63 



But if the increment of growth for 1913 be examined some points 

 of interest appear. 



For the drift net herrings the 1913 increments are here given :— 



SAMPLES. 



For the herrings having two winter rings there is a steady 

 increase of growth until sample I is reached, and then another 

 gradual increase to sample L, and finally almost a jump from L to 

 M. Sample B may be neglected as it contained only four herrings 

 of this year class, and each had made little groAvth for the year. 



Herrings with three and four winter rings show a fairly steady 

 increase of growth until reaching samples J and K respectively, 

 when there are decreases, with subsequent gradual increases. 



These variations are taken as showing a change in the composi- 

 tion of the herring shoal off our coast. 



In the first week of September and the beginning of the second 

 week, immigrants with a smaller increase of growth for 1913, and 

 probably later migrants into the North Sea joined the shoal, the 

 younger herrings coming first. The new arrivals with three winter 

 rings were not present in great numbers in sample J, 6th September, 

 but in sample K, 8th September, they, together with those immi- 

 grants having four winter rings, had become the chief fish of these 

 year classes. 



The herrings with five winter rings were present in such small 

 numbers that a similar immigration cannot be shown for them. 



At the end of the second week of September, sample M, there 

 was to the E.S.E. of the Tyne another change in the composition 

 of the shoal, which may have accounted for the sharp change in 

 the 1913 growth increment of the herrings with two winter rings 

 and the high percentage, ca. 43 per cent., of this year class of 

 herrings. 



For the Dogger Bank and Yorkshire coast trawled herrings 

 the 1913 growth increments for those with two, three and four 

 winter rings show too much variation to allow of a distinction 

 being drawn between them. But the herrings of three and four 



