1909. | SUBDIVISIONS OF THE BODY-CAVITY IN BIRDS. 229 
take place in the Varanide as in snakes, by the gradual growth 
of connective tissue from the ventral surface of the lung round 
to the dorsal side so as to fill up any cavity that originally 
surrounded it. 
In the Chelonia, Bertelli describes the exclusion of the lungs 
from the general body-cavity by the development of a septum 
which he takes to be the homologue of the “diaframma ornitico.” 
In the adult Testudo greca, this forms the ventral boundary of 
the lungs-—round which it is fused with the lateral body-walls— 
as a layer of fibrous connective tissue covered by peritoneum. 
This condition 1s connected with that usually found among 
reptiles by intermediate stages, seen in Vhalassochelys caretta, 
where this layer is very thin and does not completely shut off the 
lungs from the other viscera, and in Hmys lutaria, where it is 
less developed, and the lungs for the greater part project free 
into the pleuro-peritoneal cavity. In an embryo of Testudo, 
32 days after oviposition, the relations of the lung rudiments and 
the dorsal and ventral pulmonary ligaments are essentially the same 
as those in a chick of 72 hours’ incubation (text-fig. 24,p.212); and 
for some time development proceeds in the same way as in a bird, 
the lungs extending laterally and dorsally, the growth in the 
former direction tending to close off the pleural from the rest of 
the peritoneal cavity, and that in the latter to reduce the size of 
the pleural part of the celom. Later, the thick layer of tissue on 
the ventral surface of the lungs, which separates these organs 
from the underlying liver, meets and fuses with the lateral body- 
walls, and thus constitutes an almost complete septum across the 
peritoneal cavity. At the same time, the lungs unite with the 
dorsal and lateral walls of the pleural cavity so as to obliterate 
the ccelom in this region. 
The diaphragm in 7'’estudo is thus formed, according to Bertelli, 
out of the same constituents as is the diaphragm of birds, and has 
the same position and relations. Also the gradual obliteration of 
the pleural cavities takes place in the same way in the tortoise 
embryo as in the chick ; although in the latter it is never com- 
plete, and there remains throughout life a narrow but distinct 
cavity bridged across here and there by strands of connective 
tissue. Since, however, the growth of the lungs in a dorsal and 
a lateral direction goes on at the same time, the obliteration of 
the cavity round the lungs, and the completion of the diaphragm, 
take place practically simultaneously; so that by the time the 
diaphragm is constituted in order to close off the pleural part 
from the rest of the peritoneal cavity, no pleural cavity remains. 
It appears, therefore, to be more likely that the condition seen in 
Testudo is to be compared with that occurring in the Varanide 
rather than with that found in Birds. 
From the structure of several Crocodiles which I had the 
opportunity of examining, I believe that the arrangement of the 
ccelomic septa in this group of reptiles approaches most nearly to 
the avian condition, as Huxley and Beddard have already pointed 
