102 
Fishery Bulletin 109(1 ) 
validation (Campana, 2001; Cailliet and Goldman, 
2004). Campana (2001) defines validation as either the 
validation of the periodicity of growth increments, or as 
the validation of the age estimate made by reader(s). 
In a recent review, Campana (2001) noted that of the 
372 papers in which age validation was reported, only 
15% of the papers actually incorporated validation of 
the absolute age of wild fish. Therefore, when conduct- 
ing an age and growth study it is necessary to specify 
whether one or both validation goals are met and which 
validation method is used. 
Three methods of validation are typically used. These 
are edge analysis (also known as marginal increment 
or zone analysis), carbon dating, and mark-recapture of 
tagged fish injected with a calcium-chelating fluoresc- 
ing chemical such as the antibiotic, oxytetracycline 
hydrochloride (OTC). All of these methods have been 
used to validate elasmobranch ages (Campana et al., 
2002; Smith et al., 2003; Cailliet and Goldman, 2004; 
Ardizzone et al., 2006; Chen et al., 2007). The most 
commonly used and most reliable validation method is 
tagging with OTC (Campana, 2001). 
In studies of elasmobranchs by OTC -injection, sharks 
are measured, injected with OTC, tagged, and released. 
The date of release is then noted. Once the shark is 
recaptured, the number of vertebral band pairs distal 
to the fluorescent mark is compared with the time be- 
tween release and recapture of the shark. This experi- 
mental approach can be problematic if recapture rates 
are low. Validating elasmobranch age estimates can be 
time consuming because recapture rates are generally 
low and there is a long period of time between collect- 
ing a sufficient number of samples and analyzing the 
vertebrae. 
We developed a Leslie matrix-based demographic 
model to evaluate the ability of T. megalopterus to sus- 
tain increased levels and patterns of fishing pressure. 
We also validated the age of T. megalopterus using 
oxytetracycline to estimate growth and mortality rates, 
important demographic model parameters. 
Materials and methods 
General overview 
In 1994, a tagging program incorporating researchers 
and trained volunteer fishermen was initiated at the 
Port Elizabeth Museum in order to obtain age valida- 
tion, movement studies, and population dynamics of the 
spotted gully shark. A total of 402 wild sharks (113 male, 
230 female, and 59 of undetermined sex) were tagged, 
injected with OTC at a dosage of 50 mg/kg (Tanaka, 
1990), and released. A total of 53 sharks were recap- 
tured once, and one shark twice. The date for the first 
recaptured fish was unfortunately unrecorded, but it was 
evident that it had been re-injected with OTC from the 
presence of an additional fluorescing mark in its verte- 
brae. An additional 12 display sharks were held in the 
Bayworld Aquarium in Port Elizabeth and injected with 
OTC. All 12 display sharks and 11 tagged wild sharks 
were sacrificed for vertebral analysis. 
In addition, a total of 129 spotted gully sharks were 
collected opportunistically from ski-boat fishermen, fish- 
ing competitions, and research cruises over a 21-year 
period (1984-2009) between Cape St Francis and Cof- 
fee Bay, South Africa. Vertebrae were collected from a 
subsample of 96 sharks. Total length (TL) and sex were 
recorded for all these sharks. 
Age determination 
Between five and eight vertebrae were removed from 
the trunk region in the vicinity of the first dorsal fin, 
soaked in 4.5% sodium hypochlorite for 15-45 minutes 
to remove excess connective tissue, and were either 
stored in 70-80% ethyl alcohol or frozen (Yudin and 
Cailliet, 1990). Cleaned vertebrae were embedded in 
polyester casting resin and sectioned with a diamond- 
bladecl saw along the sagittal plane to a thickness 
of 0.6 mm (Natanson et al., 2006; Rizzo et al. 1 ), and 
mounted on glass slides with DPX mountant (Lasec, 
South Africa). 
Band pairs, defined as one optically opaque and 
one optically translucent band were counted by us- 
ing a dissecting microscope with transmitted white 
light (460-490 nm). OTC-injected specimens were also 
viewed with an Olympus BX60 microscope (Olympus, 
Johannesburg, South Africa) under ultraviolet trans- 
mitted light (510-550 nm). Each specimen was aged 
twice, three weeks apart by a single reader, without 
prior knowledge of the length or sex of the specimen. 
Counts were accepted only if both counts were in agree- 
ment. If the estimated number of bands differed by 
two or less, the specimen was recounted and the final 
count was accepted as the agreed upon number; if not, 
the specimen was discarded. If the third count did not 
concur with one of the previous two counts, the sample 
was rejected. 
An age-bias plot was used to graphically assess the 
readings and their associated agreement (Campana, 
2001; Natanson et al., 2006). A t-test was conducted on 
the slope of the age-bias plot (the linear regression of 
the second against the first age readings) to test the 
null hypothesis that the slope was equal to one. Com- 
parisons of reader accuracy for each age were made by 
using a paired t-test, and a j 2 -test of symmetry was 
used to test for systematic bias in the determination of 
age (Hoenig et al., 1995). 
The variability of the within-reader age estimates 
was estimated with an index of average percent error 
1 Rizzo, P., S. Gancitano, C. Badalucco, S. Enajjar, C. Mancusi, 
A. Mosteiro Cabanelas, B. Saidi, and L. Sion. 2004. Contri- 
bution to guidelines for age determination of chondrichthyes 
fish from the Mediterranean Sea (application to selected 
species). Report of the MedSudMed training course on age 
determination of selacean fish; 22 November-01 December 
2004, Mazara del Vallo, Italy, 22 p. [Available from FAO- 
MedSudMed Project , Mazara del Vallo, Italy 
