62 /. H. Pozvers 



form diet, frequently upon a single food organism. (July little 

 by little do older individuals acquire more varied habits, their 

 stomachs, in rare instances, yielding various Entomostraca, in- 

 sect larvae from Chironomus up to the dragon fly, adult water 

 insects, and fair-sized younger larvae of their own species. 



Of all the points in the foregoing description, the most doubt- 

 ful one may appear to be the assumption that young amblystomas, 

 which normally feed upon organisms no larger than daphnids, 

 should so readily take up the practice of swallowing animals 

 nearly as large as themselves. The readiness with which the\' do 

 so does seem to vary very much with different individuals show- 

 ing no other differences than slight variations in- aggressiveness 

 of disposition. I tried many single specimens, isolated in small 

 aquaria in the laboratory, beginning with them at 6 to 8 cm. in 

 length and using either smaller larvae of their own species or 

 frog and toad tadpoles to induce them to change their diet. Some 

 refused entirely ; some enjoyed a cannibalistic meal or two, only 

 to relapse into starvation rather than attempt another ; some 

 made splendid attempts, but failing at first, forever lost their 

 courage. Only a few possessed the true heroic fire, dauntless, 

 no matter how great the difficulties or how frequent the failure. 

 I have seen such a larva- return continuously, aside from brief 

 periods of rest, to the task of swallowing a tadpole nearly as 

 large as itself, for twenty-four hours, working away for a few 

 minutes to an hour or more at each attempt. If the tadpole was 

 killed before being swallowed, no further notice was taken of it ; 

 but as soon as another was supplied it was attacked with as much 

 spirit as before. Such observations supply the full rationale of 

 the variation. One can fairly see the head-expansion within a 

 single day ; and no wonder, when the whole masticatory and 

 respiratory framework is stretched as if it were rubber and held 

 in this position for the utmost time that the animal can endure, 

 and then, after recuperation, the whole thing is repeated again 

 and again. With so plastic an animal as this larva, adaptive de- 

 velopment follows inevitably under such circumstances, and the 

 whole bulk of nutrition is drained off to the head and adjoining 

 parts. The meaning, too, of the starved or even emaciated con- 



258 



