Zhao and McGovern: Variation in sexual maturity and sex ratio of Rhomboplites aurorubens 
845 
Table 6 
Comparison of percentages of females among periods with 
the same latitude (32°00'-32°59' N) and gear type. All data 
were pooled from various depth classes. % = female percent. 
n = the total number of male and female fish. The null 
hypothesis (H 0 ) = sex ratio is independent of period. Blanks 
indicate few samples available for comparison or the null 
hypothesis of independence between sex ratio and length 
was rejected 
in Table 5. df = 
degrees of freedom. 
Traps 
Hook-and-line 
Trawl 
(TL=200-349) 
(TL=200-299) 
(TL=150- 
249) 
Period 
% n 
% 
n 
% 
n 
1979-81 
70.6 
211 
58.8 
204 
1982-84 
72.6 453 
68.4 
244 
1985-87 
66.5 
227 
61.1 
190 
1988-90 
72.5 727 
67.6 
321 
1991-93 
71.1 554 
64.4 
45 
chi-square 
0.382 
1.199 
0.204 
P 
0.826 
0.878 
0.652 
df 
2 
4 
1 
Reject H 0 
No 
No 
No 
Table 7 
Comparison of percentages of females among gear types 
with the same latitude (32°00'-32°59' N). All data were 
pooled from various depth classes. % = female percent, n = 
the total number of male and female fish. The null hypoth- 
esis (H 0 ) = sex ratio is independent of gear type. Data were 
pooled from periods that were used in Table 6. Data with 
different length ranges were used in A, B, and C. (A) TL 
ranges were the same as used in Table 6 for each gear; (B) 
TL = 200-249 mm for all gear types; (C) TL = 186-540 
mm, mean TL = 254 mm for traps, TL = 142-560 mm, mean 
TL = 261 mm for hook-and-line, and TL = 112-256 mm, 
mean TL = 
203 mm for trawl. 
Gear 
A 
B 
C 
% 
n 
% 
n 
% 
n 
Traps 
72.1 
1734 
72.4 
919 
72.1 
1786 
Hook-and-line 68.0 
1048 
68.6 
544 
66.2 
1395 
Trawl 
59.9 
394 
60.0 
255 
61.0 
415 
chi-square 
23.460 
14.558 
24.938 
P 
<0.001 
0.001 
<0.001 
df 
2 
2 
2 
Reject H 0 
Yes 
Yes 
Yes 
ranges (Table 7B: a single TL range for all gear types; 
Table 7C: all data available were used with full TL 
ranges). Vermilion snapper caught by all gear types 
had an unequal sex ratio that was female-biased 
(traps and hook-and-line: P« 0.001, trawl: P<0.005). 
Discussion 
Maturity schedules 
Although data used for maturity analysis were lim- 
ited to those collected in the same season (May and 
June) of each period, growth may occur, and stages 
of maturity may change, within two months. An im- 
mature fish in May may become mature in June. If 
more fish were collected in June of recent years than 
in 1979-81, the percent mature at a certain age would 
be overestimated for recent years. The monthly dis- 
tribution of observations, however, was similar 
among periods with more than 80% of the fish smaller 
than 170 mm being collected in May. Relative differ- 
ences in maturity schedules among periods remained 
valid throughout the study. 
The essential underlying assumption of the matu- 
rity analysis is that length, age, and reproduction 
conditions are correctly measured or determined. In 
1985, 1986, 1987, and 1988, only fork lengths (FL) 
were measured. These FL were converted to TL with 
TL (mm) = 1.115PL - 0.254 (Zhao et al., 1997). For 
other years, observed TL was available. Because the 
method for ageing vermilion snapper by means of 
otolith sections has been validated by Zhao et al. 
( 1997) and because the persons who read otoliths in 
Zhao et al.’s study also read otoliths in our study, 
incorrect ageing of vermilion snapper is not consid- 
ered a source of bias. During 1979-86, gonads were 
examined macroscopically by experienced biologists 
by means of clearly defined gross morphological stag- 
ing criteria. These criteria were confirmed to be ac- 
curate by histological examination during 1978-80. 
All gonads, since 1987, were examined by using reli- 
able histological techniques (Cuellar et al., 1996). 
Therefore, it is believed that sex and maturation were 
correctly determined. In general, sex and maturity 
stages can be more reliably determined during the 
spawning season than in off-seasons. Data used for 
maturity analysis were only from May and June, the 
first part of the spawning season, during which er- 
rors in maturity determination were not expected. 
Furthermore, in the present study, all maturity 
stages of mature fish were pooled in only one matu- 
rity state, i.e. mature. As long as immature and ma- 
ture fish could be distinguished, inaccurate classifi- 
cation of mature substage would not introduce a bias 
in estimates of age and size at maturity. 
This study indicated that both age and length at 
sexual maturity of vermilion snapper declined over 
time. This decline may have resulted from increased 
fishing pressure, because the total landings consis- 
