12 



period of formation of each winter ring. It has been found, 

 however, and not without some degree of misgiving, that when the 

 growth of the herring is thus expressed in terms of the growth of 

 the scale the calculated sizes at each winter ring form a descending 

 series v,dth age. It is to a consideration of this apparent con- 

 traction in size that this paper is mainly directed. 



The anomaly will be best understood by giving an example, 

 Table I.:— 



TABLE I. 



-Summary of Calculated Sizes in Centimetres of Samples A — L, 

 Northumberland School, 1913. 



It will be seen from the table that the mean calculated size 

 appears to suffer contraction in each successive year group, and 

 it will be noted that the shrinking in size particularly characterises 

 the results during the period of formation of the first three winter 

 rings. 



Other examples will be found in the paper by Lea, and the 

 tables he gives also show that the apparent contraction 

 takes place even when the attempt is made to follow the succes- 

 sive year groups in successive years. In other words, the apparent 

 change in the rate of growth is not confined to one sample nor to 

 the samples of one season. 



With a view to obtaining the necessary information for 

 contrasting the growth of the scale with the growth of the herring 

 measurements have been made of the scales of two samples of 

 herring. Sample 7 was obtained on July 20th S.E. of the Tyne, 

 and sample 8 on July 27th, 1914, N.E. of the Tyne. In Table II. 

 the sizes of the scales are given, and in Table III. the sizes at each 

 winter ring as calculated by the method now under review. 



