202 



Fishery Bulletin 105(2) 



lower Laguna Madre), which formed a weakly sup- 

 ported clade. 



A statistically significant (P<0.001) correlation be- 

 tween D^. and geographic distance was observed (r=0.90) 

 and thus supported the isolation-by-distance hypothesis. 

 The greatest geographic distance between adjacent col- 

 lection sites was observed between Charlotte Harbor 

 and St. John's River; however, when the St. John's River 

 sample was excluded from the analysis, a significant 

 positive relationship was still evident (r=0.80, P<0.001). 

 The correlation between D^. and geographic distance 

 within the western Gulf of Mexico (Texas and Louisi- 

 ana) remained positive (r=0.42j and statistically sig- 

 nificant (P=0.05). 



Discussion 



Spotted seatrout inhabiting a series of sites from the 

 lower Laguna Madre of Texas to St. John's River on the 



Atlantic Coast of Florida were genetically differentiated 

 by analyses of allele frequencies of five microsatellite 

 markers. Samples from Florida waters were strongly 

 differentiated from spotted seatrout from Louisiana 

 and Texas, and the statistically significant correlation 

 between geographic distance and genetic distance was 

 due primarily to these differences. Within Florida, the 

 Charlotte Harbor spotted seatrout is genetically more 

 similar to the Atlantic Coast spotted seatrout from 

 St. John's River than to the neighboring Tampa Bay 

 spotted seatrout, which is genetically more similar to 

 Texas and Louisiana than to other Florida fish of the 

 same species. Differences in allele frequency between 

 Charlotte Harbor and Tampa Bay samples represented 

 the greatest discontinuity observed in this study. This 

 putative population structure is congruent with Wiley 

 and Chapman's (2003) findings of distinct population 

 subdivision among Atlantic Coast spotted seatrout, 

 although details of that structure may be difficult to 

 reconcile. Wiley and Chapman (2003) found spotted 



