142 



Fishery Bulletin 104(1) 



seems unlikely, because the ratio of males and females 

 across three years of red snapper catch data was 0.97 

 and did not differ significantly from unity (Nieland^-). 

 Variation in population number and in year-class 

 strength, alternatively, seems likely, given the annual 

 differences in commercial and recreational landings and 

 the annual differences in abundance of age-0 and age- 

 1 red snapper, respectively (Schirripa and Legault^'*). 

 Variance in reproductive success is far more difficult to 

 assess but can include mating systems (Nunney, 1993) 

 that lead to differences in reproductive success between 

 males and females, and a "sweepstakes" process (Hedge- 

 cock, 1994) where size-dependent fecundity, combined 

 with random but family-specific early mortality (Hauser 

 et al., 2002), leads to a large variance in the number of 

 (surviving) offspring per parent. The latter could be ef- 

 fected in red snapper by nonrandom removal of related 

 subadults or juveniles either by localized overfishing 

 or by shrimp trawling. Finally, variance in productiv- 

 ity among habitats across the northern Gulf can be 

 inferred from subregional differences in red snapper 

 growth rates (Fischer et al., 2004) and from subregional 

 ecological differences (Gallaway et al., 1998) that dis- 

 tinguish the northeastern Gulf from the northwestern 

 Gulf. Future ecological and behavioral studies to gen- 

 erate estimates of variance in individual reproductive 

 success or variation in productivity among localities are 

 clearly warranted. 



Demographic stocks 



The differences in variance effective size (A''^,i) among the 

 geographic samples of red snapper indicate present-day 

 differences in demographic dynamics that may include 

 the number of individuals that produce surviving off- 

 spring, and hence by inference, census size. The factors, 

 ecological or otherwise, promoting these demographic 

 differences are difficult to assess but likely relate in 

 some way to variation in food availability, habitat qual- 

 ity, or mortality (or a combination of all three factors). 

 Accordingly, one might expect one or more of these 

 factors to differ among the sample localities, given the 

 differences in variance effective size among localities. 

 In addition, one might expect other demographic param- 

 eters to differ as well. 



Our study was part of a larger, multidisciplinary proj- 

 ect that involved studies of age-and-growth and repro- 

 duction of red snapper at the three localities. The age- 

 and-growth studies of Fischer et al. (2004) documented 

 that fork length, total weight, and age-frequency dis- 

 tributions differed significantly among localities. Red 

 snapper sampled at the Texas locality were significantly 



'- Nieland, D. 2005. Personal commun. Coastal Fisheries 

 Institute, Louisiana State University, Baton Rouge, LA 

 70803-7503. 



" Schirripa, M. J., and C. M. Legault. 1999. Status of the 

 red snapper in U.S. waters of the Gulf of Mexico. Report 

 SFD-99/00-75, 86 p. Southeast Fisheries Science Center, 

 75 Virginia Beach Drive, Miami, FL 33149-1099. 



smaller at age and reached smaller maximum size than 

 did red snapper sampled at the Louisiana and Alabama 

 localities; fish sampled at the latter two localities did 

 not differ in size-at-age or maximum size. There also 

 was a significantly higher proportion of smaller, young- 

 er fish at the Texas locality than at the other two. The 

 studies of reproductive capacity (Woods et al., 2003) 

 involved only fish sampled from the Louisiana and 

 Alabama localities but revealed that females sampled 

 from the Alabama locality reached sexual maturity at a 

 smaller size and younger age than did females sampled 

 from the Louisiana locality. The differences in growth 

 rate are likely a function in part of the more produc- 

 tive, nutrient-rich waters found at the Louisiana and 

 Alabama localities and caused by the plume produced 

 from the Mississippi River (Fischer et al., 2004), a hy- 

 pothesis reinforced by the observation (Grimes, 2001) 

 that between 70-80'7f of fishery landings in the north- 

 ern Gulf of Mexico come from waters surrounding the 

 Mississippi River delta. The differences in female age 

 and size at maturity, alternatively, are thought to indi- 

 cate a stressed population and to reflect a compensatory 

 response to growth overfishing or declining population 

 size, or a response to both (Trippel, 1995; Woods et al., 

 2003). Collectively, the life-history differences and the 

 differences in genetic-based estimates of effective size 

 strongly suggest that red snapper at the three localities 

 represent three different, demographic stocks. 



A critical issue is whether the demographic differ- 

 ences in life history observed among red snapper in the 

 northern Gulf are genetic or phenotypic (environmen- 

 tally induced) in origin. Most discussions of stock struc- 

 ture in commercially exploited marine fishes involve 

 an explicit genetics component (Gold et al., 2001a), and 

 typically, the absence of genetic heterogeneity within a 

 fishery leads to management planning for a single unit 

 stock. However, life-history traits can change rapidly in 

 response to environmental pressures (e.g., size-selective 

 fishing), and it has been hypothesized that the pool of 

 genotypes that code for life-history traits is a highly 

 dynamic property of populations, and moreover, that lo- 

 cal adaptation(s) differentiating populations can evolve 

 even in the presence of extensive gene flow (Conover et 

 al., 2005). Thus, demographically different stocks could 

 differ genetically, but not necessarily in selectively neu- 

 tral markers that respond primarily to the interaction 

 between gene flow and genetic drift. The issue also is of 

 importance to management planning because phenotypi- 

 cally plastic responses due to environmental differences 

 generally can be reversed fairly quickly, whereas genetic 

 responses are typically much slower (Hutchings, 2004; 

 Conover et al., 2005). 



The geographic differences in red snapper in growth 

 rates and shifts in timing of female maturity in all 

 likelihood are due to a mix of genetic and environmen- 

 tal factors, as are most life-history traits in a variety 

 of animal species, including fishes (Mousseau and Roff, 

 1987; Conover and Munch, 2002). A significant genetic 

 component to growth rate is well documented in a va- 

 riety of fishes under aquaculture (Dunham et al., 2001) 



