592 
Fishery Bulletin 96(3), 1 998 
tral); minimum straight-line carapace length (MSCL, 
nuchal notch to notch between postcentrals); mini- 
mum curved carapace length (MCCL, nuchal notch 
to notch between postcentrals); and straight-line 
carapace width (CW) at the widest point. Straight- 
line carapace lengths and width were measured to 
the nearest 0.1 inch with forester’s calipers. Curved 
carapace length was measured to the nearest 0.1 cm 
with a flexible fiberglass measuring tape. Weight 
(WT) was measured to the nearest 0.25 lb with a 
spring scale. All measurements were performed by 
the author to avoid individual differences in mea- 
suring technique (Bjorndal and Bolten, 1988) and 
were converted to metric units for analysis. Notes 
on the condition of the turtle were recorded when 
the animal was injured or deformed (e.g. tag scars, 
carapace wounds, etc.). 
Turtles were double tagged on the trailing edge of 
the fore flippers with no. 681 Inconel tags (June 1986 
to May 1988; May 1994 to October 1995), with Jumbo 
Roto plastic tags (June 1988 to October 1991), or with 
both (May 1992 to September 1993). Beginning in 
1988, two holes were drilled in specific marginal 
scutes of Kemp’s ridley turtles in order to identify 
the year of capture (1988 — left postcentral, 1989 — 
right postcentral, 1990 — left 12th marginal, 1991 — 
right 12th marginal, 1992 — left 11th marginal, 
1993 — right 11th marginal, 1994 — left 10th mar- 
ginal, and 1995 — right 10th marginal). Passive inte- 
grated transponder (PIT) tags were applied to the 
left front flipper of Kemp’s ridley turtles from June 
1992 to October 1995. Turtles were immediately re- 
leased after data collection approximately 100 m 
down-current from the netting site. 
Data analysis 
Kemp’s ridley turtles. Standard straight-line cara- 
pace length was used in the analyses of size distri- 
bution and growth. Means are followed by ± one stan- 
dard deviation unless noted otherwise. Turtle catch 
per unit of effort (CPUE) was standardized accord- 
ing to Shaver (1994) with the formula 
E J Nets * Length ^ 
where E = the netting effort in hours fished by a 1- 
km tangle net; 
Nets = the number of tangle nets fished; 
Length = the length (m) of a net; and 
Hrs = the number of hours fished. 
Kemp’s ridley turtle morphometric relationships 
were investigated by regressing carapace width on 
length and log-transformed weight on length. Con- 
version formulae for Kemp’s ridley turtle carapace 
lengths were calculated by regressing paired 
straight-line and curved carapace lengths. Turtles 
with carapace wounds or deformities were not in- 
cluded in regression equations. 
Yearly growth rates for Kemp’s ridley turtles were 
calculated with the formula 
G = 
' A Length ' 
v Days , 
365, 
where G = the growth rate in cm/yr; 
A Length = difference between the recapture length 
and the initial length; and 
Days = the number of days at large. 
Marine turtle life history stages were defined accord- 
ing developmental habitats and carapace lengths 
(Schmid, 1995). The term “juvenile” was reserved for 
immature turtles in the epipelagic stage of develop- 
ment. A turtle was considered “subadult” after re- 
cruiting to its respective coastal-benthic habitat and 
“adult” when sexually mature. Loggerhead turtles 
greater than 80 cm (Carr, 1986), green turtles greater 
than 83 cm (Witherington and Ehrhart, 1989), and 
Kemp’s ridley turtles greater than 60 cm (Pritchard 
and Marquez M., 1973) were considered adult based 
on sizes of nesting females. 
Capture records were analyzed to evaluate species 
composition within the Cedar Keys study area, 
length-frequency distribution of each species, and 
patterns of seasonal occurrence. Additional analyses 
of morphometries, growth rates, population esti- 
mates, and dietary composition were performed for 
Growth rates were grouped and analyzed in terms 
of the recapture interval duration, recaptures be- 
tween versus recaptures within netting seasons, and 
size classes of recaptured turtles. Growth rates were 
assigned to 10-cm size classes on the basis of mean 
of the initial and recapture carapace measurements 
(Bjorndal and Bolten, 1988). The von Bertalanffy 
growth interval equation was fitted to the recapture 
data with a nonlinear least-squares regression pro- 
cedure (SAS Institute Inc., 1989). The von Berta- 
lanffy growth interval equation (Fabens, 1965) for 
recapture data is: 
CL 2 = a - (a - CL x )e ~ kt , 
where CL 9 = the carapace length at recapture; 
a = the asymptotic length; 
CLj = the length at first capture; 
