AND AVES OF NORTH AMERICA. 



220 



to a transverse section of the opening, as in some species of Python. The floor of the neural canal is somewhat be- 

 low the plane of the anterior zygapophyses. The zygosphen is much elevated, and its summit concave. It is nar- 

 rower than the articular cup, a feature which appears to be peouliar to fossil serpents. Its anterior surface is 

 deeply excavated, and the lateral margins extend from the base downwards and outwards to the superior edge of the 

 cup, thus continuing the sides of the neural canal forward. The zygantrutn is comparitively shallow, and has its great- 

 est depth on the medial line. Its base extends downwards and backwards until directly over the superior margin of 

 tli<; ball, protecting also from above the neural canal posteriorly. The neural spine is much oompressed, triangular in 

 outline at its base, and does not extend to the anterior margin of the zygosphen. The base of the neural arch is 

 Strengthened by a thick rounded ridge, which unites the zygapophyses on each side;. It begins below the plane of the 

 anterior zygapophysis, and ascending slightly, joins the posterior zygapophysis above the articular face. Remnants 

 of two hypapophyses connected by a prominent ridge exist, and show that the anterior process was much the smaller 

 and its base somewhat separated from the margin of tin; cup. 



The principal dimensions of this vertebra, which is apparently an anterior dorsal, are as follows:— 



Lines, 



Length of centrum from edge of cup to convexity of ball, 14.25 



Transverse diameter of cup, •'• 1 " 

 Vertical diameter of cup, '■'' 



Distance from top of zygosphen to lower margin of cup, 13.25 

 Vortical diameter of ball, 7.25 



Width of neural canal in front, 8.75 



Height of neural canal in front, 2-" 



From the Eocene mail of Shark River, Monmouth Co. N. .T. 



Prof. Marsh regards this and the two species of serpents already described, as representing a genus distinct from 

 Palseophis which he characterizes as follows, "Among the most apparent differences between this genus and Paleeophis, 

 to which it is closely related, are the following:— The base of the neural spine in Dinophis is not co-extensive with the 

 supporting arch, but rises a short distance baok of the anterior margin of the zygosphen, as in the existing Python. 



The ridge connecting the anterior and posterior zygapophyses is much more developed, and its continuous crest more 

 rounded. The sloping sides of the arch above the neural canal are more deeply concave, and an obtuse ridge separ- 

 ates the concavity from the anterior zygapophysial notch. The neural canal has a Sharp longitudinal ridge on each 

 side, which gives its transverse section a, trifoliate outline, as in Python. The zygosphen is more excavated ante- 

 riorly, and its summit is concave. The zygapophyses are more extended outwardly, and the hypapophysial ridge is 



more prominent." 



The most important character here cited appears to be the non-coincidence of the base of the neural Spine with 

 the neural arch. This charaoter is true of P. Uttoralis to such a, slight degree as to render its importance problemati- 

 cal. The other characters adduced, are of a Comparative nature, so that until the limits of the proposed genus, if 

 valid, can be more fully defined, I retain the species in I'a.la'ophis. 



AMF.H.I. IMIILO. SOO. — VOL. XTV. 



•58 



