150 



DEPARTMENT OF THE yATAL SERVICE 



Table 38. — Growth of Newfoundland herring compared with that of the herring in 

 the sample from North Sydney by help of total averages for year-class 1904 in 

 the case of Newfoundland herring, averages for the year-classes 1904 and 1903 

 taken together in the other case. 



9. Comparison with old fish from West Ardoise, West of Port Hood, and Locke- 

 port. — In order to procure material for comparison of the Newfoundland herring with 

 fish of old age from the Atlantic coast, it was necessary to collect older specimens 

 from the different samples. Taking all those of the 1904 and 1903 year-classes toge- 

 ther we have a total of thirty-four fish, one of which is from the sample taken at sta- 

 tion 42, west of Port Hood. Averages and standard errors for these thirty-four speci- 

 mens have been calculated, and comparison made accordingly, with the results shown 

 in table 39. 



Table 39.- — Growth of Ne^^'foundland herring (year-class 1904) compared with that of 

 herring from Nova Scotia (year-class 1904 and 1903 together). 



Locality. 



Newfoundland 

 Nova Scotia 

 Difference fD) 



d 



14 



11-24 



510 



15-5 



0-99 

 7-76 

 0-77 



3-1 



5-43 

 4-97 

 0-46 



2 1 



3 30 

 304 

 0-26 



2-74 

 205 

 0-69 



211 

 1 36 

 0-75 



12-5 



1-47 



in 



0-.36 



7-2 



108 

 0-87 

 0-21 



5-2 



There is here, as will be seen, a very marked difference in growth between the 

 Newfoundland fish and those from the southern portions of the Canadian Atlantic 

 coast. The differences here again are most distinctly apparent in the growth dimen- 

 sions /j and ^5 — t^ and resemble those found in the case of the sample from North 

 Sydney, t^ and t^ being greatest in the fish from the Atlantic coast, the remaining 

 dimensions less. 



10. Graphical illustratic'ns of the results obtained. — Figs. 40 and 41 show, in gra- 

 phical form, the results of comparison between the growth of the Newfoundland 

 herring and the growth of those in the remaining samples. In order to avoid a figure 

 complicated with too many curves, we have here shown, in fig. 40, the growth of the 

 Newfoundland herring compared with that of those in the samples from the gulf of 

 St. Lawrence (Magdalen islands and Northumberland strait) and in fig. 41 with that 

 of the Atlantic fish (North Sydney and southern samples). The upper portion of 

 each figure presents in graphical form the data already utilized, i.e., the average 

 calculated increments, while the lower portion showing ordinary growth curves, gives 

 the actufil length of the herring at the time of the formation of each winter ring on 

 the scales. 



