Francis and Duffy Length at maturity in three pelagic sharks 



495 



was 274 cm and the longest immature 

 female was 300 cm. Median maturity 

 was estimated by probit analysis to be 

 280.1 cm (CI 267.5-292.9 cm), but sam- 

 ple sizes were very small over the tran- 

 sitional range (Fig 5). The nonoverlap 

 of the CIs between males and females 

 showed that median maturity differs 

 significantly between the sexes. 



Blue shark 



The relationship between CLI and FL 

 was essentially linear in blue sharks, 

 and no apparent inflections were evident 

 (Fig. 7). The smallest male with sperma- 

 tozeugmata was 164 cm; 50% of males 

 contained spermatozeugmata by 194 cm, 

 and 100% by about 260 cm. 



Samples of male blue sharks examined 

 by us were small (Table 2). Maturation 

 occurred over a wide length range: the 

 smallest mature male was 157 cm and 

 the largest immature male was 237 cm 

 long. The direct estimate of median 

 maturity was correspondingly variable 

 (192.1 cm, CI 178.1-206.3 cm) (Fig. 7). 



The UWI increased abruptly above 

 about 170 cm in some sharks, all of 

 which were pregnant (Fig. 8). Other non- 

 pregnant sharks up to about 220 cm FL, 

 which were presumably subadults, had 

 UWIs less than 2%. Pregnant females 

 ranged from 166 to 252 cm (mean 203 

 cm) (Fig. 8). 



Only 26 females were available for di- 

 rect maturity estimation (Table 3). The 

 smallest recorded mature female was 

 142 cm, but this seems exceptional; the 

 next smallest was 172 cm. The longest 

 immature female was 185 cm. The num- 

 ber of sharks in the maturation length 

 range was inadequate for determining 

 median maturity (Table 3), although we 

 have shown the probit analysis ogive in 

 Figure 8. 



Discussion 



Maturity estimates 



To be sexually mature, a male shark must be able to 

 produce viable sperm and have the means to deliver 

 them to a female. Similarly, females must be able to 

 produce viable eggs and nourish the developing embryos 

 through to parturition. An assessment of the degree 

 of development of all parts of the reproductive system 

 and the presence or absence of reproductive products 

 is the best way to determine sexual maturity. We used 



this approach to score the maturity status of individual 

 sharks and thereby derive direct median maturity esti- 

 mates. However, the sample sizes available for this 

 approach were sometimes small, and confidence limits 

 ranged from unrealistically low (because of lack of over- 

 lap of immature and mature sharks) to high; therefore it 

 was not possible to rely entirely on these estimates. 



We supplemented our direct maturity estimates with 

 measurements or assessments (made by observers) of 

 some key components and products of the reproductive 



