The number of dorsal spines in the numer- 

 ous samples of young taken in the Hudson River 

 in 1954 also gives a picture of relative constancy. 

 The sample from each locality on the Hudson 

 River (table 6) is composite and represents 

 varying numbers and sizes of subsamples taken 

 mostly from mid-July to October. The maximum 

 distance between the upstream sample taken at 

 Coxsackie and the downstream sample from 

 Palisades is approximately 100 miles. No trends 

 in mean values are observed. A test indicates 

 that the samples are not significantly different 

 at the 5 percent probability level and all samples 

 could have been drawn from the same population . 

 The lower Hudson River from the Palisades to 

 Peekskill is much wider and more saline than 

 the reaches above Beacon. When the same data 

 are compared as up- vs. downstream samples 

 (table 7), the means are identical and a test of 

 the samples indicates that there is no significant 

 difference. The means of up- and downstream 

 samples of young taken in 1953 (table 8) are 

 close and the difference is not significant. 



is significant at the 1 percent level and indicates 

 that the two very probably were not drawn from 

 the same population . 



The same indication of heterogeneity 

 was found for 1954 samples from Chesapeake 

 Bay tributaries (table 10) and again the James 

 sample is obviously different. When tested 

 against all the rest (Choptank to York in table 10) 

 X 2 = 112.78, which is significant at the 1 percent 

 level and indicates that the two samples very 

 probably were not drawn from the same popula- 

 tion . 



In order to compare the dorsal spine 

 values of East Shore tributaries, West Shore 

 tributaries (less the James River system), and 

 the James River, the samples were regrouped 

 in table 11 . A test of these composite samples 

 indicated a significant difference at the 1 percent 

 level. The East and West Shore means are close 

 and are obviously different from the James 

 sample . 



Thus it has been shown that no statistical- 

 ly significant differences occur in up- vs. 

 downstream samples of young of the 1954 and 

 1955 year classes from the James River nor for 

 the 1953 and 1954 year classes from the Hudson 

 River. 



The difference in mean values for the 

 first dorsal spines between the James River 

 sample and those taken elsewhere along the 

 coast, especially with those of other Chesapeake 

 Bay tributaries, is noteworthy. The data (tables 

 9-10) for samples of young taken in 1955 and 

 1954 from tributaries of Chesapeake Bay dis- 

 close high values for both year classes from the 

 James River. There is consistency within the 

 system . The Chickahominy River considered 

 alone has a high mean value of 9.30 (27 speci- 

 mens) in 1955 and 9.49 (61 specimens) in 1954. 



A test of the 1955 samples (table 9) gives 

 a significant difference at the 1 percent level 

 and indicates they very probably were not all 

 drawn from the same population. Because an 

 inspection of the mean values indicates that the 

 James sample is outstandingly different, it was 

 tested against that of the composite sample from 

 the Elk to the York (table 9); X 2 = 50.66 which 



The data on dorsal spines for Chesapeake 

 Bay tributaries are arranged in geographic blocks 

 from north to south (table 12). A test gives a 

 value which is significant at the one percent level 

 and which indicates heterogeneity. It is obvious 

 from an examination of the mean values that the 

 James River sample differs most markedly from 

 the other three composite samples. 



On the basis of the dorsal spine character 

 it is concluded that in the Chesapeake Bay area 

 there are two subpopulations. The James River 

 subpopulation is differentiated from all others 

 studied within and outside Chesapeake Bay. 



Samples from the Hudson River have the 

 next highest count; when the data for samples of 

 young of the 1954 year class are compared with 

 those from the James River (table 13) a test 

 shows the difference to be statistically significant 

 and the population to be heterogeneous . 



Soft Rays in the Second Dorsal Fin --The 

 modal number in samples studied throughout the 

 range of the striped bass is 12 rays except for 

 those from the Hudson River where it is 11 ex- 

 cept in very few samples . The range is from 9 

 to 14 but counts of other than 12 or 11 are un- 



89 



