average separation of 81 percent. A similar 

 comparison of the James with the Rappahannock, 

 which has a slightly lower mean value, sepa- 

 rates 67 percent of the James specimens from 

 87 percent of those from the Rappahannock, for 

 an average of 77 percent. When the James 

 sample is compared with that of the Nanticoke, 

 which has the lowest mean value except for the 

 James, 67 percent of those from the James are 

 separated from 68 percent of those from the 

 Nanticoke, or an average of 67.5 percent. 



For the 1954 year class a line drawn be- 

 tween character index numbers 13 and 14 (table 

 22) permits a separation of 66 percent of the 

 James sample from 82 percent of the York sample, 

 an average of 74 percent. Where the numbers 

 are adequate for this type of treatment for the 

 Upper Bay samples, the lowest mean value is 

 found in the Wicomico. 66 percent of the James 

 specimens may be separated from 64 percent 

 of the Wicomico sample, an average of 65 per- 

 cent. 



Each of the fin ray characters studied 

 furnish a statistical basis for the separation of 

 the James River subpopulation from that in the 

 adjacent York and Rappahannock Rivers . The 

 latter subpopulation is also separable, on the 

 basis of second dorsal soft rays, from the sub- 

 population of the Upper Bay tributaries except 

 that of the Pocomoke. A meristic index based 

 on first dorsal fin spines and on dorsal and 

 anal soft rays supports this conclusion. 



What is the cause of these differences; 

 are they wholly or in part environmentally con- 

 trolled? Evidence is provided that first dorsal 

 spines are relatively constant throughout a wide 

 range in the Hudson River and through a consid- 

 erable geographic expanse of the James River. 

 On this basis it is assumed that the character 

 is perhaps less affected by environmental fluctua- 

 tions than are some of the other fin ray counts . 



Soft dorsal and anal rays show a trend to- 

 ward an increase in downstream samples in the 

 James River but the differences are usually not 

 significant. A significant increase is noted in 

 downstream samples in the Hudson River which 

 perhaps are a reflection of average temperature' 

 and/or salinity differences. However, these 

 character differences are relatively small 



compared with those which are considered to 

 denote races. A more precise solution of the 

 problem of the time of fixation of, and the effect 

 of temperature and other factors on meristic 

 characters in our important fishes awaits needed 

 experiments of the type described by Taning 

 (1953). 



How do the more direct results obtained 

 by tagging contribute to the concept of three 

 subpopulations in the tributaries of Chesapeake 

 Bay? Pearson (1938: 842) reported on the re- 

 turns of striped bass from 26 to 40 cm. long 

 tagged in July and August, 1931 off Annapolis, 

 Maryland. Over a two-year period 29.1 percent 

 were recovered. Of 89 recoveries only 9 were 

 taken south of the point where they were origin- 

 ally marked and released. Most were captured 

 in the Upper Bay from Magothy River and Love 

 Point north to the Susquehanna and Elk Rivers. 

 The point of greatest concentration was in the 

 vicinity of Rock Hall near the entrance to the 

 Chester River. The most distant points of re- 

 covery were single specimens from the Wicomico 

 and Potomac Rivers . Much additional data on 

 migration within the Bay from the studies of 

 Vladykov and Wallace (1938 and 1952) showed 

 that striped bass tagged in the middle Bay area 

 remained in the area where they were originally 

 tagged during late summer and fall . However, 

 it was a heterogeneous population and by the end 

 of October some started to move southward, es- 

 pecially along the western shore of the Bay; a few 

 reached the Rappahannock and James Rivers and 

 some left the Bay. They concluded that the Chop- 

 tank and Susquehanna Rivers were the main 

 spawning areas for the bass originally tagged at 

 Galesville, Flag Pond and Tilghman in the middle 

 region of the Bay . Data for fin ray counts also 

 support the view of an Upper Bay subpopulation . 

 Of considerable numbers tagged in October in the 

 Potomac River only a small percentage was re- 

 captured elsewhere which may indicate a local 

 population. The characters used in this study do 

 not confirm this view . 



Those tagged in the James River were al- 

 most all recaptured there and Vladykov and 

 Wallace (1938 and 1952) assumed that this was a 

 separate population. This finding is supported 

 by the present study. 



Previous tagging returns indicates little 



93 



