assumption is made that the populations are the 

 same, although it is quite clear that future 

 studies of other characters may show differ- 

 ences . Studies involving pectoral ray numbers, 

 the number and characteristics of scales, num- 

 bers of gill rakers, and measurements of body 

 parts are under way on samples taken in Chesa- 

 peake Bay and elsewhere as part of the Atlantic 

 Coast Cooperative Striped Bass Program . 



This and earlier studies were made pos- 

 sible by the cooperation of several agencies 

 and many fishery biologists William H. Mass- 

 mann of the Virginia Fisheries Laboratory 

 furnished specimens from the York River sys- 

 tem that stimulated the study by Raney and de 

 Sylva (1953) which suggested that different 

 races exist in the Hudson River and Chesapeake 

 Bay. He has continued to help and has actively 

 participated in the field work in Virginia . The 

 use of the vessel "Virginia Lee" which was 

 made available by Director J . L . McHugh of the 

 Virginia Fisheries laboratory is gratefully 

 actcnowledged. C. E. Richards and Jesse Hobbs 

 of the Laboratory assisted with the collecting 

 during July, 1954 and 1955. In Maryland the co- 

 operation of Edgar H. Hollis and Harold A. 

 Davis, Jr. of the Department of Tidewater Fish- 

 eries made available the bulk of the specimens 

 studied from tributaries of the upper Bay. L. 

 Eugene Cronin, Romeo Mansueti, Richard E. 

 Tiller, and Earl T. Walker of the Maryland De- 

 partment of Research and Education, Sobmons, 

 have made special efforts to get material from 

 the Patuxent and Potomac rivers. Through the 

 interest of John R, Greeley and W. Mason 

 Lawrence of the New York State Conservation 

 Department, support was received for field 

 work on the Hudson River during 1954. Warren 

 F. Rathjen and Lewis C. Miller assisted with 

 field work and made available collections taken 

 during 1954. Cornell graduate students Earl E. 

 Deubler, Jr., C. Richard Robins and William S. 

 Woolcott assisted during several field expedi- 

 tions and all counts taken on specimens obtained 

 in 1953 and 1954 were made by the latter. 

 Charles F. Cole made the counts on most of 

 the 1955 material. Robert Minturn Lewis made 

 some of the statistical calculations but the chi - 

 square tests were run at the Cornell Computing 

 Center through the courtesy of its Director, 

 Richard Lesser. Advice on statistical methods 

 has been received from Douglas S. Robson and 



Robert G. D. Steel of Cornell University, James 

 R. Westman and Kenneth W. MeinKen of Rutgers 

 University, and J. L. McHugh of the Virginia 

 Fisheries Laboratory. Helpful suggestions for 

 the improvement of the manuscript were made 

 by Gerald B. Talbot of the U. S. Fish and Wild- 

 life Service, Donald P. de Sylva of the University 

 of Miami Marine Laboratory, William E . Fahy 

 and Earl E. Deubler, Jr. of the University of 

 North Carolina Institute of Fisheries Research, 

 and John R. Greeley of the New York State Con- 

 servation Department . 



The terms species, subspecies, and race 

 seem adequate to describe the major levels of 

 differentiation found in the striped bass . As here 

 used, race implies a lesser rank than subspecies. 

 The definitions of the terms population and sub- 

 population as proposed by Marr (first paper in 

 this group) are followed. 



There seems to be only one species of 

 striped bass . Its closest relative, the white 

 bass, Roccus chrysops (Rafinesque), inhabits 

 fresh water in the Mississippi River and Great 

 Lakes drainages. The population of striped 

 bass found in tributaries of the Gulf of Mexico 

 is 100 percent differentiated in number of lateral 

 line scales from the nearest Atlantic coastal popu- 

 lation of the St. Johns River, northeastern Florida. 

 The lateral line scales in 1 1 specimens available 

 from tributaries of the Gulf of Mexico range 

 from 62 to 68, mean 65 . 91 (S . D. 1.97, S.E.0.59), 

 and in the St. Johns River sample of 42 specimens, 

 they range from 52 to 58, mean 54.33 

 (S.D. 1.44, S.E. 0.22). Data for lateral -line 

 scale counts were given for samples from other 

 southeastern localities by Raney and Woolcott 

 (1955: 446). Perhaps there has been no exchange 

 between the St . Johns River and Gulf of Mexico 

 populations since the emergence of the Florida 

 peninsula because the tropical waters of south- 

 ern Florida serve as a barrier. 



Along the Atlantic Coast the apparent lack 

 of coastal migration in the region of southern 

 North Carolina to northeastern Florida has made 

 free gene exchange unlikely. Two races present 

 in the southeast which were designated by Raney 

 and Woolcott (1955: 449) as the St. Johns River 

 race and the Santee -Cooper race. In the small 

 sample from the Coastal Plain of South Carolina 

 characters are intermediate as would be expected 



86 



