•88 



It would appear also that the shoal increased in extent after 

 the restrictions of 1915. It extended in 1916 as far south as the 

 mouth of the Tees, evidence of mixing of growth groups being 

 obtained from this region. The extent northwards is difficult 

 to estimate from the first year growth, but from a consideration 

 of age composition the Fame Islands in 1916 seem to be its northern 

 limit. In 1917, the shoal, viewed from the first year gcoAvth, 

 extended north of St. Abbs, and gave evidence of mixing about 

 the region of the Coquet. If the shoal had moved north to this 

 extent in 1917 and had its southern limit in the vicinity of the 

 Coquet, the same cannot be said with any certainty to have 

 obtained in 1918. Sample 1 of 1918 may be taken a;^ giving slight 

 indications of mixing, and the fish with four winter rings of samples 

 4 and 10 had a first year growth which reminds one of the growth 

 of the samples from the southern portion of the region in 1916. 



The distribution along the coast according to the growth 

 attained at the formation of the first winter ring, as observed in 

 the samples of 1916, is of interest, but it cannot at present be 

 c<»nnected with any other phenomenon. 



Return of Herrings to the Shoal. — By considering the 

 first 3^ear growth of the different age groups in their successive 

 years, e.g., by taking the herrings of two winteL* rings in 1913, 

 and so on to the fish with five winter rings in 1916, an indication 

 will be given of the extent to which the herrings of the Northum- 

 berland coast return to the same ground with increasing age. 

 This has been done for four j-ear groups, the fish with two winter 

 rings in 1913, 1914, 1915 and 1916^ 



All the samples for which the first year growth has been 

 determined are not included in their respective years. The aim 

 has been to fairly represent the shoals, and where changes have 

 been observed, as sample 10 of 1914 and sample Dl of 1916, which 

 contained 25 per cent, of fish ready for spawning, and where a 

 year class is very poorly represented these samples have been 

 omitted. Whilst the omitting of a sample has been considered 

 the fairest way of considering the matter with our present know- 

 ledge, yet in seme cases little difference is shown in the final 

 percentages determined. 



The herrings with two winter rings in 1913 had a much larger 

 first year growth than those of three, four and five winter rings 



