218 MISS M. POOLE ON THE DEVELOPMENT OF THE  [ Mar. 2, 
‘« diaframma primario ” because he says it isa temporary structure 
which will later give rise to the “ diaframma definitivo” and also 
to the ventral walls of the intermediate air-sacs. At this stage— 
ten days after incubation—he says the anterior intermediate air- 
sacs lie enclosed in the primary diaphragm, and by splitting it 
into two layers, form with their dorsal walls the definitive 
diaphragm and with their ventral walls the so-called “ diaframma 
toraco-addominale.” The posterior intermediate and the posterior 
air-sacs likewise split the primary diaphragm into two layers; and 
behind this he describes them as lying in the lateral abdominal 
walls, covered by their peritoneum, and bounded externally by the 
connective tissue which surrounds the muscles of the body-wall. 
Now I do not find any signs of these air-sacs actually penetrating 
into the walls of the peritoneal cavity so as to lie between its 
bounding celomic epithelium and the musculature of the body- 
wall. According to my observations, the air-sacs always remain 
enclosed in the backward prolongations of the post-pulmonary 
septa, and this tissue becomes closely attached—throughout its 
entire thickness—to the lateral body-walls. Between the post- 
pulmonary tissue surrounding the air-sacs and the connective 
tissue of the body-wall, in all my sections there is a darkly- 
staining streak, showing—I believe—where the two layers of 
ceelomic epithelium have fused together (text-figs. 29, 30, 31, cf). 
And I differ a little from Bertelli as to the formation of his 
“ diaframma definitivo” from the antero-dorsal portion of the 
“ diaframma primario ” ; though this divergence of opinion depends 
entirely upon my own conception of the essential nature of a 
coelomic septum. I consider that sucha septum should be covered 
on both surfaces by peritoneum, and really separate one division 
of the celom from another; and therefore I prefer—following 
Butler—to regard the whole post-pulmonary mass of tissue, with 
the intermediate and posterior air-sacs enclosed between its 
antero-dorsal and postero-ventral walls—the “ diaframma ornitico” 
and the so-called *‘diaframma toraco-addominale” respectively of 
Bertelli—as a single septum. Except for the presence of the costo- 
pulmonary muscles in the antero-dorsal half of this septum, 
there is nothing to distinguish it, as the true diaphragm, from 
the postero-ventral portion, in the first stages when the splitting 
into two layers occurs with the development of the air-sacs. 
Besides which, the ‘“diaframma ornitico” of Bertelli only 
separates the pleural cavities from the cavities of the inter- 
mediate air-sacs, and is consequently covered only on one face 
by peritoneal epithelium; the other surface being invested by 
the endoderm of the air-sacs. What I have described as the 
post-pulmonary septum, on the other hand, does really separate 
the pleural from the peritoneal cavity, and is covered ventrally, 
as well as dorsally—except where the septum is continuous with 
the tissue of the lungs—by ceelomic epithelium. The interposition 
of the air-sacs between its two surfaces, does not, I believe, 
essentially alter the nature of the whole. My observations 
