596 
Fishery Bulletin 96(3), 1998 
Table 2 
Seasonal water temperatures and Kemp’s ridley turtle, Lepi- 
dochelys kempii, carapace lengths at Corrigan Reef from 1986 
to 1995 (standard deviations given in parentheses). 
Season 
Mean 
water 
temperature 
Mean 
carapace 
length 
n 
Size range 
Spring 
22.5°C 
(3.0) 
44.1 cm 
(7.4) 
24 
26.8-58.6 cm 
Summer 
30.4°C 
(0.6) 
45.5 cm 
(5.9) 
142 
28.2-54.4 cm 
Fall 
22.9°C 
(4.2) 
43.1 cm 
(6.6) 
85 
27.3—56.6 cm 
Winter 
15.7°C 
(2.0) 
0 
was not significantly different from the distributions 
of 1986 through 1990. The observed shift in size-class 
distribution for 1991 may have been caused by 
changes in fishing conditions that year. Beginning 
in July 1991, a smaller mesh (25.4-cm bar) net was 
deployed either singly or in combination with the 
larger mesh (30.5-cm bar) net that was used the pre- 
vious years. The smaller mesh net may have resulted 
in the increased capture of 30-40 cm turtles in 1991, 
although the frequency of 40-50 cm turtles increased 
in the following years. Also, during August 1991, a 
massive influx of pelagic Sargassum occurred along the 
west coast of Florida and the majority of netting effort 
was conducted in the months following this unusual 
event. It is not known how this latter condition may 
have affected the relative frequency of carapace lengths, 
either by increasing the frequency of 30—40 cm turtles 
or decreasing the frequency of 40-50 cm turtles. 
Carapace regression equations 
There was a strong correlation between carapace 
width and carapace length (r=0.9883, n= 227) for 
Kemp’s ridley turtles. Regression of width on length 
resulted in the equation 
CW = -3.7415 + 1.0530 ( SSCL ). 
A strong correlation (r=0.9886, n=225 ) was calculated 
for the weight-to-length data transformed with the 
natural logarithm. Regression of these variables re- 
sulted in the equation 
In WT = -8.1570 + 2.8128 (In SSCL). 
Conversion equations were computed between the 
straight-line and curved carapace length measure- 
Table 3 
Formulae for converting between straight-line and curved 
carapace measurements of Kemp’s ridley turtles, Lepido- 
chelys kempii. TSCL = total straight-line carapace length, 
SSCL = standard straight-line carapace length, MSCL = 
minimum straight-line carapace length, and MCCL = mini- 
mum curved carapace length. 
Converted 
length 
Conversion formula 
n 
r 2 
TSCL 
1.0118 SSCL + 0.0650 
227 
0.9990 
TSCL 
1.0214 MSCL + 0.3800 
227 
0.9982 
TSCL 
0.9720 MCCL + 0.2203 
107 
0.9920 
SSCL 
0.9874 TSCL -0.0199 
227 
0.9990 
SSCL 
1.0094 MSCL + 0.3155 
227 
0.9990 
SSCL 
0.9598 MCCL + 0.1941 
107 
0.9929 
MSCL 
0.9773 TSCL - 0.2896 
227 
0.9982 
MSCL 
0.9897 SSCL -0.2658 
227 
0.9990 
MSCL 
0.9495 MCCL - 0.0864 
107 
0.9933 
MCCL 
1.0205 TSCL + 0.1368 
107 
0.9920 
MCCL 
1.0345 SSCL + 0.1162 
107 
0.9929 
MCCL 
1.0462 MSCL + 0.3913 
107 
0.9933 
ments (Table 3). These equations will allow for com- 
parisons between studies with different measuring 
techniques. 
Growth analyses 
Twenty-one Kemp’s ridley turtles were recaptured a 
total of 24 times, yielding 24 annual growth rates. 
However, 83% of the recaptures occurred within a 
year of initial tagging and extrapolating annual 
growth rates from short-term recapture intervals will 
amplify any error associated with the measurements. 
The removal of short-term recaptures decreased the 
range of annual growth rates and increased the pre- 
cision of the mean growth rate estimate (Table 4A). 
Subsequent analyses of growth by netting seasons 
and size classes were confounded by short-term re- 
captures. The mean growth rate of Kemp’s ridley 
turtles recaptured within netting seasons (see Table 1 
for months fished each year) was significantly larger 
(X 2 =7.93, df=l, P=0.005) than that of turtles recap- 
tured between netting seasons (Table 4B). However, 
the duration of all recaptures within netting seasons 
was less than 180 days (mean=49.6 ± 44.5 days) and 
the annual growth rates may have been overesti- 
mated owing to extrapolation error. Although mean 
growth rates did not vary significantly when com- 
pared by size class (F=0.753, P=0.484), Kemp’s rid- 
ley turtles in the 40-50 cm size class appeared to 
have a higher mean growth rate than those in the 
30-40 cm and the 50-60 cm size classes (Table 4C). 
