96 



DEPARTMENT OF THE NAVAL SERVICE 



another " kind " of x^orwegian herring, the so-called large herring, can hardly be 

 explained save by the supposition that a certain year-class of herring have throughout 

 the whole of the time from 1908 been more numerously represented in the stock than 

 others, and that as time went on, and the fish grew older, one ring was formed on the 

 scales for each successive year. 



These observations exhibit great similarity to the results in the following experi- 

 ment, which may be easily carried out with fresh-water fish : A number of fish are 

 caught, all those with a certain number of rings on the scales sorted out and liberated 

 in a pond where no fish of the same species are' found. A year later they are recap- 

 tured, scale samples taken and examined, to ascertain whether the scales now exhibit 

 one ring more than at time of liberation. They are then again set free, to be taken 

 up once more a year after for renewed examination. The investigations actually made 



5 4 5 6 7 



Age groups 



10 II 12 13 14 15 16 17 18 



Fig. 21. 



lack the absolute certainly as proof which such a test experiment would have; on the 

 other hand, the length of the period here embraced, and the absolute uniformity of 

 the results from all samples, are points of no little weight. Throughout the whole ^f 

 the time from April, 1908, to February, 1916, not a single sample was found to furnish 

 any exception to the general regularity. Not until February, 1916, was a sample 

 brought in which lacked the 12-ringed fish present in such great numbers in 

 the remaining twenty-four samples from the winter in question. This single instance, 

 however, does not impair the proof-value of the material as a whole, for the regularity 

 observed can naturally not be expected to continue indefinitely. On the contrary, it 

 would be natural to expect that the character of the stock in this respect will in a 

 short time exhibit noticeable change. 



The samples of grown Norwegian herring, from the year 1910, inclusive, contain 

 a considerable quantity of specimens exhibiting a remarkable arrangement of the 

 winter rings. As a general rule, the distance between the rincs decreases from the 



