144 



Fishery Bulletin 100(1) 



ment of the oviducal gland, ovary, and uterus. 

 Specimens were considered juvenile if they had 

 undeveloped sexual organs, filiform uteri, and 

 no vitellogenic activity in their ovaries. Pre- 

 ovulatory females had relatively larger ova- 

 ries with orange vitellogenic follicles, but no 

 uterine eggs. Ovulating females had uterine 

 eggs and ripe ova still in the ovary, whereas 

 in gravid specimens ovulation was complete. 

 Postpartum females showed similar-size ova- 

 ries, oviducal glands, and ovarian follicles as 

 those of gravid specimens, but had flaccid uteri 

 that were still slightly enlarged (compared 

 with other nongravid females) indicating re- 

 cent parturition. Individuals that had simi- 

 lar-size uteri as those of pre-ovulatory and 

 ovulating individuals, but smaller ovarian fol- 

 licles with little or no vitellogenic activity were 

 termed "resting" females. Male maturation was 

 evaluated according to development of the tes- 

 tes and claspers. Individuals with relatively 

 short, flexible claspers, and filiform ampullae 

 of the ductus deferens were considered juve- 

 niles. Adults were characterized by elongate 

 and calcified claspers and relatively large epi- 

 didymides (compared with juveniles). 



Statistical analyses 



Size-frequency distributions of males and fe- 

 males were compared by using a two-sample Kol- 

 mogorov-Smirnov test (P=0.05). Chi-squared 

 goodness-of-fit tests were used to examine the 

 hypothesis of an equal sex ratio between num- 

 bers of juvenile and adult C. acronotus sampled 

 and among embryos in gravid females. To help 

 define time of parturition, analysis of variance 

 (ANOVA) was used to investigate the hypoth- 

 esis of an increase in TL of embryos from near- 

 term females captured in November and December. To 

 provide a balanced analysis, three gravid females captured 

 on 10 December 1998 were analyzed with three individu- 

 als captured almost one month earlier on 11 November 

 1998. Data were assessed for normality by using the Sha- 

 piro and Wilk procedure (Zar, 1996), tested for heterosce- 

 dasticity with Cochran's test, and then analyzed in the 

 appropriate one-nested factor ANOVA (Undei-wood, 1981). 

 Gravid females were considered a random-effects factor 

 nested in months and the four embryos for each gravid 

 female were the replicates. 



Results 



The Kolmogorov-Smirnov test detected significant differ- 

 ences in size-frequency compositions between males and 

 females; proportionally more larger-size females ( 121-131 

 cm TL) were captured. Both sexes showed two distinct 

 cohorts. The first consisted of juveniles (46-65 cm) cap- 

 tured by gill nets at depths between 5 and 10 m, and 



the second included larger specimens (i.e. 85-131 cm), 

 and mostly adults, captured on bottom longlines in deeper 

 water ( 10 to 60 m). The ratio of juvenile males to females 

 was not significantly different from 1:1 (X"=0.027,P>0.011, 

 whereas significantly fewer adult males than females were 

 sampled (ratio of 1:2.34) (x-=14.08.P<0. 01). 



Female maturation and reproduction 



Juveniles ranged in size from 46 to 101 cm TL and had 

 narrow oviducal glands, light ovaries with undeveloped 

 white follicles, and thin uteri (Fig. 1, A-C, Table 1). Pre- 

 ovulatory specimens ranged in size from 103 to 129.5 cm 

 TL, had well-developed oviducal glands (width of between 

 1.9 and 3 cm), and mature ovaries with thick orange fol- 

 hcles (1.5 to 2.8 cm in diameter) (Fig. 1, A and B, Table 1). 

 The shortest pre-ovulatoi\v specimens (e.g. 103 and 

 106 cm TL), had oviducal glands that were substantially 

 wider (i.e. >2x) than those in the longest juveniles ( 101 cm 

 TL) (Fig. lA, Table 1); therefore sexual maturity probably 

 was approached within this size range. This should be con- 



