128 
Fishery Bulletin 109(1 ) 
300 400 500 600 700 800 900 1000 1100 
Total length class (mm) 
Figure 2 
Proportions of immature bisexual, immature female, mature female, and mature male Hawai- 
ian grouper ( Hyporthodus quernus) present in each 10-mm length class of the 600-specimen 
subsample used for histological examination. Numbers above bars indicate numbers of speci- 
mens examined. 
five mean and median body lengths of all adult female 
specimens were 703 and 700 mm TL. The smallest and 
largest observed adult female individuals were 328 and 
977 mm TL. The smallest female whose ovaries con- 
tained hydrated oocytes was 492 mm TL. 
Sex change Body size at adult female-to-male sex 
change was appraised using all recognizably mature 
adult fish. The body length-at-female-to-male sex change 
relation was best described by the logistic equation 
P 50 = 100 / (1 + exp (0.0153— 13. 653 
with SE a =0.002, SE b =1.858, r 2 = 0.80, PcO.0001, and 
n-397. 
Body length (L 50 ) at female-to-male sex change was 895 
±20 mm TL (95% Cl) (Fig. 5). The mean and median 
body lengths of all adult male specimens were 895 and 
891 TL, respectively. The smallest and largest adult 
male fish encountered were 753 and 1103 mm TL. The 
five transitional fish ranged from 760 to 913 mm TL. 
Many (38%) of the mature specimens were between the 
sizes of the smallest male and the largest female fish. 
Spawning seasonality 
The slopes of the relations between gonadal and body 
weights and between gonadal weights and body lengths 
did not differ among females in prespawning, spawn- 
ing, and postspawning maturation stages (ANCOVA of 
GFBW and maturation stage effects on GW: GFBW- by- 
maturation and TL-by-maturation interaction effects — 
F 2 327=2-36 and 2.30, respectively; accept H 0 : slopes 
equal at P=0.10). The intercepts of these same rela- 
tions, however, differed (ANCOVA: maturation stage 
effects — F 2 3 29 =114.0 and 113.2, respectively; reject H o : 
intercepts equal at P<0.0001). The relation between 
gonadal weight and body length provided the better fit 
and TL was used in subsequent analyses of monthly 
patterns. In these analyses we used RGI s, in which 
the pooled slope of the GW-to-TL relation (2.69266) was 
the scaling factor. RGIs indicated that most females 
began ripening in December and maintained elevated 
ovarian masses of 1-3% GFBW through April (1-way 
ANCOVA: month effect on RGI — F u 324 =6.86; reject 
H o : all months equal at PcO.0001). Ovary weights 
peaked in January-March (REGW multiple-range test: 
January-March highest; P<0.03). Analyses with simple 
