12 BERTRAM G. SMITH. 



aquatic. The branchial slit is guarded at its pharyngeal opening 

 by two valve-like flaps, in position and appearance like degener- 

 ate gills, which prevent the entrance of water from without. 

 Their respiratory service, owing to their small size, is probably 

 very insignificant. 



Feeding Habits. — Examination of stomach contents shows 

 that crayfishes form the principal food of Cryptobranchus ; fishes 

 are eaten only occasionally. Out of a dozen specimens exam- 

 ined in August, nine were found to contain crayfishes ; only three 

 contained fishes. Cryptobranchus sometimes takes a hook baited 

 with earthworms, if cast near it. In captivity it will eat almost 

 any small moving animal, or pieces of meat moved along a little 

 to one side of the head. A specimen kept in an aquarium in the 

 Zoological Laboratory of the University of Michigan is reported 

 to have eaten frogs on several occasions, and once a toad. One 

 of my newly-captured specimens seized a young Necturus about 

 eight inches long, but soon released it. The adult Crypto- 

 branchus will eat the eggs or larvae of its own kind. Specimens 

 kept in a creek aquarium in running water do not refuse food, 

 even immediately after their capture. 



Like many other amphibians, Cryptobranchus eats its own 

 shed epidermis. The epidermis usually comes off in ragged 

 pieces, giving the animal a tattered and uncouth appearance. 

 The epidermis from the feet comes off like the fingers of a glove. 

 The mouth is sometimes used to aid in the removal of the frag- 

 ments. The practice of eating the shed epidermis is probably 

 quite common amongst the Urodeles. I have observed a Triton 

 (Diemyctyhis) viridescens pulling off and eating the glove-like epi- 

 dermis from the foot of another specimen. Ritter ('97) states 

 that the shed epidermis of Triton {Diemyctylus) torosns forms an 

 important article of diet for the animal. 



Cryptobranchus apparently takes no notice of its prey until the 

 latter approaches within a very few inches of its mouth and a 

 little to one side, then seizes it by a remarkably quick sidewise 

 movement. Perhaps the striking distance corresponds to the 

 distance at which it can see clearly, or it may be the result of 

 experience in catching wary prey. The enormous size of the 

 mouth may compensate for the rather deficient eyesight (see 



