CAXADI.W FISHERIES EXPEDITIOS, 191 ',-15 



149 



Table 36. — Growth of Newfoundland herring compared with that of the herrings in 

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

 class 1904. 



Sample. 



Newfoundland (nine samples) 



St. George's Bay 



Difference (D) .' 



D 



d 



1 08 

 84 

 24 



3 4 



There can thus be no doubt that the exceptional sample differs strougly both with 

 regard to age composition and growth, from the remaining Newfoundland samples. 

 The difference in growth is again of a similar character to that noted between the 

 Newfoundland samples and those from the Magdalen islands, the greatest differences 

 are found in the growth dimensions ^,, t., t^ and t.. Here, likewise, t^ is greater in 

 the exceptional sample, while the other dimensions are smaller. 



Table 37 shows the exceptional sample compared with each of the remaining New- 

 foundland samjiles separately; the order of magnitude of the fraction -, being given 

 for the differences between the five first growth dimensions. 



D 



Table 37. — Distribution of values of fraction ~ arising by comparison between the 



a 



divergent sample from St. George's Bay and each of the nine other samples 



from Newfoundland. 



It will be seen that the exceptional sample differs strongly from each one of the 

 remainder. 



8. Comparison with the sample from North Sydney. — In view of the great resem- 

 blance which exists between the sample from North Sydney and the exceptional sam- 

 ple from St. George's bay, it might appear sui>ertluous to make any further comparison 

 between the former and the Newfoundland samples. For the sake of completeness, 

 however, this has been done in table 3S, and we find that the Newfoundland samples 

 differ from this sample exactly as they were seen to do from the exceptional sample 

 from St. George's bay. 



