86 



Fishery Bulletin 101(1) 



thus concluded that males mature in three years and fe- 

 males in four. In our study, males reached full maturity at 

 2.4 to 2.6 years of age, making them functionally mature 

 at the third breeding season following birth. Females were 

 found to mature at 2.2 to 2.5 years, which would also re- 

 sult in full maturity just prior to the third postnatal breed- 

 ing season. Although it was noted in both previously cited 

 studies that males matured six months to one year earlier 

 than females, no such discrepancy in age at maturity be- 

 tween the sexes was apparent in our study. 



Differences between studies 



The differences between this and previous studies on R. 

 terraenovae are likely a combination of many contributing 

 factors. These studies were conducted in different regions 

 at separate times and may reflect clinal or temporal differ- 

 ences (or both) between Gulf of Mexico and northwestern 

 Atlantic R. tej-raenovae populations. However, there are 

 other contributing factors that must be considered as well, 

 most notably differences in data collection and analysis 

 techniques. 



Parsons' ( 1985) growth curves were based on males and 

 were grouped into age classes (not assigned actual ages). 

 His von Bertalanffy parameters were then derived by us- 

 ing the Ford and Walford plot method (Parsons, 1985), re- 

 quiring the use of mean lengths of each age class. This age 

 class grouping does not take into account growth since the 

 deposition of the last increment, and may therefore bias 

 the Ford and Walford plot by pulling the data to a faster 

 asymptote (Branstetter and McEachran, 1986; Branstet- 

 ter, 1987a;). This bias produced a low L, (706 mm PCD 

 and /f,(-2.01 years) in Parsons' estimates (Branstetter and 

 McEachran, 1986; Branstetter, 1987a). This phenomenon 

 was not evident in VEGE estimates based on age classes 

 in our study, which were very similar to estimates based 

 on actual ages (Table 2), and was probably due to the fact 

 that iterative fitting of age data to the VBGE by computer 

 software (an option unavailable to Parsons at the time of 

 his study) is less sensitive to unaddressed growth than the 

 graphically based Ford and Walford plot method. 



Although the aging technique used by Branstetter 

 (1987a) was similar to that of our study (counts on lon- 

 gitudinal sections of cervical centra). Parsons' (1985) 

 aging technique took ring counts from the face of centra 

 that had been removed from a more posterior region of 

 the vertebral column than the region chosen in our study. 

 It has been stated by several authors (Branstetter and 

 McEachran, 1986; Martin and Cailliet, 1988; Kusher et al.. 

 1992) that increment counts made from sections of verte- 

 bral centra are generally preferable to those taken from 

 the face of unsectioned centra. Sectioned centra allow for 

 better documentation of the increment structure near the 

 edge because the increments become narrower and more 

 difficult to delineate with increasing age (Branstetter and 

 McEachran, 1986; Martin and Cailliet, 1988; Kusher et 

 al., 1992). This distinction is critical when the potential 

 consequences of age underestimation (including overesti- 

 mation of K', growth rate, and maximum sustainable yield) 

 are considered. 



Based on comparison of our work to that of previous 

 studies (Branstetter, 1981, 1987a; Parsons, 1983a, 1983b, 

 1985), there may be differences between the Gulf of 

 Mexico and southeastern U.S. Atlantic populations of 

 Atlantic sharpnose sharks. The question then becomes 

 whether these differences are clinal or temporal in nature. 

 Clinal variation, for instance, may explain the differences 

 noted in size and age at maturity in female/?, terrae/iovae. 

 Simpfendorfer ( 1993 ) noted differences in size at maturity 

 between populations of R. taylori in Australia, as did Par- 

 sons (1993) and Carlson et al. (1999) between populations 

 of Sphyrna tiburo and Carcharhinus acronotus. respec- 

 tively, off the Gulf coast of Florida. However, the extended 

 time frame between the current and previous studies ( 15 

 to 20 years), also opens the possibility that the differences 

 are related to a temporal change in population structure 

 of the species across the entire Gulf and Western Atlantic 

 region. In the earlier studies, data were collected during a 

 time when fishing pressure (both directed and indirected) 

 on R. terraenovae was lower than at present, and fisheries 

 were shown to have dramatic effects on shark populations 

 in less time (Anonymous'). The differences noted between 

 the studies may thus be a manifestation of temporal 

 changes in population structure of the species as a whole 

 over the last two decades. A more current study on Gulf of 

 Mexico R. terraenovae is needed to properly address these 

 potential population differences. 



Conclusion 



Small shark species such as R. terraenovae tend to show 

 rapid growth in the first few years of life and a dramati- 

 cally slower growth rate once maturity is reached. This 

 aspect of their growth complicates age estimation by ver- 

 tebral increments because the most recent marks in older 

 specimens are so closely spaced that accurate counting 

 and measurement become problematic. The overlapping 

 of increments in these older specimens or the lack of iden- 

 tifiable increment formation altogether due to asymptotic 

 growth may lead to an underestimation of ages in large 

 adults. Althhough the maximum age demonstrated in our 

 study was ll-i- years, the actual life span ofR. terraenovae 

 may be longer 



The life history parameter estimates that have been pre- 

 sented in our study are based on one of the largest short- 

 term samples collected for any study of elasmobranch life 

 history to date. The most significant aspect of this study is 

 the documentation of differences in size and age at matu- 

 rity between female R. terraenovae in the Gulf of Mexico 

 and females off the southeastern U.S. coast. A difference in 

 age of maturity of one year in an animal with a relatively 

 short life span, such as R. terraenovae, can have a dramatic 

 effect on the outcome of population models (see Cortes, 

 1995). Although the documentation of age at maturity dif- 

 ferences by different researchers may be highly susceptible 



Anonymous. 1993. Fishery management plan for sharks 

 of the Atlantic Ocean, 167 p. U.S. Dep. Commerce., NOAA, 

 NMFS, Silver Spring, MD 20910. 



