811 



Abstract— Tagged neonate and small 

 juvenile Carcharhinus obscurus were 

 released between March 1994 and 

 June 1996 in southwestern Australia. 

 Length and time-at-liberty data from 

 304 usable recapture data were used to 

 examine growth rates of male, female, 

 oxytetracycline injected, noninjected, 

 and all individuals combined. Recap- 

 tures were made up to 4.7 years after 

 release. Four methods of analysis were 

 employed: Gulland and Holt, Fabens, 

 Francis, and length-at-age. Length-at- 

 age analysis was possible because the 

 majority of the animals released had 

 open umbilical scars and so were of 

 known age. The four methods pro- 

 duced varying results. One method was 

 not able to estimate growth param- 

 eters, another produced inaccurate 

 estimates of von Bertalanffy growth 

 parameters, and two methods indicated 

 that a linear growth model described 

 growth better than the von Bertalanffy 

 model. Although each produced differ- 

 ent results, the three successful meth- 

 ods estimated that the growth rate 

 up to age 5 ranged from 8 cm/year 

 to 11 cm/year. These growth rates 

 agreed closely with those reported for 

 young C. obscurus. Length-at-age anal- 

 ysis indicated significant differences in 

 the growth rates between males and 

 females, and between oxytetracycline- 

 injected and noninjected males. Results 

 from the Francis method did not show 

 significant differences between males 

 and females, or injected and nonin- 

 jected animals. The coefficient of varia- 

 tion of growth variability ranged from 

 0.24 to 0.40, mean measurement error 

 ranged from 0.0 cm to 0.94 cm, and 

 the standard deviation of measurement 

 error ranged form 2.1 cm to 2.4 cm. 

 The usefulness of each of the methods 

 is discussed — the more detailed meth- 

 ods providing a better understanding 

 of tag-recapture data. 



Growth rates of juvenile dusky sharks, 

 Carcharhinus obscurus (Lesueur, 1818), 

 from southwestern Australia estimated 

 from tag-recapture data 



Colin A. Simpfendorfer 



Western Australian Marine Research Laboratories 



PO Box 20 



North Beach 



Western Australia 6020, Australia 



Present address: Center lor Shark Research, Mote Marine Laboratory 



1600 Ken Thompson Parkway 



Sarasota, Flonda, 34236 

 Email address: colins(a)mote,org 



Manuscript accepted 26 June 2000. 

 Fish. Bull. 98:811-822 (20001. 



Tag-recapture data are a useful source 

 of information on the growth of animals. 

 The simplest form of analysis is to esti- 

 mate an average growth rate, normally 

 for a specific size range, from the period 

 at liberty and the growth increment. 

 This approach has been used in studies 

 examining elasmobranch growth. For 

 example, Thorson and Lacy (1982) esti- 

 mated the average growth rate of adult 

 sawfish (Pristis perotteti). Pratt and 

 Casey (1983) provided average growth 

 rates for shortfin mako sharks (Isurus 

 oxyrinchus) in 20-cm size groups, and 

 Casey et al. (1985) provided estimates 

 of giowth rates for sandbar sharks 

 [Carcharhinus plumbeus) in 20-cm size 

 groups. Although this approach provides 

 information on growth rates that can be 

 compared with those predicted by other 

 techniques, it does not provide estimates 

 of parameters for growth functions (e.g. 

 the von Bertalanffy growth function). 



A more useful approach is to fit 

 growth functions to tag-recapture data. 

 A range of techniques have been devel- 

 oped to undertake this type of anal- 

 ysis. Early techniques used relatively 

 simple approaches (e.g. Walford. 1946; 

 Gulland and Holt, 1959; Fabens, 1965) 

 to estimate parameters of the von Ber- 

 talanffy growth function. Species of 

 elasmobranchs for which these types 

 of techniques have been used include 

 three species of Raja (Holden, 1972), 

 C. plumbeus (Wass, 1973; Casey and 

 Natanson, 1992), Galeorhinus galeus 

 (Grant et al, 1979). Squatina califor- 

 nica (Cailliet et al., 1992) and Galeo- 

 cerdo cuvier (Natanson et al., 1999). 

 With the advent and proliferation of 



powerful computers, more sophisticated 

 techniques have been developed to deal 

 with tag-recapture data and to allow 

 the fitting of a range of growth func- 

 tions and the estimation of confidence 

 intervals (Francis, 1988). These growth 

 functions have included both the widely 

 used von Bertalanffy curve, as well as 

 other forms (e.g. Schnute, 1981; Fran- 

 cis, 1995). In addition, functions that 

 allow for the estimation of seasonal 

 growth patterns (e.g. Francis, 1988), 

 growth variability (e.g. Francis, 1988, 

 Wang et al., 1995), and measurement 

 error (e.g. Francis, 1988) have been 

 developed. These more complex func- 

 tions allow researchers to extract more 

 information from tag-recapture data, 

 as well as identify factors that may be 

 important in determining the growth 

 of individual animals. These types of 

 analysis have not been widely used in 

 elasmobranch studies to date. However, 

 Francis and Francis ( 1992 ) and Francis 

 and Mulligan (1998) used the Francis 

 (1988) method for Mustelus lenticula- 

 tus and G. galeus, respectively. 



Tagging studies of elasmobranch pop- 

 ulations have also been used to vali- 

 date age data based on vertebrae. In- 

 dividuals are injected with a marker 

 that is incorporated into calcifying struc- 

 tures so that the number of bands laid 

 down between tagging and recapture 

 can be determined, and the periodicity 

 of band formation validated (Cailliet, 

 1990). The most commonly used marker 

 in these studies is oxytetracycline (OTC ) 

 at dose rates of approximately 25 mg/kg 

 (Gelsleichter et al., 1998). OTC is also a 

 powerful antibiotic with recommended 



