Almost all stripped bass (Morone saxatilis) along the mid-atlantic coast originate from two estuaries: the Hudson River and Chesapeake Bay. Both stocks winter in coastal waters, where they may be subject to bycatch fisheries (e.g.., the Cape Hatteras, North Carolina, dogfish fishery). Virtually nothing is known regarding the stock composition of wintertime aggregations of stripped bass. During the mid-1970's, it was estimated that about 90% of the stripped bass found along the northeast U.S. coast were from the Chesapeake Bay; by the late 1980's the Hudson River was contributing more than half of stripped bass found along the coast. In 1993, the historical high for the Maryland juvenile stripped bass survey was recorded; thus, relative contributions of the two stocks may have shifted back toward domination by the Chesapeake Bay stock. We analyzed genotypic frequency data derived from composite information on single-copy nuclear DNA and mitochondrial DNA major length variation of stripped bass collected from three latitudinally-widespread locations: during winter 1995-1996 off the New Jersey coast, and during winter 1997-1998 at the mouth of Delaware Bay and off Cape Hatteras, North Carolina.  We also determined genotypic frequencies for stripped bass from three reference spawning tributaries of the Chesapeake Bay not previously characterized, the Nanticoke, Patuxent, and Pocomoke rivers. Contingency analysis indicated an absence of statistical heterogeneity among the three wintertime coastal collections. Stock composition analysis of the aggregate wintertime collections (N=356) showed a strong Chesapeake Bay contribution (80%) and a lower Hudson River contribution (20%). Stock composition analysis of the contemporaneous Delaware Bay and North Carolina collections in aggregate, showed relative contributions of about 16% for the Hudson River stock and 84% for the Chesapeake Bay stock. Discrete stock composition analyses on the three collections suggested a significantly higher Hudson River contribution for the New Jersey collection, but the asynchronicity and statistical questions render The validity of such analyses as uncertain. We also examined the utility of additional mitochondrial DNA and nuclear DNA markers (microsatellites).