846 



Fishery Bulletin 92(4). 1994 



Table 3 



Nested analysis of variance on numbers of hydrated oo- 

 cytes per gram ovary weight among ovarian regions (six 

 total-three per lobe), between ovarian lobes (right and left), 

 and among six female red drum captured 26 September 

 1988 from the northern Gulf of Mexico. MS=mean square. 



Source of Variation df 



MS 



F-value P>F-value 



Individuals 



Lobes 



Ovarian regions 



Error 



Total 



5 



6 



24 



36 



72 



42,580,790 



242,773 



2,290,678 



4,266,679 



49,380,920 



69.86 

 0.33 



0.78 



0.0001 

 0.9155 

 0.7318 



Table 5 



Comparison of red drum seasonal spawning frequencies 

 (SF, expressed as average days between successive 

 spawnings) estimated with the postovulatory follicle (POF) 

 method of Hunter and Macewicz ( 1985 ) and the time-cali- 

 brated (TO method of Fitzhugh et al. (1993). Day-0 fe- 

 males are those evidencing yolk coalescence or hydration 

 of oocytes; day-1 females are those with postovulatory fol- 

 licles from previous day's spawning. All specimens collected 

 by purse seine from 13 August through 8 October. 



POF method 



TC method 



Year 



Mature 

 females 



Females 

 with POF 



SF 



Day-0 Day-1 

 females females 



SF 



tion within classes. Exclusion from the regression 

 analyses of those specimens captured during Oc- 

 tober, based on the possibility of declining output 

 toward the end of the spawning season as sug- 

 gested by Fitzhugh et al. (1988), produced r 2 val- 

 ues ranging from 0.23 to 0.39. 



Spawning frequency 



Red drum spawning frequencies estimated with 

 the postovulatory follicle method were highly vari- 

 able among years (Table 5) ranging from one 

 spawning event every 2.8 days in 1986 to one 

 spawn every 80.0 days in 1991. A total of 65 of 

 572 sexually mature females captured during the 

 1986-1992 spawning seasons evidenced postovu- 

 latory follicles for a seven season average spawn- 

 ing frequency of 8.8 days. Spawning frequencies 

 calculated with the time-calibrated method (Table 

 5) showed less variation and gave more plausible 

 estimates. Except for the 1992 spawning season, 

 when sampling was limited to three dates during 

 the spawning season (28 August, 3, 12 Septem- 

 ber), spawning frequencies of one spawn every 2- 

 4 days were predicted. 



Discussion 



Aspects of red drum reproductive biology in the 

 Gulf of Mexico have been variously inferred from 

 visual observation of gonadal development, from 

 larval and juvenile abundances and lengths, and 

 from histological documentation of ovarian devel- 

 opment. Given both the disparities and subjectiv- 

 ity inherent among, and even within, these meth- 

 odologies and the expanse of the Gulf of Mexico, it 

 is not surprising that published accounts of red 

 drum reproduction vary widely and, perhaps, geo- 

 graphically. We agree with West (1990) that his- 

 tological methods produce the most accurate and 

 most reliable results in assessing ovarian devel- 

 opment and predicting reproductive variables. We 

 also stress that our findings should not be bm;idly 

 applied to red drum populations throughout the 

 Gulf of Mexico. 



The sex ratio of the 2,364 mature individuals 

 taken by purse seine is undoubtedly most reflec- 

 tive of the offshore spawning stock of red drum in 

 the northern Gulf of Mexico. Given that most of 

 our specimens taken by purse seine were captured 

 either just prior to or during the spawning season 

 and considering the substantial sample size, our 

 data establish a 1:1 sex ratio in schools of pre- 

 spawning and spawning red drum. This supports 



