148 



DEPARTMENT OF THE NATAL SERVICE 



table is thus totally different from that given by table IG (p. 136), showing the inter- 

 nal differences exhibited by the Newfoundland samples when compared one with 

 another. 



Comparison with the sample from Northumberland strait. Table 34 corresponds 

 to table 32 and shows the total averages of the Newfoundland samples compared with 

 total averages for the samples from Northumberland strait. 



Table 34. — Growth of Newfoundland herring compared with that of herring from 

 Northumberland Strait by help of total averages for year-classes 1904 and 1903 

 respectively. 



The table here shows that the Newfoundland samples, when compared with those 

 from Northumberland strait, differ from these in very much the same way as they 

 were seen to do from the Magdalen islands samples, the difference for t., however, is 

 greater, and must be considered as significant. In view of the great differences appa- 

 rent, I have not considered it necessary to show comparison between the separate sam- 

 ples; such a table would, roughly speaking, be merely a repetition of table 33. 



7, Comparison with exceptional sample from St. George's Bay. — Table 35 shows 

 the total averages of the nine similar samples from Newfoundland compared with the 

 corresponding averages for the exceptional sample, the 1904 and 1903 classes being 

 taken together in the latter. 



Table 35. — Growth of Newfoundland herring compared with that of the herrings of 

 the divergent sample from St. George's Bay by help of total averages for year- 

 class 1904, in Newfoundland samples, and those for year-classes 1904-1903 in 

 the divergent sample. 



The table shows the great differences between the exceptional sample and the 

 remaining ones from Newfoundland; these differences are indeed so marked as to be 

 distinctly apparent even when making comparison with the few (seventeen) specimens 

 of the 1904 year-class contained in the former sample as shown in table 36. 



