1886. ] Aquatic Respiration in Soft-shelled Turtles. 235 
molecular oxygen stored up in the tissues, and (0) from the air in 
the lungs. Analysis of the air taken from -the lungs after the 
turtle had been wholly submerged for ten hours, showed only a 
slight trace of either oxygen or carbon dioxide. So far as this 
single experiment goes, we conclude that if the lungs were 
moderately filled with air upon the immersion of the turtle, the 
amount of oxygen that might be taken from the air in the lungs 
would fully account for the excess of carbon dioxide found in the 
water. That the aquatic respiration is due almost entirely to the 
pharynx and but slightly to the skin, is shown: (a) By anzsthetiza- 
tion, the turtle becoming anesthetized four or five times as quickly 
when kept entirely submerged in ztherized water as when allowed 
to come to the surface as frequently as it desired. (4) When the 
turtle’s skin was completely covered with vaseline and the turtle 
kept wholly submerged, the amount of oxygen removed from the 
water and of carbon dioxide added thereto was nearly as great as 
when the skin was unvaselined. 
In some at least of the hard-shelled turtles (Chelydra and 
Chrysemys) similar movements of the hyoid apparatus occur 
when they are submerged, and water is seen to enter the nos- 
tril and be expelled therefrom as in soft-shelled turtles. 
The pharynx expands and contracts with considerable regu- 
larity in all of the turtles, so far as we know, when they are in 
eair. These movements appear like those of the frog, but in 
the turtles they are unnecessary for filling the lungs. In frogs, 
however, they are necessary for this purpose, although as shown 
by Townson (1794), pharyngeal movements often occur in the 
frog without any air being forced into the lungs. 
As these movements are of undoubted use in respiration for 
the soft-shelled turtles in water, it seems probable that they may 
of use in respiration for all turtles in tHe air, that is, the mem- 
brane lining the pharynx probably acts as a respiratory organ 
whether the medium bathing it and containing free oxygen be 
_ air or water. 
These movements and their object, respiration, then seem to | 
connect, physiologically at least, the turtles on the one hand with 
the lower vertebrates—Amphibia and fishes—and on the other 
hand with the higher forms, viz., dog and man, for Garland has — oe ; 
shown that in the dog, and also in man, occur rhythmical me 
Fyngeal movements which draw air into the pharynx and expel it 
