Crabtree and Bullock: Life history of Mycteroperca bonaci 
745 
Figure 7 
(A) A transitional gonad from a 1030-mm-TL black grouper, 
Mycteroperca bonaci , captured in September 1994 with prolif- 
erating spermatogenic tissue in sperm cysts (S), primary growth 
stage oocytes (PG), degenerating vitellogenic oocytes (VO), and 
PAS-positive melanomacrophage centers (PAS). (B) A 947-mm- 
TL black grouper captured in November 1995 with proliferat- 
ing spermatogenic tissue in sperm cysts (S), primary growth 
stage oocytes ( PG), degenerating vitellogenic oocytes (VO ), and 
PAS-positive melanomacrophage centers (PAS). (C) The same 
ovary as in B showing the presence of mature sperm in periph- 
eral sperm sinuses (SS). Scale bars = 200 microns. 
however, as annuli became progressively more 
closely spaced, age estimation became more dif- 
ficult. Additional work is needed to assess the 
accuracy of our age estimates for older grouper; 
consequently, our age estimates predicted lengths 
at age, and growth model parameters should be 
used with caution. From their analyses of mar- 
ginal increments, Manooch and Mason (1987) 
also reported that black grouper from the Florida 
Keys form a single annulus each year. They found 
that annuli were usually formed during March- 
May, about a month earlier than we estimated. 
This difference is probably a result of differences 
in interpretation of the appearance of an annu- 
lus present on the otolith margin. Other conge- 
ners also form annuli during late spring and early 
summer. Collins et al. (1987) reported that gag, 
M. microlepis, off the southeastern U.S. coast 
form annuli during late spring to late summer, 
and Hood and Schlieder (1992) suggested that 
gag form annuli during summer in the eastern 
Gulf of Mexico. Bullock and Murphy (1994) re- 
ported that yellowmouth grouper, M. inter- 
stitialis, form opaque bands during late summer 
and early fall. Matheson et al. (1986) found that 
scamp, M. phenax, in North Carolina waters form 
an annual mark during April or May. 
We rejected 12.5% of the otoliths we sectioned 
as unreadable, and this is a possible source of 
bias to our growth model parameters. Otolith 
weight is a useful predictor of age of black grou- 
per (r 2 =0.952), and the otoliths that we rejected 
were generally heavier and thus probably older 
than most of those that we considered readable. 
In addition, although the length-frequency dis- 
tribution of fish whose otoliths were rejected was 
not significantly different from that of all fish 
whose otoliths were readable, the significance 
level of the test (P=0.062) was close enough to 
0.05 to cause us to suspect that the distributions 
could have been different. If so, we may have re- 
jected as unreadable more otoliths from large 
grouper than from small grouper. We may also 
have tended to exclude slower-growing older 
grouper from our sample of aged fish, and this 
could have biased our growth models. However, 
when we recalculated growth models and in- 
cluded all fish regardless of CV, the resulting 
growth-parameter estimates were all within one 
standard error of those in Table 3. Thus any bias 
to the growth model caused by our rejection of 
otoliths from older grouper appears to be negli- 
gible. Our choice of a threshold CV of 12% was 
arbitrary, but growth model parameters did not 
appear to be sensitive to the choice of CV thresh- 
