Cuellar et al.: Reproductive seasonality, maturation, fecundity, and spawning frequency of the vermilion snapper 



649 



year. From visual observations of histological slides, 

 it appeared that the duration of HO's and new POF's 

 might be more similar than the duration of new POF's 

 and old POF's. Thus, we decided to test the popula- 

 tions statistically to see if this might be true. There 

 was no difference in the frequencies of HO's and new 

 POF's (Mann-Whitney, n=23, P=0.675), but there was 

 a significant difference in the frequencies of new 

 POF's and old POF's (Mann-Whitney, n=23, P= 

 0.00598). Thus, HO's and new POF's probably have 

 similar durations that allow us to combine the cat- 

 egories statistically to increase the sample size. An 

 overall average was recalculated by summing only 

 two categories ((HO+new POF)/2, and old POF) and 

 by dividing by two. This overall average was then 

 multiplied by days in the spawning season to deter- 

 mine a new spawning frequency. Combining catego- 

 ries results in a spawning frequency of approximately 

 every five days or 35 times per year. The duration 

 between new POF's and old POF's was long enough 

 that the combination of categories was not statisti- 

 cally possible. 



Discussion 



Hunter et al. ( 1992) listed four lines of evidence that 

 indicate determinate fecundity (see Materials and 

 Methods section). Vermilion snapper, however, is an 

 indeterminate spawner because it does not demon- 

 strate any of the four characteristics given for deter- 

 minate fecundity. Each line of evidence is addressed 

 in detail in the following section. 



The oocyte size-frequency distribution is continu- 

 ous in vermilion snapper (without a hiatus between 

 any oocyte stages) except for the hydrated batch. 

 Traditional evidence for determinate fecundity is 1) 

 the obvious gap in oocyte maturity stages or size 

 classes among the oocytes that are maturing during 

 the season and 2) the reservoir of less mature oo- 

 cytes present year-round in the ovary (Yamamoto, 

 1956). Although the absence of such a hiatus is evi- 

 dence for indeterminate fecundity, in some cases fish 

 with a continuous oocyte distribution are still con- 

 sidered to have determinate fecundity (Hunter and 

 Macewicz, 1985). In such cases, the product of batch 

 fecundity multiplied by spawning frequency needs 

 to be compared with the standing stock of mature 

 oocytes at the beginning of the season. For example, 

 a 300-g vermilion snapper in this study produces 

 approximately 126,408 oocytes on the basis of the 

 potential fecundity equation, whereas a 300-g female 

 produces approximately 1.7 million oocytes accord- 

 ing to the product of BF multiplied by spawning fre- 

 quency. It is quite obvious from this study that the 



product of BF multiplied by spawning frequency must 

 be used to estimate annual fecundity. 



The mean oocyte diameter of vermilion snapper 

 stayed constant over the spawning season. Species 

 with determinate fecundity would produce larger 

 oocytes as the season progressed because, as matu- 

 ration continued during the spawning season, no new 

 oocytes would be recruited into the advanced stock, 

 and eggs that were present would be developing and 

 increasing in size. In contrast, indeterminate spawn- 

 ers are continually producing oocytes during the 

 spawning season, with smaller immature oocytes 

 replacing those that mature and are spawned, re- 

 sulting in a constant MOD. 



Total fecundity and BF in vermilion snapper did 

 not consistently decline over the spawning season 

 as would be expected if the vermilion snapper was a 

 determinate spawner. Total fecundity and BF actu- 

 ally increased through August and declined in Sep- 

 tember in this study. No trends in TF existed over 

 the months of the spawning season. Atresia also had 

 no significant effect on fecundity estimates to account 

 for a decrease in the number of oocytes. Similarly, 

 Davis and West ( 1993 ) found that the BF of Lutjanus 

 vittus did not decline over the spawning season. Ver- 

 milion snapper in the northern central Gulf of Mexico 

 appear to have a similar trend of peak spawning in 

 late summer according to larval abundances found 

 in that area (Comyns and Lyczkowski-Shultz 3 ). 

 Grimes ( 1976) found peak spawning activity (accord- 

 ing to gonsomatic index [GSI] values) in August off 

 the Carolinas, and Nelson (1988) found peak activ- 

 ity in the Gulf during the summer. This trend to- 

 ward peak spawning during middle to late summer 

 would be expected for an indeterminate spawner with 

 continuous development of oocytes over the repro- 

 ductive season. 



The order of spawning batches was not consistent 

 with the determinate fecundity prediction. Of 750 

 females examined histologically, three had hydrated 

 oocytes (HO's) in the ovary in the absence of any other 

 advanced-stage yolked oocyte (AYO) batch, and only 

 16 had a batch of early migratory nucleus-stage oo- 

 cytes (MNO's) and a late HO batch. Many of the fe- 

 males sampled from September contained all mature 

 oocyte stages, except in the HO batch. 



According to Hunter and Macewicz ( 1985), deter- 

 minate fecundity is generally found in species in 

 cooler climates and with shorter spawning seasons. 

 Many fishes (snappers, grunts, jacks, dolphin, some 



3 Comyns, B. H., and J. Lyczkowski-Schultz. 1993. Spawning 

 and early life history of snappers in the northcentral Gulf of 

 Mexico. Proceedings of the sixth annual MARFIN conference; 

 Atlanta, Georgia, 12-13 October 1993, p. 1-3. 



