58 
Fishery Bulletin 1 13(1) 
stage 3, the ovary is developed, has a small volume and 
is orange, and has visible oocytes at the ovary wall. At 
stage 4, the ovary is largely inflated, fills the whole 
carapace cavity, and is bright orange, with oocytes 
clearly visible along the ovary wall. Finally, at stage 
5, the ovary is developed but has a flaccid appearance 
and is cream to yellow; there may still be some oocytes 
on the ovary wall — a specific peculiarity of this stage. 
Individual fecundity was quantified volumetrically 
(Vazzoler, 1996) by using females with eggs in the early 
stages of embryonic development (orange in color). This 
method avoided underestimation caused with females 
at later stages, when there was the possibility of loss of 
eggs with catching and handling, or early larval hatch- 
ing (DeMartini and Williams, 2001). The laboratory 
protocols were the same as those of Vazzoler (1996); 
eggs were removed from pleopods by dissociation in 
10% NaCIO, and were then vigorous stirred. The oo- 
cytes from each egg mass were transferred to a 1-L 
beaker and homogenized with a glass rod, and seven 
5-mL subsamples were removed with a pipette. The oo- 
cytes were placed in gridded petri dishes for counting 
under a stereomicroscope ( 20 x). We estimated individ- 
ual fecundity (F) according to the method proposed by 
DeMartini and Williams (2001), counting the number 
of eggs in subsamples, discarding the extreme values 
(minimum and maximum), and calculating the mean 
individual fecundity of the 5 remaining subsamples (fs) 
with the following equation: 
F = (tv x fs) / dv, (1) 
where tv = total volume of the egg mass with a dilution 
of 1 L; and 
dv = the dilution volume of the egg subsample or 
5 mL). 
The relationship of TL to the number of eggs (NE) was 
calculated with the following equation: 
NE = (b x TL) — a, (2) 
where a = the intercept; and 
b = the slope of linear regression. 
Data analysis 
The statistical analyses for this study were performed 
with R software, vers. 2.13.0 (R Development Core 
Team, 2011) (Ihaka and Gentleman, 1996), and the 
fisheries were mapped with the Surfer contouring 
and 3-D mapping package, vers. 8 (Golden Software, 
Golden, CO). We adopted the TL as the size indicator 
for the hooded slipper lobster, as was done in previ- 
ous studies of lobsters of the genus Scyllarides (Hearn, 
2006; Pessani and Mura, 2007). 
The relative growth of each sex was assessed by a 
regression analysis of the empirical points for mor- 
phometric relationships given by the power function 
y=ax b , and then a i-test was used to check for possible 
differences in allometry (Huxley, 1950). The following 
morphometric relationships were evaluated: ALxTL, 
CLxTL, LAxTL, AWxTL, CWxTL and WxTL (Abele, 
1982). These relationships indicated the possible differ- 
ences between the linear regressions obtained for each 
sex by comparisons with an analysis of covariance of 
constants a and b in the linear regressions of the bio- 
metric relationships (Zar, 1999; Faraway, 2002). 
To represent the allometric relationships from the 
power function in a linear form, the dependent ty-axis) 
and independent (x-axis) variables were transformed 
by natural logarithms (In). This transformation, 
lny = lna + (bx lruc), (3) 
facilitates the use of a least squares fitting technique 
with linear regression: 
y = a + (b x x). (4) 
This mathematical procedure transforms the curvilin- 
ear relationship into a linear equation or equations and 
enables graphic comparisons of linear growth phases 
for the evaluation of biometric distinctions between 
developmental stages (juvenile and adult). Therefore, 
the linearized data (lny=lna + [ 6 xlrLr]) for each relation- 
ship were assessed with the segmented library (Mug- 
geo, 2008) in R (vers. 2.13.0) to identify significantly 
different growth rates during ontogeny, according to 
procedures used by Pardal-Souza and Pinheiro (2012). 
This method of regression analysis seeks to partition 
the independent variable into intervals at break points, 
separating line segments that are fitted to each inter- 
val. The mean length at which 50% of females reached 
maturity (L 50 ) was estimated, under sigmoid adjust- 
ment, from the proportion of individuals with mature 
gonads (females in stages 3-5) in size classes of 1 cm 
TL (Hovgard and Lassen, 2000): 
P = 1 / (1 + e- r(i - L 5 o>), (5) 
where P = the probability that a female is mature; 
r = the slope; 
L 50 = the mean TL where the probability of mature 
females is 50%; and 
L = the mean TL. 
Data on the carapace color were analyzed in relation 
to size class, sex, the area of capture, and fishing gear. 
Five major fishing areas were established by mapping 
the fishing grounds of 2 types of vessels with different 
storage types (double trawlers that store their catch 
on ice and vessels that use pots and traps and store 
catch in refrigerators) on the basis of their similar lati- 
tudes and depths of fishing grounds for lobster species. 
Size composition and sex also were evaluated by fish- 
ing area, season, and month of capture. Size distribu- 
tion was compared between sexes by using a t-test (Zar, 
1999). An analysis of variance was used, separately for 
each variable, to evaluate the occurrence of different 
color patterns by sex (exclusivity or predominant pat- 
terns), fleet (differences in storage type), capture pe- 
riod (month and season), and fishing area, as well as to 
