358 
Fishery Bulletin 96(2), 1 998 
labile 1 
Sample sizes and sex ratios of bottlenose dolphins, Thrsiops 
truncatus, stranded along the Texas coast from January 
1981 through December 1990 and of the subsample for 
which age was estimated. 
Sample 
Males 
Females 
Sex 
unknown 
Sex ratio 
(M:F) 
Total 
373 
292 
233 
1.3 : 1 
Aged 
78 
81 
36 
1 : 1.04 
<1.0 yr old 
22 
10 
7 
2.2 : 1 
1-20 yr old 
46 
39 
22 
1.2 : 1 
>20 yr old 
10 
32 
7 
1 : 3.2 
of the animals. The age structure of males and fe- 
males was compared by using the nonparametric 
Kolmogorov-Smirnov (K-S) test (Sokal and Rohlf, 
1981). 
Growth was determined by fitting the nonlinear, 
least-squares Gompertz model to length-at-age data 
(SAS Institute, 1985): 
Reproductive data were available for 57 females 
from the total sample of stranded females from which 
some data were collected (n= 292). Teeth were avail- 
able for 25 of these. Sexual maturity of females was 
determined by noting 1) the presence or absence of a 
corpus of ovulation from external examination of 
ovaries or 2) the presence of a fetus or an extended 
uterus, indicating a pregnancy. Sexual maturity of 
males could not be evaluated because few data or 
samples from testes were collected. 
Vertebrae from 24 males, 25 females, and 12 dol- 
phins of unknown sex were collected. All were exam- 
ined for state of physical maturity. From this sample, 
teeth were available for only 11 males, 12 females, 
and 8 specimens of unknown sex. Physical maturity 
was determined by noting the degree of fusion of the 
epiphysis to the centrum of thoracic and lumbar ver- 
tebrae. Three categories were used to classify the 
specimens: a) not fused (immature), b) fusing (ma- 
turing), and c) fused (mature). 
Results 
Sit) = A(exp(-6exp(-£))), 
( 1 ) 
Age and sex composition 
where S 
A 
b 
k 
t 
a measure of size (cm); 
asymptotic length; 
the constant of integration; 
the rate of growth constant; and 
age (yr) (Laird, 1966). 
Total body length was measured to the nearest cen- 
timeter in a straight line from the tip of the upper 
jaw to the notch of the fluke (Norris, 1961). Predicted 
asymptotic lengths for males and females were com- 
pared by using the approximate Gtest (Sokal and 
Rohlf, 1981). These lengths were then compared with 
the corresponding asymptotic lengths for male and 
female bottlenose dolphins from Sarasota Bay, 
Florida (Read et al., 1993). 
Length at birth was estimated in three ways: 1) as 
the mean length of 21 specimens from the entire data 
set (n= 898) determined to be neonates on the basis 
of the presence of remains of the umbilical cord or 
folded dorsal or caudal fins; 2) as the mean length of 
21 specimens (no overlap with the previous sample 
of 21) estimated to be <0.1 year old because the neo- 
natal line in tooth sections had not yet formed or 
had just started to form (Hohn and Hammond, 1985); 
3) as the predicted value from the Gompertz model. 
Calving season was also identified in three ways: as 
the date of stranding of 1) neonates, 2) specimens 
<0.1 year old, and 3) specimens measuring 90-120 cm 
in length. 
Estimates of age were obtained for 78 males, 81 fe- 
males, and 36 specimens of unknown sex (Table 1; 
Fig. 1). For an additional 10 dolphins (5 males, 3 fe- 
males, and 2 dolphins of unknown sex), only mini- 
mum ages, ranging between 12+ to 20+ years, were 
obtained because poor-quality tooth sections pre- 
cluded accurate age estimates. These 10 specimens 
were excluded from all analyses. Occlusion of the pulp 
cavity was not seen in any of the tooth sections. 
The oldest male and the oldest specimen of un- 
known sex were 33 years old and the oldest female 
was 41 years old (Fig. 2). Age distributions for males 
and females were significantly different (K-S test, 
P<0.01). Specimens less than 1 year old accounted 
for a relatively large part (20%) of the aged sample 
and were skewed towards males, but with increas- 
ing age, the proportion of females in the sample in- 
creased significantly (chi-square (% 2 ) P- 0.003, for 
data stratified by age group 0-0.9, 1.0-9. 9, 10.0-19.9, 
and >20 years; Table 1). Because of the relatively 
large number of males less than one year of age, the 
K-S test was rerun with that age class excluded for 
both males and females. The difference was still sig- 
nificant (K-S test, P<0.01). 
Growth 
The Gompertz model gave predicted asymptotic 
lengths of 263.5 cm for males and 244.7 cm for fe- 
