Wintner and Cliff Age and growth determination of Carcharhinus limbatus 



141 



at birth, but during the first summer growth. 

 This provides a more plausible explanation 

 for the nature of initial band deposition found 

 in C. limbatus from southern Africa. The 

 narrow, translucent band found immediately 

 after the change in angle is deposited during 

 the first winter growth, and subsequent 

 opaque bands are the result of summer growth. 

 Clearly, the relationship between change in 

 angle and birth requires investigation. 



Centrum edge analysis (where the ratio of 

 translucent to opaque last bands differed sig- 

 nificantly from the expected ratio) and the 

 inability to perform marginal increment 

 analysis did not shed light on the periodicity 

 of band deposition. The inability to perform 

 the latter on sharks older than four years is 

 not unique to the present study. Killam and 

 Parsons ( 1989) used only sharks with two or 

 three growth rings for this analysis. In our 

 study, the shark nets tended to select for large 

 sharks; consequently few small C. limbatus 

 were sampled and only five animals younger 

 than four years were available. 



All age estimates from vertebral growth rings are 

 based on the assumption of an annual growth-ring 

 deposition. According to Cailliet (1990), the annual 

 periodicity of calcified growth zones has been proven 

 for only five species and partly confirmed for 25 spe- 

 cies, including C. limbatus (Branstetter, 1987, a and 

 c; Killam and Parsons, 1989). A project was initiated 

 recently in which all shark species, including 

 C. limbatus, found alive in NSB nets were injected 

 with oxytetracycline. In this study, growth ring peri- 

 odicity for C. limbatus from southern Africa could 

 not be confirmed, owing to a lack of holding facilities 

 for large sharks and to the absence of any recaptured 

 sharks injected with oxytetracycline. 



A linear relationship between vertebral diameter 

 and animal length was also found in C. limbatus by 

 Branstetter ( 1987a) and Killam and Parsons ( 1989) 

 and has been found in several other shark species 

 (Cailliet et al., 1983b; Schwartz, 1984; Branstetter, 

 1987b). The existence of this relationship in the 

 present study justified the use of the Dahl-Lea 

 method of back calculations. These back calculations 

 were not used to confirm growth-ring periodicity, 

 because this method is based on several assumptions 

 and has inherent problems (Smith, 1983) but was 

 used to test for Lee's phenomenon and to derive a 

 comparison with observed values. Unfortunately, the 

 use of mean age estimates in this study created dis- 

 crepancies in back calculations. The number of ob- 

 served lengths per age class, plus the number of back- 

 calculated lengths in the next age class, does not al- 



ways add up to the total number of observed lengths. 

 This phenomenon is evident in Table 2. 



Mean back-calculated lengths were lower than 

 observed values, as was found by Killam and Par- 

 sons (1989). Branstetter ( 1987a) found that observed 

 lengths were slightly greater than back-calculated 

 values. Lee's phenomenon was encountered in the 

 present study. Even the exclusion of slow-growing 

 individuals in age classes with few individuals did 

 not affect the phenomenon. Killam and Parsons 

 ( 1989) noted that it appeared to occur at some ages, 

 but no consistent trend was identified. Branstetter 

 (1987a) found no evidence of Lee's phenomenon in 

 C. limbatus. 



Age and growth estimates 



The mean back-calculated birth size was slightly 

 lower than the size of the neonates, as found by 

 Branstetter (1987a) and Killam and Parsons (1989) 

 (Table 3). On the basis of neonate sizes, C. limbatus 

 from KwaZulu-Natal is born at a slightly larger size 

 than are those from the Gulf of Mexico. Branstetter 

 (1987a) remarked that in general C. limbatus is 

 smallest in the northwestern Atlantic and largest in 

 the Indian Ocean. This is not only evident in birth 

 sizes but also in sizes at maturity and maximum sizes 

 (Table 3). 



On the KwaZulu-Natal coast, C. limbatus of both 

 sexes mature at similar ages, males at six years and 

 females at seven years. They mature at similar 



