78 



Fishery Bulletin 101(1) 



Lq and Cq are biologically derived intercepts that repre- 

 sent the fish length and centrum radius, respectively, at 

 which the proportionality between fish length and centrum 

 growth are initiated. For the purposes of this study, mean 

 body length and centrum radius at birth were used as the 

 biologically derived constants (Sminkey and Musick, 1995). 

 The observed age-class data were used to estimate 

 "actual ages" to 0.1 year These were calculated by the 

 number of circuli present plus growth since the formation 

 of the last circulus. All specimens were given a 1 June 

 birth date, which approximates the middle of the pupping 

 season. This process corrected for growth since the last in- 

 crement, preventing the potential overestimation of size- 

 at-age that might result from analyzing the data by year 

 class alone. All three types of length-at-age data (observed 

 age class, observed actual age, and back-calculated age) 

 were fitted to the von Bertalanffy growth equation (VBGE; 

 von Bertalanffy, 1938): 



L, = LJ1 



-Kit- 



'o'), 



where L, = length at age t; 



L^ = asymptotic length; 



K = growth coefficient; and 



Tq = theoretical age at zero length. 



Each of the three types were analyzed for sexes combined, 

 as well as for each sex separately. The parameters for the 

 VBGE were estimated through a stepwise Gauss-Newton 

 iterative fitting process computed by JMP statistical 

 analysis software (Anonymous, 1998). 



Results 



The sharpnose shark was abundant throughout the year 

 in coastal waters within the sampling area. The ratio of 

 males to females in the overall sample was not signifi- 

 cantly different from a 1:1 ratio (chi-square test, « = 1091. 

 a=0.65, v=l,;^2=1.39,P=0.24). 



Linear regression of TL, NTL, and FL on PCL resulted 

 in the following equations: 



TL = 29.804 + 1.279PCL 

 NTL = 31.678 + 1.254PCL 

 FL= 11.249 -H 1.075PCL 



(;i = 1009, ;-=0.99,P<0.0001); 



(n=493,r2=0.99,P<0.0001); 



(n = 1083, r2=0.99, P<0.0001). 



Reproduction and maturity 



Size-at-maturity estimates were based on observations 

 of 526 males and 564 females. The smallest fully mature 

 male was 600 mm PCL, and the largest immature male 

 was 615 mm PCL. All males greater than 615 mm PCL 

 and 36'" '" -ales from 600 to 615 mm PCL were fully 

 mature. The onset and completion of maturity in male R. 

 tcrraenorae were demonstrated by the onset of develop- 

 ment in the claspers and siphon sac (Fig. 3). Males began 

 to mature at 500 mm PCL. The maturation of claspers and 

 siphon sac reached completion approximately one year 

 later, at 600 to 615 mm PCL. 



The smallest maturing female was 509 mm PCL and con- 

 tained one maturing oocyte five mm in diameter The second 

 smallest maturing female was 529 mm PCL. The smallest 

 gravid female was 591 mm PCL. The largest immature fe- 

 male, based on lack of embryos or uterine scarring, was 611 

 mm PCL. Females from 591 to 611 mm PCL were either 

 gravid (63%) or contained large (>10 mm diameter) matur- 

 ing oocytes and were close to their first ovulation (377^ ). All 

 females greater than 611 mm PCL were mature. 



Mean GSI and mean ovarian egg diameter (MOD) both 

 demonstrated prominent peaks during the calendar year 

 Male GSI values were highest in April and high values 

 were also present in March and May (Fig. 4). However, 

 the seminal vesicles remained turgid and full of semen for 

 some time following the seasonal testicular degeneration 

 which began in May. Female MOD values were highest in 

 May and June. An increase in standard error along with a 

 drop in mean value for the month of June (Fig. 4) demon- 

 strated that ovulation began at that time. The extremely 

 low MOD in July indicated the completion of ovulation. 



Litter sizes ranged from one to eight, and generally in- 

 creased with female PCL (Fig. 5). Mean litter size was 3.85 

 embryos, and significantly more embryos were found in 

 the left uterus (mean=2.19) than in the right (mean=1.65; 

 chi-square test, «=558, a=0.05, v=4. ^2=62.62, P<0.0001). 

 Nonlinear regression of litter size on female PCL resulted 

 in the following equation («=278, /■'-=0.51, P<0.0001): 



Litter size = -11.07 -i- 0.021 PCL + 1.37 

 X lO-^lPCL- 710.9)2 



Rhizoprionodon terraenovae were born at approximate- 

 ly 212 mm PCL. The smallest free-swimming neonate was 

 190 mm PCL, and the largest full-term embryo was 242 

 mm PCL. Most pupping occurred from mid-May to early 

 June. However, a small number of neonates appeared 

 as early as mid-April. Consequently, mean embryo total 

 length was at a minimum in July and at a maximum in 

 June (Fig. 6). The sexes of uterine embryos were not sig- 

 nificantly different from the expected 1:1 ratio (chi-square 

 test, ?!=844, a=0.05, v=l, ;f-=0.076, P=0.78). 



Age and growth 



Separate linear regressions of PCL on centrum radius 

 (CR) for males and females were not significantly different 

 (ANCOVA, P=0.065l and were therefore combined (Fig. 7) 

 to yield the following formula: 



PCL = 61.80 -I- 124.48Ci? (r2=0.963, n=812, P<0.0001). 



The regression line slightly overestimated centrum radius 

 for the largest individuals (>700 mm PCL) of both sexes. 

 Data transformation, as well as nonlinear regression, 

 failed to increase the r^ value, and only the largest speci- 

 mens were affected. 



Nonlinear regression of total body weight on length was 

 significantly different between males and females (AN- 

 COVA after log-transformation, P<0.001), and resulted in 

 the following equations: 



