204 
sent out through the right aorta, while the left aorta is purely arterial. 
This is again the same scheme as we found in the mammal and results 
in the head receiving a better quality of blood. 
Experiments were performed on three species of turtles to ascertain 
if this condition prevailed. The plastron removed and the heart laid bare, 
a double ligature was passed through the transverse pericardial sinus 
and arranged to tie one at the distal, the other at the proximal edge of 
the sinus. Next cornstarch granules suspended in normal salt solution 
were introduced into the auricles during diastole; the auricle allowed to 
contract, giving time to have the distal ligature ready to tie off; the distal 
ligature was tightened during ventricular systole and immediately the 
proximal one—isolating three columns of blood in the three vessels. These 
were bled separately into watch glasses containing dilute acetic acid and 
examined for the granules. It was found that granules injected into the 
right and left, and in both auricles simultaneously, were always recovered 
from all three efferent vessels. It must also be remembered that in the 
turtle the fetal circulation is not unlike that found in the fetal bird—the 
posteaval vein conveys the oxygenated blood, and if this segregation of 
blood occurred as described in the adult, the head would receive only 
venous blood. This objection aiso holds good in the Crocodilia, where, 
according to Wiedersheim (6), the condition is as follows: “The blood from 
the right ventricle passes into the pulmonary artery as well as into the 
left aortic arch and, according as the septum ventriculorum is complete 
or incomplete, is either entirely venous (Crocodilia) or mixed (other rep- 
tiles). A complete septum ventriculorum thus appears for the first time 
in crocodiles, in which, consequently, the right ventricle contains unmixed 
venous blood and the left ventricle unmixed oxygenated blood, although, 
as will be seen presently, an admixture takes place in the systemic ar- 
teries.” Again, according to this scheme, the head will receive a better 
quality of blood because the carotids arise from the left aortic arch, but 
again the objection as to the manner of transformation from the fetal 
crocodile to the adult crocodile heart would arise. This form certainly 
needs careful investigation. The purely venous blood would far exceed 
the purely arterial, and the mixture at the foramen of Panizza might be 
very complete. 
The amphibian circulation is naturally described on the basis of the 
segregation of blood and must therefore fall into two classes, the anural 
and the urodele. The description of the anural circulation is delightfully 
