in both years, was significant (P < 0.05) in only one. 

 Although male CPH decreases significantly with 

 depth whereas female CPH decreases significantly 

 with latitude, CPH is not a strict linear function of 

 depth and latitude; therefore, linear approximations 

 mask aspects of the variability. The important point 

 is that both male and female CPH generally increase 

 with an increase in latitude or a decrease in depth, 

 but at more southerly latitudes or at the shallowest 

 depth, male CPH is considerably higher than female 

 CPH (Fig. 5). 



Different trends in CPH between sexes suggested 

 that the sex ratio of golden king crab varied spatial- 

 ly. To investigate this further we examined the 

 variation in proportion of males within trawl hauls 

 having at least five crabs. The proportion of males, 

 based on combined 1981 and 1982 observer data, 

 is shown by latitude and depth in Figure 5. When 

 the proportion of males was regressed against 



latitude and depth (using weights equal to the num- 

 ber of crabs within each trawl haul), the latitude 

 coefficient was negative and highly significant (P 



< 0.01) in both years; and the depth coefficient, 

 although negative in both years, was significant (P 



< 0.05) in only one. 



From a biological perspective, the latitudinal 

 decrease in the proportion of males is difficult to ex- 

 plain; therefore, we considered possible sampling 

 bias that could lead to an apparent change in the 

 proportion of males. Since males are considerably 

 larger than females in the central area but nearly 

 the same size as females in the northern area, the 

 proportion of males might vary with latitude due to 

 size selectivity of the trawls. This hypothesis was 

 tested by comparing the proportion of males be- 

 tween the northern and central areas considering 

 only crabs within an equal size range. To eliminate 

 a possible confounding effect due to a sexual differ- 

 ence in growth rate that begins at maturity, we 

 restricted the comparison to crabs <90 mm. Based 



579 



