1912] Ditmars: Feeding Habits of Serpents 225 



long and sharp spines are a common prey but are swallowed without 

 difficulty as the snake's venom renders the muscles quite inert. The 

 quarry is swallowed head first, the bristling spines folding against the 

 body while deglutition goes on. 



The Dipsadormorphinae: Despite the elaborate array of species 

 of this subfamily, it is difficult to obtain a representative series of 

 living examples for study. Many of the species are small and secre- 

 tive; others are arboreal and difficult to capture, and amid these 

 conditions they escape the attention of the collector of living reptiles. 

 A number of living examples of the Dipsas type, such as Tarbophis, 

 Trimorphodon, Lycognathus, Dipsadomorphus, Himantodes and Sibon 

 have been under the writer's observation. All the specimens studied 

 were oviparous and showed a preference for small, soft-scaled lizards 

 and batrachians; usually feeding at night. With Tarbophis and 

 Trimorphodon, lizards are preferred. The specimens of Lycogna- 

 thus, Dipsadomorphus and Himantodes preferred small frogs. Exam- 

 ples of Tarbophis and Lycognathus were induced to take very young 

 mice and birds, but these were refused if even thinly clad with pelage 

 or feathers. With the strictly arboreal genera, such as Dryophis 

 and Oxybellis there is an unvarying preference for lizards. No speci- 

 men of these latter genera was noted to feed upon batrachians, and 

 but occasional examples exhibited any interest in freshly caught 

 birds or small rodents from the nest. 



Food of the Proteroglypha 



While there is marked specialization among the snakes of this 

 series in the possession of venom-conducting teeth, placed anteriorly, 

 and temporal glands for the copious storage of what appears to be the 

 most deadly of the venoms existing among all the known types of 

 poisonous reptiles, the character of the food seems to be analagous to 

 that of the Aglypha — the wholly non-venomous Colubrine forms. 

 However, in making up a general average of the species of the Pro- 

 teroglypha observed alive, the number showing a preference for warm- 

 blooded prey is slightly in excess over those feeding upon other rep- 

 tiles or batrachians. One peculiarity may here be noted. The 

 writer has had under observation no serpent of this series that indi- 

 cated a tendency to feed exclusively upon warm-blooded prey — a 

 trait already noted among several genera of the Colubrinae, 



