IS ON AMPHIBIAN MIGRATION. 



the 1910 work. It is probable from this that the number in any 

 one pond remains fairly constant. It Avill be noticed that of the 

 native animals set loose in Experiment 11 only one got back. 

 These were set loose, however, as will be seen in text-fig. 5, in the 

 worst possible position 1 could find from which the Newts — even 

 natives^ — might be expected to return. The slope led away down 

 the field from the pond and the field v/^as separated from the 

 trough by a drop of several feet and a hard road three yards across. 



A decisive result brought out by these homing experiments is 

 that Newts certainly have no very definite faculty or instinct for 

 detecting water even from a short distance (see Table III.) ; and 

 in this connection attention may be drawn to the length of time 

 usually taken by the native Newts in finding water from short 

 distances. They perhaps rely on a combination of small clues, 

 for, in addition to the downhill tendency and a small homing 

 faculty, a supposed sensitiveness to atmospheric moisture at a 

 distance and to slight variations in temperature, I can suggest as 

 auxiliary factors (but can adduce no evidence for or against) : — a 

 sense of smell for detecting stagnant water and a sense of hearing 

 for discovering other Amphibia already in the water, or, in those 

 Amphibia which live in streams, for discovering running water. 



Either all or a few of these may provide the animals with 

 clues, small though some of them would be. In this paper a com- 

 bination of the downhill tendency and a homing faculty is 

 considered as the chief factor. 



I was very anxious to make more experiments with natives, 

 but the stock became exhausted and this brought my work to a 

 prematui'e close. 



Conclusion. 



This paper does not pretend to present a conclusive case on 

 the subject of amphibian migration. But the results of the 

 experiments are suggestive, and the opinions I have formed 

 from them are: — 



1. That Newts are helped to find water in which to breed 

 by their marked tendency to walk downhill. Those animals 

 hibernating on the slopes around their pond would simply have 

 to walk downhill. A good many young Newts I have found 

 hibernating in this way, close to the water of their native pond. 

 It may be a wise precaution, because young Newts of the year 

 would not have any knowledge of locality. 



2. That there is a small homing faculty in Newts, which 

 probably consists in a topographical knowledge of the locality 

 where the breeding pond is situated. 



3. That they are able to " home " from only short distances 

 and that, therefore, both their topographical knowledge and theii- 

 terrestrial wanderings in autumn and winter must be rather 

 circumscribed. 



4. That, speaking generally, the same Newts return to the same 

 pond each year. 



5. That those Newts which wander away from the precincts of 



