of the, Sidmon in Fresh Water. 171 



and occasionally do, take and swallow worms and other wriggling objects 

 is well known. But the swallowing of a few worms can do but little to 

 make good the enormous changes going on in the fish, even if, when 

 swallowed, they are digested and used. 



The evidence we have adduced may be summarised as follows : 



1st. There is no reason why salmon should feed during their stay in 

 fresh water. When they leave the sea they have in their bodies a 

 supply of nourishment not only sufficient to yield the material for the 

 growth of ovaries and testes, but to afford an enormous supply of energy 

 for the muscular work of ascending the stream (pp. 33, 93, and 120). 



2nd. During the stay of the fish in fresh water the material accumu- 

 lated in the muscles steadily diminishes, and there is absolutely no 

 indication that its loss is made good by fresh material taken as food 

 (pp. 83, 93, and 120). 



3rd. The marked and peculiar degenerative changes which the lining 

 membrane of the stomach and intestine undergoes during the stay of the 

 fish in fresh water shows that during this period the organs of diges- 

 tion are functionless (p. 13). 



The absorption of food stuffs is not a mere mechanical process, but is 

 chiefly dependent on the activity of the cells lining the alimentary 

 canal, and in the river these essential cells degenerate and are shed. 



It is a point of no little interest that before the fish again reaches the 

 sea, after spawning, the lining membrane of the alimentary canal 

 undergoes complete regeneration (p. 17 and 20), while the distended con- 

 dition of the gall bladder seems to indicate that the bile-forming func- 

 tion of the liver is again becoming active. 



4th. The very low digestive power of extracts of the mucous mem- 

 brane of the stomach and intestine, not only in fish from the upper- 

 reaches in which the degenerative changes abovereferred to have occurred, 

 but in fish coming to the mouth of the river and with the lining mem- 

 brane still intact, seems to indicate that the salmon has practically 

 ceased to feed before it makes for the river mouth (p. 23). It is to be 

 regretted that, although every effort has been made, no specimen of a 

 salmon stomach containing food has been procured. It is highly 

 desirable that the digestive activity of such stomachs should be compared 

 with the activity of those examined in this series of observations. 



5th. The changes in the bacteria of the alimentary canal also throw 

 light upon the question (p. 36). Generally speaking both in the 

 estuary and in the upper reaches, the number of organisms varies 

 directly with the temperature of the water. This is just what might 

 be expected, since the number of organisms in the water largely depends 

 upon its temperature. 



But setting this aside, it is found that while in the gullet there is no 

 great difference between the number of organisms in fish in the 

 estuaries and in fish in the upper reaches, there is a markedly greater 

 number of organisms in the stomach and intestine of fish in the upper 

 reaches. This is especially the case with the putrefactive organisms 

 which are the most readily destroyed by free acids. The greater abundance 

 of these in the upper water fish is strongly indicative of the absence of 

 the free acid which is formed in the stomach of fish while digesting food, 

 and which if present would destroy them. 



Miescher concluded that organisms are less numerous in the 

 fish in the upper waters, but his conclusion is not supported by any 

 evidence. 



6th. Our observations confirm these of Miescher as regards the 

 absence of food from the stomach. In not one of the 104 fish sent to the 

 Laboratory during 1896 and the spring of 1897, was any trace of food 



