838 



Fishery Bulletin 97(4), 1999 



100 mm less than the average length of 464 mm re- 

 ported by Nelson (1988). In the SAB, the average 

 total length of 6-year-old fish had decreased from 451 

 mm in 1972-74 (Manooch and Huntsman, 1977) to 

 363 mm in 1991-94 (Harris and McGovern, 1997). 

 Both Harris and McGovern (1997) and Hood and 

 Johnson^ concluded that the observed decrease in 

 average length at age was likely due to size selective 

 fishing. 



Mortality 



The age of full recruitment to both the recreational 

 and commercial fisheries has increased since the 

 early 1980s. Zastrow (1984) and Nelson (1988) re- 

 ported the age at full recruitment to be 4 years, less 

 than our estimate of 6 years. The reported age of full 

 recruitment to the SAB hook-and-line fishery was 4 

 years before 1981 (Grimes, 1978; Huntsman et al., 

 1983; Zhao et al., 1997) but increased to 6 years by 

 1993 (Cuellar et al., 1996; Zhao et al., 1997). This 

 increase in age of recruitment has been associated 

 with a decrease in length at age attributed to selec- 

 tive fishing on faster-growing members of each age 

 group (Zhao et al., 1997). Our estimates of mortality 

 (0.469-0.489) are within the range given by 

 Schirripa^ for the years 1986-1995 in the GOM 

 (0.466-0.791) and less than the estimate from the 

 SAB headboat fishery (0.67) reported by Huntsman 

 etal. (1983). 



Reproduction 



Our observed sizes of maturation were less than those 

 reported by Nelson ( 1988) for the western GOM. We 

 found that most females were sexually mature at 200 

 mm TL (age 1), and we did not examine any imma- 

 ture males (the smallest male we sampled was 199 

 mm TL). The smallest mature female and male re- 

 ported by Nelson (1988) were 234 mm TL and 291 

 mm TL, respectively. In the SAB, Collins and 

 Pinckney (1988) found that at 160 mm TL, 60% of 

 females and 90% of males were mature. Cuellar et 

 al. (1996) sampled females as small as 186 mm TL 

 and males as small as 197 mm TL and did not find 

 any immature fish in their samples. These lengths 

 at maturity are in contrast to the results of Grimes 

 and Huntsman (1980) who found that SAB vermil- 

 ion snapper matured between 186 and 324 mm TL 

 (ages 3 and 4). There is an indication that increas- 

 ing fishing pressure may depress vermilion snapper 

 size and age at maturity. Zhao and McGovern ( 1997) 

 noted a decrease in the size at maturity for males 

 and females in the SAB from 1970 to 1992. Before 

 1982, 31% of males and 5% of females were mature 



E 

 E. 



c 

 a 



2 

 o 



0) 



2 4 6 8 10 12 14 



Age (yr) 



Figure 9 



Mean empirical total length at age for vermilion snapper 

 from the eastern and western Gulf of Mexico. Scale-based 

 ages from the western Gulf of Mexico by Zastrow ( 1984; 

 circle, solid line) and Nelson (1988; square). Scale-based 

 ages from Barber (1989) from the eastern (diamond) and 

 western (hexagon) Gulf of Mexico. Whole-otolith-based ages 

 from Barber (1989) from the eastern (triangle) and west- 

 ern (inverted triangle) Gulf of Mexico. Sectioned-otolith- 

 based ages from this study (circle, dashed line). 



at 140 mm TL. After 1982, all males and 37% of fe- 

 males were mature at 140 mm TL. They suggested 

 that this change may have been caused by the in- 

 creasing and selective fishing pressure that occurred 

 during the 1980s. 



On the basis of the presence of ripe females and 

 increases in GSI, we believe that vermilion snapper 

 in the GOM spawn from May to September. Nelson 

 ( 1988) and Collins'' also reported summer spawning. 

 In our samples, ripe gonads were present from late 

 spring to early fall but were most common from May 

 to July. In addition, GSIs were highest during the 

 summer months. Spawning seasonality in the GOM 

 is similar to that reported for the SAB, where fish 

 spawn from the late spring to early fall (Grimes and 

 Huntsman, 1980; Cuellar et al., 1996). Boardman and 

 Weiler (1979) found that in Puerto Rican waters, 

 spawning takes place year round. 



Batch fecundity was positively correlated with fish 

 length. Nelson ( 1988) and Collins^ also found a posi- 

 tive relationship between batch fecundity and length 

 for GOM fish. Their estimates were larger than those 

 obtained in our study (61,600-392,000 and 33,550- 

 415,161 oocytes, respectively); however, the fish 

 sampled were also larger ( 209-510 and 248-375 mm 

 TL, respectively). Our estimates of batch fecundity 

 were similar to SAB estimates reported by Cuellar 



