Fitzhugh et al.: Reproductive biology of Pogonias cromism Louisiana 



25! 



sissippi River. Drumming behavior associated with 

 spawning was noted in Caminada Pass, Louisiana in 

 April, further documenting inshore reproductive activ- 

 ity (Donald Baltz, Coastal Fish. Inst., LA State Univ., 

 Baton Rouge, pers. commun. ). 



The dynamic of changing spawning locations may 

 have a seasonal component related to water tempera- 

 ture. Mok & Gilmore (1983) analyzed sound produc- 

 tion from black drum in Florida waters and noted "loud 

 drumming," which they associated with spawning, oc- 

 curring at 18-20°C. They also noted cessation of this 

 drumming during a temperature drop to 13-15°C. Pe- 

 ters & McMichael ( 1990) provided more direct evidence 

 for onset of spawning at 16-20°C. By correlating sea- 

 sonal water temperature with larval birthdates, peak 

 births were calculated to have occurred in March when 

 temperatures reached 21-24°C. Although we did not 

 have precise locations and temperatures for commer- 

 cial catches, our samples of hydrated-oocyte and POF- 

 bearing females indicate that spawning may have pre- 

 dominated in outside waters in February (e.g., trawl 

 landings) and moved to inside waters as seasonal tem- 

 peratures increased (haul-seine and gillnet landings 

 in March and April ). Other factors not examined may 

 influence spawning, including moon phase and tidal 

 period (Peters & McMichael 1990). 



Postovulatory follicles probably last longer than 24 h 

 at sea temperatures encountered in coastal Louisiana 

 waters in February and March (19-22°C). Hunter & 

 Macewicz (1985) found POFs for 3-4 d from northern 

 and peruvian anchovies (Engraulis mordax and E. 

 ringens) spawning at 13-19°C. Based on a higher 

 spawning temperature of 23-24°C, Hunter et al. ( 1986) 

 found 24 h-old follicles in skipjack tuna that appeared 

 similar to those in northern anchovy held 48 h. Our 

 day-1 POFs appeared similar to 24 h POFs for skip- 

 jack tuna shown in Hunter et al. (1986). We estimated 

 black drum follicle duration to be at least 32 h old, due 

 to the presence of recent POFs and older-degenerating 

 POFs together in the same histological sections (i.e., 

 0-8 h plus 24 h, respectively) which is consistent with 

 spawning on successive nights. If follicles are identifi- 

 able well past 24 h, our estimate of average duration 

 between spawning would increase. 



Our estimates of spawning frequency, once every 3 

 or 4d, are similar to other sciaenid species. Tucker & 

 Faulkner (1987) report a daily spawning fraction of 

 0.35 (once every 3d) for captive spotted seatrout. 

 Brown-Peterson et al. (1988) calculated an average 

 daily percentage of wild seatrout in ripe condition to 

 be 27.5% (indicating spawning once every 3.6 d) over a 

 6 mo reproductive season. Red drum have displayed a 

 spawning fraction of 0.68 (once every 1.5 d) in captiv- 

 ity over a 76 d period (Arnold et al. 1977). 



The pattern of appearance of vitellogenic oocytes sup- 

 ports our contention that oocyte recruitment contin- 

 ued during the reproductive season. Therefore, the 

 yolked-oocyte population was not deterministic or rep- 

 resentative of annual fecundity (Hunter et al. 1985). 

 Of females examined histologically in February, 73% 

 showed evidence of recent spawning (day-0 females), 

 yet the proportion of vitellogenic oocytes from females 

 did not reach a maximum until March. In contrast, a 

 species with a determinant oocyte development pat- 

 tern could exhibit multiple spawns but the proportion 

 of vitellogenic oocytes would decrease following onset 

 of spawning (Hunter et al. 1985). With continuous 

 recruitment of batches of oocytes, the traditional 

 method to determine fecundity by enumerating 

 vitellogenic oocytes prior to onset of reproduction (e.g., 

 Bagenal 1968) would underestimate potential fecun- 

 dity. This method necessitates counting of hydrated 

 eggs just prior to ovulation, i.e., determining batch 

 fecundity and number of spawns in the season (Hunter 

 etal. 1985). 



Previously, little fecundity information has been re- 

 ported for black drum (Sutter et al. 1986). Pearson 

 (1929) estimated fecundity at 6 million eggs for one 

 110 cm gravid female (107cmFL, 16.4 kg eviscerated 

 weight)* based on extrapolation of wet weight for 60 

 eggs. Our computation of batch fecundity was 1.6 mil- 

 lion eggs for a 6.1kg female (mean eviscerated weight 

 of 25 females). Figure 4 and comparison with Pearson's 

 result suggest that batch fecundity is a function of 

 body size, but we found this relationship to be quite 

 variable (Fig. 4). While these data do not appear to fit 

 a linear relationship as closely as for smaller sciaenids 

 (DeMartini & Fountain 1981, Brown-Peterson et al. 

 1988), we only sampled hydrated females measuring 

 660-876 mmFL. Because females >1000mm are occa- 

 sionally landed (e.g., Beckman et al. 1990), a broader 

 range of sizes could illuminate the functional relation- 

 ship between size and fecundity. Batch size may vary 

 also with the reproductive period and may be higher 

 earlier in the spawning season. Our sample size was 

 too small to detect changes in batch fecundity over the 

 breeding season. However, the pattern of vitellogen- 

 esis and GSI provides evidence that spawning peaked 

 in March. The proportion of females in spawning con- 

 dition was also highest in March. Conover ( 1985) dem- 

 onstrated a quadratic batch-fecundity relationship for 

 Atlantic silversides during the breeding season. He 

 postulated that this pattern may occur when optimal 



* Calculated from TL-FL regression reported in Murphy & Taylor 

 1 1989). The relationship between eviscerated body weight (BW) and 

 fork length (FL) is given by the equation: Ln BW = 2.8835 Ln FL - 

 10.382 <n = 100, r 2 =0.964, range 459-1122 mmFL). 



