DeMARTINI and FOUNTAIN: OVARIAN CYCLING FREQUENCY IN QUEENFISH 



H 1 1 \ 1 h 



cs 



FIGURE 4.— Mean gongd indices of adult female queenfish of five length classes collected March- August 1979. Sample sizes and 95% 



confidence limits of means are indicated. 



hydrated oocytes (Yamamoto and Yamazaki 1961; 

 Yamamoto and Yoshioka 1964; Macer 1974; Htun- 

 Han 1978), as illustrated in Figure 5. Histological 

 results verified our conclusions based on the 

 external appearance of ovaries (Table 4). 



Spawning frequency thus can be inferred from 

 the proportion of daytime sample fish whose 

 ovarian eggs are in a hydrated state. Throughout 

 the 1979 spawning season, an average of 13.69^ of 

 the females (of all sizes) present in daytime 

 samples were in a ready-to-spawn condition (Table 

 5); thus females spawn on average every 7.4 d. 

 Spawning frequency was similar for females of all 

 body sizes (Table 5). Confidence intervals of these 

 percent frequency occurrence data are necessarily 

 asymmetrical, since female spawners are conta- 

 giously distributed among sample fish and, like 

 hydrated-state females of the northern anchovy, 

 Engraulis mordax (Hunter and Goldberg 1980), 

 their sampling frequency is best described by the 

 normal-log negative binomial distribution. 



Sex Ratio 



The breeding adult sex ratio of S. politus was 

 1.04 males to 1.00 females. The mean percentage of 

 females among 16,794 adults was 49% with a 95% 

 confidence interval of ±0.8%. The sex ratio was 

 more skewed in favor of males in samples of fish 

 containing one or more females in ready-to-spawn 

 condition (Table 6). 



Batch Fecundity 



The number of eggs liable to be produced per 

 spawning, i.e., the potential batch fecundity, was 

 estimated for fish collected throughout the 1979 

 spawning season. The Gilson's fluid-fixed ovaries 

 of daytime sample fish of a complete range of body 

 sizes were used. Potential fecundity varied as 

 the cube of standard length, and was proportional 

 to body size in a consistent manner throughout 

 the season (Figure 6). Bimonthly regressions of 



551 



