The Mechanism of Pulmonary Respiration in Araphibians with Gill Clefts. 67 



removed from the oral cavity, as in Necturus. This preparatory 

 stage is usually followed by two or three respiratory acts, during 

 which the mouth and gill clefts remain closed and air passes in and 

 out through the nasal cavity. The first of these is a normal respira- 

 tory act, including aspiration, expiration and inspiration, but the 

 movements of the floor of the mouth do not have great amplitude 

 and relatively little air enters the lungs. In the supplementary acts, 

 which follow in close succession after the first, expiration is usually 

 omitted and aspiration is followed immediately by inspiration, the 

 mouth being well filled with pure air which is sent directly to the 

 lungs. During these stronger inspirations a conspicuous pulse-like 

 wave passes from the anterior to the posterior part of the abdomen. 



After the last inspiration the surplus air remaining in the mouth 

 is expelled, usually through the gill clefts, and the animal then 

 resumes its acquatic respiration. 



The foregoing account is based on the study of large specimens 

 of Amphiuma. An isolated respiratory act which is common in Nec- 

 turus and other amphibians, was occasionally seen in these specimens 

 but two or three inspirations were usually employed to fill the lungs. 



Cryptobranchus allegheniensis. The pulmonary air is renewed 

 in Cryptobranchus usually by a Single respiratory act, but one or 

 two supplementary acts may sometimes be observed. Air passes 

 through the nasal cavity both in inspiration and expiration and the 

 gill clefts are closed during the entire process. In inspiration the 

 nasal passages are also closed. After the lungs are filled the surplus 

 air is expelled and acquatic respiration is renewed. 



Larva! amphibians. The Statement of von Siebold (1828), 

 Duges (1838), and Owen (1866), that larval amphibians take air by 

 the open mouth and force it into the lungs by a swallowiug move- 

 ment, suggests at once a respiratory process more or less similar to 

 that of Necturus and Siren. My observations show that such a 

 resemblance actually exists. In the large larva of Amblystoma tigri- 

 num, from the moment of taking air by the open mouth to the 

 emission of surplus air through the mouth or gill clefts, the process 

 is essentially the same as in Necturus. 



In the large larva of the bullfrog, Rana catesbiana, the same 

 general characteristics may be easily observed. The precise moment 

 of inspiration and expiration is indicated by movements of the wall 

 of the abdomen and the relation of these movements to the rise and 

 fall of the floor of the mouth can be easily determined. No air 



5* 



