History and Habits 15 



while others have many more salmon than grilse. 

 In some streams the grilse come along with the 

 first run of salmon, and in others six to eight 

 weeks later. A river which has a large propor- 

 tion of grilse may not have so many salmon as 

 one in which the grilse are scarce. All this 

 would seem to indicate that fish in the grilse 

 stage come back to their native rivers in vary- 

 ing proportionate numbers, or that the grilse of 

 one river go to another at that period of exist- 

 ence. This latter theory would account for the 

 presence of large numbers of grilse in one river 

 whilst another near by has but few, as is the 

 case in the Nepisiguit and Restigouche. It is 

 known that salmon have forsaken their own river 

 and gone up another for one year. I have seen 

 an undoubted example of this, and there is no 

 reason why grilse should not do the same. How- 

 ever, this is a sporadic action in salmon, and the 

 disparity in numbers between them and grilse in 

 different rivers is habitual. In the Tweed, ac- 

 cording to Mr. Willis Bund, in his most valuable 

 and interesting book " Salmon Problems," " from 

 1808 to 1853 in no year were less than three 

 grilse taken for each salmon, from 1S53 to 1876 

 only two grilse were taken for each salmon." 



