50 THK ANMOMY OF HYLA CKRUI FA WHITE. 



Allans (Fig. 4, A, B.) On the dorsal surface is a neural 

 spine, represented by a median longitudinal ridge, which 

 is more strongly developed posteriorly. The anterior edge 

 of the neural arch is notched slightly in the mid-line dorsally. 

 The posterior zygapophyses are small, and the anterior 

 are absent. The posterior articular face of the centrum 

 is strongly flattened dorso-ventrally. Posterior notches 

 in the pedicles are present for the anterior portions of the 

 first intervertebral foramina, l>ut there are no anterior ones. 



Second vertebra (Fig. 5). The transverse processes 

 are horizontal and well-developed. The posterior 

 zygapophyses are normal, but the anterior are small, 

 in agreement with the small size of the posterior processes 

 of the atlas. The neural spine is comparatively prominent 

 and projects in front. Posteriorly, ridges are given off 

 from it to the posterior zygapophyses. 



Third vertebra (Fig. 6). The transverse processes are 

 very well developed, being next in size to those of the 

 ninth or sacral vertebra. In the vertical plane the trans- 

 verse processes make an angle of 90 degrees with the 

 longitudinal axis of the centrum, but in the horizontal 

 plane they are directed ventrally at an angle of about 

 20 degrees. The neural spine is comparatively well 

 developed, like that of the second, but projects posteriorly. 



Fourth vertebra. The transverse processes are well 

 developed, and directed posteriorly. 



Fifth vertebra (Fig. 7). The transverse processes are 

 slender and directed posterior^ like the former. 



Sixth vertebra. In the sixth the transverse processes 

 are slender but horizontal. 



Seventh vertebra. The only difference ot importance 

 between this vertebra, and the sixth is that in the former 

 the transverse processes project anteriorl3^ 



Eighth vertebra. The transverse processes are similar 

 to those of the seventh, but they approach the vertical 

 more closel}^ than the latter. Nicholls (7) draws attention 

 to a statement of Gadow (2) in which the latter remarks 

 that the eighth vertebra is bi-concave in the procoelous 

 famiUes of Anura. Nicholls shows that this is incorrect, 



