196 The Life Story of the Fish 



which he comes in contact? And how does he pick out his 

 own small tributary as he ascends a large river? Does he 

 remember landmarks so clearly that he can retrace his route 

 years later? Or is it true, as some scientists suggest, that he 

 has a chemical sense which enables him to pick out the 

 chemical composition of his own water from all the other 

 currents which he meets, and with some of which it mingles? 

 Aside from the fact that his own home water probably differs 

 from itself in chemical composition at various times of the 

 year more than it differs from other near-by streams at the 

 same time of year, it is harder for me to believe that the 

 salmon's chemical memory can retain for years his home 

 stream's composition than to believe that his visual memory 

 can retain the landmarks. One needs, like Alice in Wonder- 

 land's queen, to be able to believe at least three impossible 

 things before breakfast. We know that birds, after a several- 

 thousand-mile round trip to their wintering grounds, return 

 in the spring to the same nest that they occupied the pre- 

 vious year. We find ourselves capable of believing that their 

 memory enables them to do this, although we could not do it 

 ourselves. We know that a salmon can travel a thousand 

 miles from his own stream and still find his way back to the 

 small tributary of that stream in which he was born. If we are 

 not capable of believing that he does this by means of a topo- 

 graphic sense stronger than any we ourselves know, then 

 we must believe that it is by means of some sense of which 

 we do not even conceive. 



Baffled in their efforts to solve the "parent stream" — pref- 

 erably the "home stream" — problem, zoologists have turned 

 to the question of what makes the salmon leave salt water 

 and come into fresh to spawn. It is easy to say that it is 

 instinct, just as it is instinct which makes a bird build a nest 

 before it lays its eggs. Scientists are no longer satisfied with 

 that answer. They want to know how instinct works. Scien- 



