Natanson et al : Growth of Galeocerdo cuvier in the western North Atlantic 



945 



sharks approach maximum sizes, 

 their vertebrae and band deposition 

 may not reflect age. Because at- 

 tempts at verification were unsuc- 

 cessful, and direct validation was 

 not possible, the determination of 

 the tiger shark growth rate was not 

 completely satisfied. 



To determine if vertebral growth 

 bands reflect age in older individu- 

 als and possibly to verify the esti- 

 mates of Branstetter et al. 1 1987), 

 we undertook a study with tag and 

 recapture data to estimate, inde- 

 pendently, von Bertalanffy param- 

 eters for the tiger shark. Verifica- 

 tion of neonatal and juvenile gi-owth 

 rates was accomplished by using 

 monthly length-frequency data ob- 

 tained over a period of seven years. 

 In addtion, data on growth of oxy tetracycline (OTO- 

 injected tagged and released tiger sharks from an 

 ongoing study were available to compare with length- 

 frequency and tagging growth data. 



Materials and methods 



Data from tiger sharks were obtained between 1963 

 and 1997 from research vessel cruises, sportfishing 

 tournaments, and the commercial shark fishery from 

 Cape Cod, MA, to the Florida east coast. Data for 

 monthly length-frequency analyses were obtained 

 from tiger sharks caught by longline in a delineated 

 area within the nursery grounds off Florida during 

 1988-94 (Fig. 1). 



Length measurements 



Measurements of total length (TL) and FL were taken 

 to the nearest centimeter (cm ) following the conven- 

 tions of Bigelow and Schroeder ( 1948). Fork lengths 

 are reported unless otherwise noted. TL to FL con- 

 versions can be calculated from the relationship 



Figure 1 



Map showing the portion of the nursery area (shaded box) from which monthly length- 

 frequency samples were obtained. 



cial fishermen who also reported shark size in TL, 

 FL, or weight. All measurements and estimates were 

 converted to FL. 



Gulland and Holt's ( 1959 ) and Fabens' ( 1965 ) meth- 

 ods were used to calculate von Bertalanffy (1938) 

 growth parameters from the tag-recapture data. 

 Techniques for calculating the parameters according 

 to Gulland and Holt ( 1959 ) came primarily from their 

 publication and additional clarification was obtained 

 from Cailliet et al. (1992). Only fish that were mea- 

 sured at both release and recapture and at liberty 

 for at least 0.9 years were included in the analysis. 

 Two of the three parameters for the von Bertalanffy 

 (1938) growth function (VBGF), k and L„, were esti- 

 mated directly with the methods of Fabens ( 1965 ) and 

 Gulland and Holt ( 1959). T^, cannot be estimated from 

 tagging data alone, rather it requires an estimate of 

 absolute size at age, such as size at birth, and was cal- 

 culated with the VEGF and solving for t,,, such that 



t.=t 



■(l//?)[ln{(L,, 



L,)IL_ 



% 



FL = (TL X 0.8761) 



13.3535 r- = 0.99n=44 



(Kohleretal., 1995) 



Li = known length at age (size at birth); 

 k = the von Bertalanffy growth constant; 



and 

 L . = the theoretical maximum attainable 



length from the VBGF 



Tag-recapture data 



During 1962-96, over 6000 tiger sharks were tagged 



-with NMFS tags (Casey, 1985) and released as part 



of the NMFS Cooperative Shark Tagging Program. 



Tags were returned primarily by sport and commer- 



The ?Q values were calculated based on an average 

 size at birth of 61 cm FL^ with t = 0. 



Longevity was estimated from the FL at which 

 >99'^'f of the L. was reached (i.e. 71n2/^) (Fabens, 

 1965; Cailliet et al., 1992). The von Bertalanffy pa- 

 rameters derived from these methods were compared 

 with growth information obtained from the length- 



