142 



DEPARTMENT OF THE NATAL 8ERYICE 



Table 24. — Averages for year classes 1903 and 1907 compared with those for other 

 year classes in the five samples from Northumberland Strait. 



Year elates. 



1011 

 1910 

 1909 

 1908 

 19!)7 

 1900 

 1905 

 1904 

 1903 

 1902 



Number 



of 



individual.s. 



5 

 43 

 51 

 23 



110 

 25 

 35 

 52 



173 

 13 



hi 



0-6 



If we are to characterize the growth of the individuals in these samples, w^e can 

 only repeat what has been said in the case of those from the Magdalen islands; the 

 similarity existing between these two waters, in respect of age composition, is sup- 

 plemented by the resemblance noticeable in the growth of the fish. We shall later on 

 examine the question as to whether there is any essential difference between these 

 waters as regards growth. 



It may not be out of place here to mention the samples containing very young 

 fish. Some of these are from the waters about Prince Edward Island and the Gaspe 

 coast, others from the neighbourhood of Souris and cape George. Table 25 shows 

 the average value for these samples. 



Table 25. — Showing averages for growth dimensions t^ — t^ for young herring in the 

 waters around Prince Edward Island. Values with an asterisk are incomplete. 



Locality and date. 



Station 33, 48" 13' N.. 63= 58' \V., July 5-C, 1915 - 



Station 3.5, 46" 48' N., 64" 32 W., July 7, 1915 



Souris, July 14, 1915 -{ 



Station 53, 45" 46' N., 62" 23' W., July 30-31, 1915... ' 



It should be borne in mind that the last growth dimension in most of these sam- 

 ples is incomplete, and cannot therefore be compared with the same in the older fish. 

 The table shows that these young fish also are characterised; by a high value for t^, 

 exactly the reverse of what was the case with the young fish from Port au Port (taken, 

 it is important to note, with the same set of gear as the present samples from the 

 Gaspe coast). As regards the differences apparent between the different samples, and 

 between the different year-classes, it should be remembered that both dissociation of 

 year-classes and net-selection must doubtless have had some effect. Making allowance 

 for these considerations, the resemblance between the samples is marked. 



