134 



Fishery Bulletin 98(1) 



Table 1 (continued) 



Gulf 



Atlantic 



Haplotype 



AP 



TB 



OF 



SA 



CH 



FB 



IR 



ML 



PO 



TP 



SC 



111 



112 



113 



114 



115 



116 



117''' 



118 



119 



120 



121 



122' 



123 



124 



125 



126 



127 



128 



129 



130' 



131 



132 



133 



134 



Total 



20 



22 



14 



20 



20 



17 



21 



22 



23 



22 



' Haplotypes with an insertion at position 160. 

 - Haplotype with an insertion at position 210. 



The D value between the ML sample and the 

 remaining pooled Atlantic samples was -0.05%. In 

 contrast, the D value between the ML sample and 

 the Gulf samples ranged from -0.02% to 0.076%, 



The analysis of molecular variance ( AMOVA) for all 

 samples yielded a (^^.j- value of -0.001, indicating that 

 no significant heterogeneity was detected between 

 any two samples. For the geographic analysis, in 

 which samples were divided into shifting regional sub- 

 sets, the variance components among samples within 

 groups and among individuals within samples were 

 not significant for any grouping. Significant values of 

 (|)f.j.were observed in five of the 10 groupings (Table 2). 

 In four of the five significant groupings, the division 

 occurred in peninsular south Florida. Overall, results 

 of the AMOVA suggest that a genetic transition occurs 

 in red drum along the Florida coast between Sara- 

 sota Bay in the Gulf and Mosquito Lagoon in the 

 Atlantic. However, the highest (/)(^.y, value and between- 

 group variance component were observed when sam- 

 ples were grouped according to their actual Atlantic 

 and Gulf locations (Table 2, grouping 1). The signifi- 



cant 0f,j. value for the AP sample versus all other sam- 

 ples suggests that an additional genetic discontinuity 

 occurs in Gulf waters off northwest Florida. 



In the Mantel test between interpopulational D 

 and geographic distance, no association was observed 

 among the Atlantic samples (P=0.20) or among 

 the Gulf samples (P=0.053). However, a significant 

 association was observed for all red drum samples 

 (P<0.01 ), reflecting the genetic transition that occurs 

 in south Florida. 



Discussion 



Genetic population structure 



Because there are few absolute barriers to gene flow 

 in the ocean, it is generally expected that marine spe- 

 cies with continuous distributions, large populations, 

 and high levels of larval and adult dispersal should 

 have very little intraspecific population structure over 

 large geographic areas (Avise, 1987; Palumbi, 1992). 



