302 THE BIOLOGY OF THE SEASONS 



may be relied upon to show how often a salmon returns to 

 fresh water, then Malloch's conclusion seems secure that 

 many skip the grilse stage altogether, and do not return 

 till they are salmon. If the assumption be granted, then 

 many of the smolts marked in 1905 returned as small 

 Spring salmon in 1907, and many came up the Tay in the 

 following year, from 20 to 40 Ib. in weight, without having 

 been in fresh water since they were marked. In any case, 

 it may be safely said that the smolts are moved by a strong 

 impulse to go to the sea, that they feed voraciously and 

 grow quickly there, and that they return sooner or later 

 to spawn. Most, if not all, of the energy which the salmon 

 show in ascending the swift streams, and in leaping up 

 the falls, is accumulated during the variable marine period. 

 For after youth there is practically no feeding in fresh 

 water. 



Again, as in so many other cases, we have a sharp anti- 

 thesis between a feeding and growing period on the one 

 hand, and a fasting and reproductive period on the other 

 hand. It is the antithesis of caterpillar and butterfly over 

 again ; but how different the guise ! Few salmon seem 

 to spawn more than once, and some die of spawning. 

 After the exhausting process, they are known as " kelts " 

 thin and lanky, and out of condition. They return to 

 the sea, regain their tone, and eat ravenously. It is certain 

 that some live to a good age, for giants of 55 and 60 Ib. 

 have been caught with the rod, and even larger ones in 

 nets. 



There are serious discrepancies, both of observation 

 and interpretation, in the current accounts of the life- 

 history of this very common creature. As we hinted at 

 the outset, the probable reason for this may be found in the 

 variability of the organism on the one hand, and in the 

 diversity of its environmental stimuli on the other. 



