Transactions of the American Fisheries Society 123:321-334, 1994 © Copyright by the American Fisheries Society 1994 Mitochondrial DNA Diversity of Roanoke River Striped Bass EDMUND J. STELLWAG AND E. STURGIS PAYNE Department of Biology. East Carolina University Greenville, North Carolina 27858-4353. USA ROGER A. RULIFSON Institute for Coastal and Marine Resources. East Carolina University Abstract. —The genetic heterogeneity of mitochondrial DNA (mtDNA) in Roanoke River striped bass Morone saxatilis was examined by restriction endonuclease analysis of mtDNA isolated from 32 specimens captured during the 1989 and 1990 spawning runs. Restriction fragment banding patterns based on restriction site polymorphisms revealed six distinct genotypes, designated I-VI, among the 32 sampled individuals. The predominant genotype, designated genotype I, was found in 59% of the individuals sampled, while genotypes denned by Ace l-Nde I (genotype II), Taq I (genotype III), Taq l-Hinf I (genotype IV), Dde I (genotype V), and Spe I (genotype VI) were found in 25%, 7%, 3%, 3%. and 3% of the individuals. Five distinct mtDNA fragment length variants, designated classes A-E, were identified from the same 32 individuals. The number and size of individual length variants agreed with those identified previously in striped bass populations from geographically distant sites. Analysis of the distribution of length variant classes A-E among genotypes I-VI (defined by restriction site polymorphisms) revealed a lack of concordance between individual length variants and specific genotypes. Geographically and genotypically independent distributions of the five length variants strongly suggest that mtDNA fragment length variants have converged in striped bass. Although mtDNA fragment length variants have been used to distinguish among regional striped bass stocks, our results suggest that caution should be exercised in phylogeographic assignment on the basis of mtDNA length variants alone. Using estimates based on restriction site polymorphisms only, we calculated a nucleotide sequence divergence value of p = 0.003, nearly an order of magnitude greater than reported previously for striped bass. Our results indicate that either the Roanoke River striped bass stock is genetically more diverse than other striped bass stocks, or that previous estimates of stock diversity are low. Striped bass Morone saxatilis is a commercially tion. Therefore, the vigor of major spawning stocks and recreationally important fish species distrib- is evaluated by measuring the yearly commercial uted along the east coast of North America from and recreational harvest (Deuel et al. 1989). Al- Canada to the Gulf of Mexico (Raney 1952). though adequate for monitoring wholesale popu- Striped bass native to river systems as far south lation fluctuations within restricted geographic as Cape Hatteras, North Carolina, are thought to regions, a harvest-based stock assessment method be anadromous, whereas populations south of is inadequate for ascertaining the effects of non- Cape Hatteras, including those along the Gulf native stocking or natural mixing of spawning coast, are thought to be riverine (Raney and Wool- populations due to anadromy. cott 1955). Three major spawning stocks occur It is possible that extensive stocking of major along the mid-Atlantic coast: the Hudson stock, spawning populations of striped bass with non- which spawns in the Hudson River, New York; native fish may have altered the native gene pool, the Chesapeake stock, which spawns in rivers and thus contributing to decline of stocks. Therefore, tributaries of the Chesapeake Bay; and the Roa- the only appropriate method by which to estimate noke stock, which spawns in the Roanoke River, regional spawning stock vigor requires a compre- North Carolina (Berggren and Liebermann 1978). hensive genetic analysis of each regional stock. Precipitous declines in the major spawning stocks Identification of genetic markers characteristic of have intensified efforts to characterize individual the strains representative of each regional stock stocks to permit evaluation of the effects of stock would provide a means to distinguish among stocks management and restoration efforts (Waldman et and would facilitate efforts to account for possible al. 1988). Despite repeated attempts to develop a intregression, the effects of stocking nonnative stock-specific identification system, the major specimens, and the harvest of spawning as well as spawning stocks continue to be identified primar- migratory stocks, ily on the basis of spawning site or capture loca- A range of techniques based on morphological, 321