64 /. H. Pozvcrs 



Functionally, too, these contrasting developments become evi- 

 dent, and even absurdly so, when the annuals are placed side by 

 side out of water. Unless glutted with food, the ordinary larva 

 can crawl considerably, but not so the cannibal. When I dipped 

 such animals out of their aquaria with a concave sieve to display 

 their contrasting developments, their ludicrously different achieve- 

 ments in locomotion never failed of effect. The ordinary larvae 

 could crawl about considerably in the sieve, and, if strong-limbed 

 or starved a little to reduce weight of the body, would sometimes 

 crawl quite out. But the most desperate efforts of the cannibal 

 could never raise the anterior end of its body well from the sup- 

 port, and it soon found itself pinned down, as it were, with its 

 head in the lowest part of the sieve, all eff'orts at locomotion 

 serving but to move the posterior portion of the body round and 

 round the stationary head. 



I have spoken of the fact that incipient cannibals may be in- 

 duced to return to ordinary feeding. It is interesting to note, 

 however, that this is, in many instances, extremely difficult, even 

 with larvae no more than 6 cm. long. I have taken such larvae 

 from clear water and found them obstinately feeding upon their 

 companions, even though the water of their aquaria was alive 

 with daphnids, which constitute the most natural food of the 

 species at all ages. In no other single instance have I found 

 amblystomas of any age refusing daphnids as food, even when 

 they were several years old and had never before seen an ento- 

 mostracan. During two seasons I placed a large number of very 

 young larvae in a pond, well stocked with ordinan- food organ- 

 isms, but in which the conditions were not favorable to canni- 

 balism. Yet even here a few of the larvae which showed but a 

 trace of the broad-headed type when liberated persisted in their 

 development, and were taken from time to time during the sum- 

 mer in the most deplorable state of cannibalistic emaciation, so 

 delicate withal that I could scarcely keep them alive for twenty- 

 four hours. 



Not less surprising than this persistence is the obvious psychic 

 or nervous change that takes place almost from the start in the 

 voung larvae which has become well launched on its career. Or- 



260 



