122 



Fishery Bulletin 93(1). 1995 



et al., 1987; Natanson, 1990). As a slow-growing, 

 long-lived species, the dusky shark may have over- 



lapping lengths at age which may obscure length 

 modes and bias the estimates of model parameters 

 (Rosenberg and Beddington, 1987; Shepherd et al., 

 1987). The vertebral method is therefore considered 

 the more robust method and the length-frequency 

 parameters are used for comparison only. 



Yoccoz (1991) has brought up questions as to the 

 validity of judging biological significance based on 

 statistical tests. He suggests that statistical signifi- 

 cance is not necessarily indicative of biological sig- 

 nificance; this appears to be the case with the dusky 

 shark. The statistically significant differences shown 

 between male and female dusky shark vertebral 

 growth curves may not reflect biological differences. 

 Examination of the length-at-age data suggests that 

 biologically the differences between male and female 

 vertebral curves are small. The age and size at ma- 

 turity differ by only two years and five centimeters 

 for males and females (Table 4). Females are pre- 

 sumed to grow ultimately to a larger size than males. 

 This means that either growth slows in males after 

 maturity or that males do not live as long as females. 



The vertebral VBGF derived in this study is very 

 similar to the curve attained by Hoenig (1979) for 

 combined sexes for the ages under consideration 

 (birth to 33 years) but is different from data presented 

 by Lawler (1976) and Schwartz (1983). Hoenig's 

 ( 1979) parameter values for the VBGF have a slightly 

 higher L ro and t and lower K than parameters de- 

 rived from vertebral analysis in the present study 

 (Table 2). Lawler (1976), using vertebral analysis to 

 determine the age of female dusky sharks, obtained 

 VBGF-parameter values markedly different from the 

 present study (Table 2). The L x in his study is more 

 than twice as large as the L x reported here and his 

 lvalue suggests a much slower growth rate. These 

 two factors combine to make Lawler's (1976) curve 

 appear as a straight line from birth to 34 years. 



