NATURAL HISTORY OF BRITISH SALMONIDJE. 133 



near the tail of the pool. Crawling to the bank's edge, and 

 cautiously putting our heads over, we could see everything 

 that went on below. In the front rank there were five large 

 fish only, one apparently of about 30 lbs., the others running 

 perhaps from 20 lbs. to 25 lbs. In the second rank were fifteen 

 ranging from 15 lbs. to 20 lbs. ; in the third twenty-three from 

 10 lbs. to 15 lbs., and in the rear rank the Grilse, some thirty- 

 two in number. All the fish were motionless except for the 

 slight action of the tail fin necessary to keep them up to the 

 stream. We then sent the keeper round to stir them gently by 

 throwing a stone in behind the Grilse, which caused the whole 

 body to move up into deep water ; but after waiting about an 

 hour we saw them begin to return to their former quarters, 

 where by degrees they took up their original formation in per- 

 fectly " dressed " ranks, and without a fish missing. 



'When the water was barely high enough for fishing, I could 

 still command a sight of the bottom of the pool, although not 

 a perfectly clear one. As the fly passed over him, a salmon 

 before rising invariably turned on his side. . , .' 



To return to the subject of salmon migration. Allowing for 

 the exceptions previously pointed out, the order in which fish 

 ascend rivers is generally somewhat as follows : The strong, 

 early runners come first. These are succeeded by the grilse, 

 and by the small ' spring salmon ' which have probably never 

 ascended at all as grilse, but have remained in the sea since 

 the smolt state— a period of from eight to ten months, as 

 noticed in ' Proved Facts,' No. 10. The tails of these spring 

 salmon are not so forked as those of grilse. A few of them 

 generally appear with the early-running fish. 



As the season advances, the larger fish and those heavy 

 with spawn begin to work their way upwards from the mouths 

 of the rivers and estuaries towards the higher reaches ; and 

 such fish continue ascending from the sea until the close of 

 the autumn, or, if the river be an ' early ' one, of the summer. 

 Even as late as November and December, and the January 



