140 



Fishery Bulletin 104(1) 



Estimates for the samples from Texas (A/py=1098) and 

 Alabama (N^,y=1235) were essentially the same, falling 

 well v/ithin the 95% confidence intervals of one another. 

 An exact, maximum-likelihood estimate could not be 

 generated for the sample from Louisiana because the 

 value of N^.y with highest likelihood was >75,000 and 

 the likelihood of higher values of N^,y could not be com- 

 puted. This estimate is more than an order of magni- 

 tude greater than the N ^• estimates for the other two 

 localities and is significantly higher than those based 

 on 95% confidence intervals. 



Discussion 



Genetic population structure 



Results obtained from pairwise exact tests indicated 

 that the majority of genetic differentiation detected 

 among the twelve spatial-temporal samples of red snap- 

 per was due to allele and genotype distributions in 

 the 1995 cohort sampled from the northwestern Gulf 

 (Texas) and in the 1997 cohort sampled from the north- 

 eastern Gulf (Alabama). In addition, exact tests among 

 cohorts sampled in 1997 were nonsignificant and F^y. 

 values among localities (both cohorts) averaged less than 

 0.001. These results indicate that the genetic differences 

 observed in the present study are temporal (between 

 cohorts within localities I and not spatial (among locali- 

 ties). A "hint" of spatial differentiation was suggested by 

 assignment tests. A total of 53% offish were reclassified 

 (assigned) to their original locality, a proportion that 

 differed significantly from that expected if genotypes 

 were distributed randomly among localities. However, 

 for 98% of the fish, none of the three localities could be 

 unequivocally excluded as the locality of origin. 



The above results are in general agreement with oth- 

 er, genetics-based studies of red snapper in the northern 

 Gulf in that little to no significant geographic hetero- 

 geneity in genetic markers, ranging from allozymes to 

 mtDNA to microsatellites, has been detected (John- 

 son^; Gold et al., 1997; Garber et al., 2004; Gold et 

 al., 2001b). The one exception was a study by Bortone 

 and Chapman* where significant heterogeneity in both 

 temporal and spatial restriction-fragment patterns of 

 the mitochondrially-encoded 16S ribosomal (r)RNA 

 gene was reported. Bortone and Chapman'* suggested 

 that the observed genetic heterogeneity likely stemmed 

 from nonrandom sampling where individuals related 

 by descent had remained in close spatial proximity to 

 one another. In general, the "consensus" inference has 

 been that gene flow among present-day red snapper 



in the northern Gulf is sufficient to offset divergence 

 by genetic drift of the (presumed) selectively neutral 

 genetic markers assayed. Such gene flow could involve 

 movement of adults (Patterson et al., 2001), hydrody- 

 namic transport of pelagic eggs and larvae (Goodyear-'), 

 or both. 



The foregoing notwithstanding, there are a number of 

 caveats (discussed in Pruett et al., 2005) to the infer- 

 ence that significant gene flow occurs among present- 

 day red snapper in the northern Gulf Briefly, tag-and- 

 recapture and ultrasonic-tracking studies (Fable, 1980; 

 Szedlmayer and Shipp. 1994; Szedlmayer, 1997) have 

 indicated that adult red snapper are largely sedentary 

 and nonmigratory. Significant movement of adults in 

 the northeastern Gulf was reported by Patterson et 

 al. (2001), but movement per se was mostly unidirec- 

 tional (west to east) and the average distance covered 

 in roughly a year was only -30 kilometers. Movement 

 of (pelagic) red snapper eggs and larvae likely occurs, 

 but neither egg nor larval type nor length of larval life 

 are effective predictors of gene flow in marine fishes 

 (Shulman and Bermingham, 1995) and larval exchange 

 rates of marine species generally appear overestimated 

 (Cowen et al., 2000). In addition, regardless of the life- 

 history stage at which gene flow might occur in red 

 snapper, movement across the continental shelf should 

 be more-or-less linear and would be expected to fol- 

 low a pattern of isolation by distance where fish from 

 proximal localities are more similar genetically than 

 fish from more distal ones. However, the correlations 

 between genetic and geographic distance expected from 

 isolation by distance have not been found (Gold et al., 

 1997; 2001b; this article). Finally, salient differences 

 in geologic structure, habitat structure, and ecological 

 conditions (Rezak et al., 1985; Gallaway et al., 1998), 

 significant differences in salinity due to freshwater 

 outflow from river systems in the northcentral Gulf 

 (Morey et al., 2003), and the present-day occurrence 

 during the summer months of a major hypoxic zone 

 that extends out along the continental shelf from the 

 Mississippi Delta westward (Rabalais et al.'°; Ferber, 

 2001) potentially could serve as barriers to movement 

 and gene flow. 



Despite these caveats, the bulk of the genetics data 

 has indicated essentially no difference among present- 

 day red snapper sampled across the northern Gulf This 

 is consistent with the unit stock hypothesis and with 

 the inference that observed genetic homogeneity is due 

 to substantial gene flow. However, it is important to 



' Bortone, S. A., and R. W. Chapman. 1995. Identification 

 of stock structure and recruitment patterns for the red snap- 

 per, Lutjanus campechanus. in the Gulf of Mexico. Final 

 report for Marfin Program Grant Number NA17FF0379-03, 

 39 p. Southeast Regional Office, National Marine Fisher- 

 ies Service, 263 13"^ Avenue South, Saint Petersburg, FL 

 33701. 



•' Goodyear, C. P. 1995. Red snapper stocks in U.S. waters 

 of the Gulf of Me.xico, 171 p. National Marine Fisheries 

 Service. SE Fisheries Center, Miami Laboratory, CRD 95/96- 

 05, 75 Virginia Beach Drive, Miami, FL 33149-1099. 



i"Rabalais, N. N., R. E. Turner, D. Justic, Q. Dortch, and W. 

 J. Wiseman. 1999. Characterization of hypoxia: topic I 

 report for the integrated assessment on hypoxia in the Gulf 

 of Mexico. NOAA Coastal Ocean Program Decision Analy- 

 sis Series No. 15, 203 p. NOAA Coastal Ocean Program, 

 1305 East-West Hwy,. N/SC12 SSMC4, Silver Spring, MD 

 20910. 



