650 



Fishery Bulletin 94(4), 1 996 



porgies, and sea basses) in the Carolinas have a pro- 

 tracted spawning season extending from early spring 

 to early fall (Hardy, 1978; Johnson, 1978; Grimes, 

 1987). The southeastern United States has temper- 

 ate and tropical characteristics, with surface water 

 temperatures from about 22°C to 29°C (Mathews and 

 Pashuk, 1986) and bottom temperatures of 15°C to 25°C 

 on the middle continental shelf (18-60 m) during sum- 

 mer (present study). Vermilion snapper fit the expec- 

 tation of being indeterminate spawners on the basis of 

 the temperate and tropical characteristics of this area 

 and on the basis of their long spawning season. 



Length was the best predictor of batch fecundity, 

 but great variability was seen in BF estimates for 

 same-size fish, perhaps because some oocytes may 

 have been spawned. This is not likely, however, be- 

 cause fish with hydrated oocytes free in the lumen 

 and fish with postovulatory follicles (POF's) were not 

 used in BF estimates. The monthly average BF ap- 

 peared to peak in August and decline in September. 

 However, time (months) did not have a significant 

 effect on BF estimates, and it is unlikely that differ- 

 ences in station locations within the sampling area 

 could account for the variation in BF. Davis and West 

 (1993) suggested that the variation seen in BF esti- 

 mates of Lutjanus vittus might be due to the collec- 

 tion of samples during various phases of the lunar 

 cycle. Lutjanus vittus seemed to follow a lunar 

 rhythm in spawning, as do many lutjanids (Grimes, 

 1987; Carter and Perrine, 1994). Thresher (1984) 

 suggested that lunar periodicities are probably uni- 

 versal in lutjanids. Environmental cues of spawning 

 events were not examined in this study; however, 

 Grimes ( 1976) found that vermilion snapper spawn- 

 ing correlated with temperature and photoperiod, but 

 not with the lunar cycle. 



Because of differences in fecundity estimation 

 methods, comparisons between the present study and 

 that of Grimes (1976), or with most other studies on 

 lutjanids, are not particularly useful (Grimes, 1987). 

 For example, Grimes (1976) estimated fecundity on 

 vermilion snapper taken off the Carolinas early in 

 the season. He counted all developing oocytes in each 

 female (the total count is approximately equivalent 

 to a PF estimate in this study). His fecundity equa- 

 tion for a 300-g fish results in a fecundity estimate 

 of 49,152 oocytes, compared with a PF of 126,408 

 oocytes for the same size fish in the present study. 



Collins and Johnson 4 used the hydrated oocyte 

 method (Hunter et al., 1985; Hunter and Macewicz, 



4 Collins, L. A., and A. G. Johnson. 1994. Spawning and an- 

 nual fecundity of gag, red snapper and vermilion snapper in 

 the northeastern Gulf of Mexico. Southeastern U.S. and Carib- 

 bean reef fish workshop: Charleston, South Carolina, 15-16 

 September 1994 [abstract). 



1985) for spawning frequency and annual fecundity 

 measurements of vermilion snapper in the northeast- 

 ern Gulf of Mexico. The spawning frequency (44 times 

 per year) of vermilion snapper in the Gulf exceeded 

 spawning frequency in this study (35 times per year). 

 Also, the annual fecundity of the vermilion snapper 

 in the Gulf (1.32 to 21.9 million eggs) was much 

 higher than the annual fecundity determined in the 

 present study for vermilion snapper in Atlantic wa- 

 ters off the southeastern United States (140,175 to 

 3.2 million eggs). The spawning season of vermilion 

 snapper in the Gulf is slightly longer than the spawn- 

 ing season of vermilion snapper in the southeastern 

 United States (April to late September), and this 

 additional time in the warmer waters of the Gulf may 

 result in the differences in annual fecundity esti- 

 mates between the two regions. The use of the hy- 

 drated oocyte method instead of the postovulatory 

 method to estimate spawning frequency may also ac- 

 count for the difference between the two studies. 



Larger fish in the Gulf study may also account for 

 some of the differences in annual fecundity estimates 

 between the two studies; however, the sizes offish stud- 

 ied by Collins and Johnson 4 were not available. The 

 largest female in the present study was 440 mm TL, 

 but the majority of the vermilion snapper were between 

 200 mm TL and 300 mm TL. It is generally accepted 

 that larger females produce more oocytes. Grimes and 

 Huntsman ( 1980) found that fecundity increased dra- 

 matically in larger (older) vermilion snapper. 



Davis and West ( 1993), using new assessment tech- 

 niques, found that the annual fecundity of a 300-mm 

 Lutjanus vittus was about 4.5 million oocytes (if it 

 spawned 90 times a season) or 7.6 million oocytes (if 

 it spawned 150 times a season) on the northwest shelf 

 of Australia. This annual fecundity estimate for 

 Lutjanus vittus is closer to the annual fecundity es- 

 timate for vermilion snapper ( 140,175 to 3.2 million 

 eggs) in this study than to the estimate for vermil- 

 ion snapper calculated by Collins and Johnson. 4 Ex- 

 amination of the annual fecundity estimates reveals 

 that the major difference between vermilion snap- 

 per and Lutjanus vittus is spawning frequency. Batch 

 sizes were relatively similar: a batch size for a 300- 

 mm L. vittus was about 71,000 oocytes and a batch 

 size for a similarly sized vermilion snapper was about 

 50,000 oocytes. The spawning season for Lutjanus vittus 

 is year round, and most spawning occurs from Septem- 

 ber to April. The longer spawning season of L. vittus 

 accounts for a major portion of the large difference in 

 spawning frequency between the two species. 



In the vermilion snapper of our study, POF's could 

 not be given an exact age because of a lack of 24-h 

 sampling and because of small sample sizes; there- 

 fore time of day for spawning could not be deter- 



