1889.] THE BODY-CAVITY IN LIZARDS, ETC. 469 



wall, without damage to the delicate septa and ligaments attached to 

 and passing into it. 



There appears to be a septum or " diaphragm " behind the lungs, 

 «?.a°, d.a° ; but this only extends for a short distance inwards from the 

 ventral body-wall, as is seen by comparing the figures 40 and 42, 

 which represent the condition in the ventral region, with figures 41 

 and 43, which are sections through the more dorsal part. There is, 

 in fact, a continuous pulmohepatic space 4, 4°, 4, 4", on either side 

 closed posteriorly by the post-hepatic septum (figs. 40, 42, and 43, 

 ft, ft, and fig. 41 ft+y & ft). I would compare the imperfect partition 

 (d.a°, d.a°, figs. 40 & 42) to the ventral portion of the avian 

 diaphragm. The septum behind the liver appears to me to be homo- 

 logous with the post-hepatic septum of Tupinamhis, or with the 

 ventral or omental part of the post-hepatic septum of the bird. And 

 perhaps the most dorsal part near the postero-dorsal extremity of the 

 lung and liver may represent the dorsal element (oblique abdominal 

 septum of the bird). 



But besides these lateral pulmo-hepatic cavities, there are, 

 anterior to the post-hepatic septum, two smaller sacs on either side 

 in the region of the liver, which must now be described. 



In fig. 42 we see on either side of the median ventral ligament 

 (m) a closed peritoneal sac (1, 1') containing a portion of the liver-lobe 

 of its side. These sacs I would roughly compare to the large ventral 

 liver sacs in the bird (figs. 29 et var. 1, 1'). However, the lateral 

 boundaries {Lob, l.ob) of these spaces in the Crocodile do not seem to 

 correspond exactly to any membranes in the bird, but to the oblique 

 ligaments of the liver described above (p. 466) in the Lacertilia, 

 as apparently complementary to the pulmohepatic ligaments which 

 are represented in all three groups. The relations of these spaces 

 (I, 1') and ligaments (l.ob, l.ob) are further illustrated in the trans- 

 verse sections (figs. 37 & 38) and in the longitudinal section (fig. 39). 



The second pair of cavities in the region of the liver are specially 

 worthy of consideration. These are the cavities 2 & 2' on the right 

 and left sides respectively in the longitudinal sections (figs. 40 & 43, 

 & 41 right side) and in the transverse sections (figs. 37 & 38). 

 I regard them as comparable to the pulmohepatic recesses of the 

 bird. That on the left side seems to be entirely closed, and since it 

 lies between the liver and the alimentary canal it is not bounded 

 antero-dorsally by the lung and mediastinal tissue, and it can only 

 be compared with the posterior and more median portion of the 

 corresponding recess in the bird. 



The space on the right side, however, much resembles the corre- 

 sponding space in the bird or lizard in its relations to the liver and 

 lung ; it is bounded on the outside by the pulmohepatic ligament (figs. 

 37, 40, 41, 43, a), which, as in the lizard or bird, passes postero- 

 dorsally into the membranous tract (y, fig. 38) that is continuous 

 with the posterior extremity of the pulmonary and mediastinal tissue 

 (oblique abdominal septum of bird). 



On the right side I found, in some of the young specimens 

 examined, a passage between the pulmohepatic recess and the 



