170 Investigations on the Life-History 



The number of male fish examined was too small to allow of general 

 conclusions being drawn. 



In the kelts examined the amounts of solids were per standard fish 



Muscle, - 946-00 



Ovaries, - - 9-28 



955-28 



It would thus seem to be the state of nutrition which is the factor 

 determining migration towards the river ; that when the salmon has 

 accumulated the necessary supply of material It tends to return to its 

 original habitat. 



From May to August probably from November to August the fish 

 leaving the sea have the amount of material stored in their muscles 

 about the same. During these months the ovaries are yet small, and do 

 not act as a reservoir for stored material. In October and November 

 the estuary fish have a smaller amount of stored material in their 

 muscles. Why have these fish not left the sea sooner ? Is it that, 

 either because they have left the rivers later, or because the supply of 

 food has been less readily obtained, the period of rapid growth of the 

 genitalia has supervened before the full accumulation of material in the 

 muscle has been accomplished ? This rapid growth of the genitalia 

 would withdraw material from the muscle and prevent its accumulating 

 there, and thus, when the necessary amount of stored material was 

 accumulated, it would be distributed between the muscles and genitalia. 



The late-coming fish, although the supply of solids in the muscles is 

 smaller, have the ovaries so large that the total store of nutrient 

 material in the fish is just about the same as in those entering the 

 estuaries in the earlier months. 



A return to fresh water is essential for the completion of reproduc- 

 tion, for it has been shown that salt water prevents the development of 

 the ova. In its natural condition the fish is impelled to migrate sea- 

 ward in search of more abundant food, but descent to the sea is not 

 necessary for the development of the genitalia. This is proved by the 

 experiments carried on at Howietoun, which show that fish, when 

 properly fed, may develop their genitalia without leaving fresh water. 

 Again, the salmon of the great American lakes spawn in the streams, 

 and yearly descend to the main lakes, as the salmon of this country 

 descend to the sea, there to feed and lay in the necessary store of 

 material. 



The course of migration and the question of to-and-fro migration 

 have been discussed on pp. 75 to 78, and it has been shown that the early- 

 coming fish press up to the upper waters of the rivers, and that on to 

 August fish continue to stream into the upper reaches ; but that the 

 fish leaving the sea in October and November do not at once ascend to 

 the upper parts of the river. It has been further shown that there 

 is strong evidence against there being a to-and-fro migration from river 

 to sea and sea to river throughout the season. 



B. Do SALMON FEED IN FRESH WATER? 



The question of whether salmon feed while in fresh water has been 

 frequently discussed. Much depends on what is meant by the word 

 "feeding." By feeding, we here mean not the mere swallowing of material; 

 but the digestion, absorption, and utilisation of that material by the body. 

 That salmon take the fly, minnow, or other shining object in the 

 rnouth is no argument as to their feeding in this sense. That they may, 



