40 THE SCOTTISH NATURALIST 



the Tay, revealed, in November and December, fish heavy 

 with spawn, Kelts, large spring fish fresh out of the sea a 

 year before they could spawn, and Grilse ready to spawn. 

 From the number of recaptures we had obtained — about 

 300 at that time — it had also become apparent that there 

 was what has been called a divided migration ; that some 

 fish after spawning went to the sea and returned again the 

 next season, while others went to the sea and remained there 

 for two or three seasons before returning. In other words, 

 a large number of Salmon disregarded the spawning season 

 and remained feeding in the sea ; so that even at the spawn- 

 ing time there are probably more Salmon in the sea than in 

 fresh water. 



In comparing weight and length results, it soon became 

 apparent that one had to classify these short- and long-period 

 fish, and avoid mixing the one class with the other. An 

 average increase of 6 lb. in the weight of short-period fish 

 had to be put against an average of 10 lb. for the long-period 

 fish. It also began to emerge from the extended observa- 

 tions that short-habit fish were much more common in certain 

 districts than in others where the long habit seemed the rule. 

 In these early fresh-water operations over 12,000 fish had 

 been marked, and although the recaptures were not very 

 numerous— which later became explained when we under- 

 stood that the great majority of our Salmon, especially on 

 the east coast, spawn only once or twice in their lives, and 

 therefore return but seldom to fresh waters — a considerable 

 amount of information apart from the homing habit had been 

 secured. I may here add that only in small west coast rivers 

 have fish been found which had spawned on three and four 

 consecutive years. In one isolated case we have found five 

 consecutive spawnings. Of iii marked and recaptured Tay 

 fish only one had been taken beyond the natural limits of 

 the Tay district. 



The examination of Salmon scales, which has told us so 

 much in recent years, only commenced in the Tay in 1904, 

 so that at the time of the Tay Smolt-marking we had not got 

 further than to realise that the scale reading corroborated the 

 marking results in every way. 



