158 



now and then by agitated swimming during which in some 

 instances the animals emitted a succession of squeaking noises. 

 Finally the caymans became passive, discharged considerable 

 air from the mouth, and dropped to the bottom where they soon 

 died. During the whole period of submergence their activities 

 were as a rule carried out quietly and without any movements 

 that could be called convulsive. The shortest period in which, 

 drowning occurred was 34 minutes, the longest 72 minutes and 

 the average for the five tests was 54 minutes. I conclude, 

 therefore, that caymans on being submerged may drown in a 

 little less than an hour. 



This period seemed to me surprisingly short and yet I had no 

 reason to suspect the accuracy of my observations. After each 

 cayman had suffered what was supposed to be death by drown- 

 ing it was taken out of the water and left for recovery in the air 

 but none ever revived. The drowning was unquestionable. 



Several good observers who were familiar with the habits of 

 our southern alligator, Alligator mississipiensis (Daudin), as- 

 sured me that these animals could remain under water without 

 drowning for periods much longer than those recorded for the 

 caymans. I therefore attempted by methods like those just 

 described to test the drowning of alligators. Through the 

 kindness oLDr. Thomas Barbour I was enabled to work on four 

 living alligators from Florida. All were small individuals meas- 

 uring in length about 26 cm. each. They were submerged 

 singly in an aquarium arranged in the same way as in the tests 

 with the caymans. On the whole the alligators were more con- 

 tinually active than were the caymans. The alligators often 

 made squeaking noises but eventually became quiet and sluggish 

 and finally subsided to the bottom of the aquarium where they 

 died. It was impossible to determine accurately the exact 

 moment of death but judged by their irresponsiveness to stimuli 

 an approximation to this could be made. Based upon this 

 approximation the periods necessary for drowning the alligators 

 varied from 5 hours and 20 minutes to 6 hours and 5 minutes 

 and averaged for the four animals 5 hours and about 40 minutes. 

 This length of time was more in accord with what had been 

 expected, and is in striking contrast with the times recorded 

 for the caymans. I know of no reason, however, for any inac- 

 curacy in the tests on the caymans and I conclude, therefore, 

 that Caiman sclerops is quickly drowned and that Alligator 

 mississipiensis may resist this fate for from five to six hours. 



Neither the cayman nor the alligator gave evidence of breath- 

 ing in any way except by lungs. Water tortoises have in addi- 

 tion to pulmonary respiration, oral and anal respiration. Oral 

 respiration is possible in consequence of the highly vascular 

 condition of the surfaces of the mouth and pharynx. S. H. 

 Gage and S. P. Gage (1886) have studied oral respiration in the 

 soft-shelled tortoises, Amyda and Aspidonectes, and have noted 

 that these animals will remain under water voluntarily as long 



