326 
Fishery Bulletin 118(4) 
The von Bertalanffy growth function was fitted to 
length-at-age data to estimate individual growth: 
L, = L(t = ee) (3) 
where L, = the length (TL, in centimeters) at age ¢; 
L,, = the asymptotic length; 
k = the growth coefficient; and 
ty = the theoretical age at length zero. 
Additionally, a likelihood ratio test (Kimura, 1980), also 
known as the likelihood ratio chi-square test, was applied 
to test for differences between males and females in 
parameters of individual growth. 
Results 
During the study period, 233 stone scorpionfish (117 males 
and 116 females) were sampled. No significant differ- 
ence was found in the ratio of males to females when 
applying a 1:1 exact binomial test (P>0.05). However, the 
number of males was lower than the number of females 
for the large fish (>30 cm TL). Overall, 70% of the 
sampled individuals ranged in size between 20 and 
30 cm TL. The lengths of the males ranged from 15.3 to 
35.5 cm TL (12.5—-28.9 cm SL), with a mean length of 
25.9 cm TL (SE 3.60). Comparatively, the lengths of the 
females ranged from 19.0 to 44.5 cm TL (15.5-36.5 cm 
SL), with a mean length of 27.8 cm TL (SE 5.17). Weights 
ranged from 64 to 1033 g for males, with a mean weight 
of 406 g (SE 181), and from 119 to 1870 g for females, 
with a mean weight of 545 g (SE 385). The analysis of 
variance revealed significant differences in mean length 
(Fy.931)=10.67, P<0.05) and mean weight (F(, 93;)=12.57, 
P<0.05) between males and females. Females were longer 
and heavier than males. 
The length—weight relationship (Table 1) was not signif- 
icantly different between males and females (F(; 999)=1.27, 
P>0.05). The b-value of the length—weight relationship of 
males and females was not significantly different from 3 
(P>0.05), indicating an isometric growth pattern. 
Otolith radius ranged between 2.5 and 5.6 mm. The lin- 
ear regression model developed to predict the relationship 
between OR and TL was as follows: 
TL = -3.53 + 3.910R. (4) 
The linear relationship between both variables was signif- 
icant (coefficient of determination [r7]=0.79, P<0.05). 
The highest percentage of otoliths with opaque edges 
was recorded in August (81%), followed by the percentage 
in December (65%), and the highest percentages of otoliths 
with translucent edges were observed in February (80%) 
and June (68%). These results indicate that the periodicity 
of annulus formation is annual and that the formation of 
a new annulus begins with an opaque band in the months 
of August-September and ends with a translucent band in 
the months of May—July. 
The detections of annuli were confirmed and annuli 
could be counted in 98% of the 233 otoliths analyzed. 
For the remaining 2% of otoliths, the annuli could not be 
counted because they were not visible. The ages of individ- 
ual specimens ranged between 2 and 10 years for males 
and between 3 and 15 years for females. The most fre- 
quent ages ranged from 5 to 7 years (63%) for males and 
from 4 to 6 years (74%) for females. The values of CV and 
APE were 6.50 and 4.59, respectively. 
The parameters of the von Bertalanffy growth function 
for males and females and the growth curves that were 
fitted to length-at-age data are shown in Figure 2. Results 
of the likelihood ratio test indicate significant differences 
in parameters of the von Bertalanffy growth function 
between males and females (y7=1.89, P<0.05); females 
grew slower (indicated by a lower k value) yet reached a 
greater maximum average length (indicated by a greater L,,.) 
than males. 
Discussion 
In this study, the maximum length recorded for stone scor- 
pionfish was 44.5 cm TL, which is comparable to the maxi- 
mum length of 51 cm TL previously reported for this species 
(Robertson and Allen, 2015). In addition, individuals larger 
Table 1 
Estimates of the parameters from regression analysis of the length—weight relationship for 
male and female stone scorpionfish (Scorpaena mystes) caught in the Gulf of California in 
Mexico during May 2015—April 2016. The parameters are the intercept (a) and slope (6) of the 
regression line. r’=coefficient of determination; ClI=confidence interval; SE=standard error of 
the mean. 
Sex Mean SE 
Males 0.0205 
Females 0.0159 
0.29136 
0.18687 
95% CI 
0.0115-0.0365 
0.0109-0.0229 
b 
95% CI r? 
Mean SE 
3.01 0.08975 
3.09 0.05696 
2.84-3.19 0.90 
2.98-3.21 0.93 
