Page:Quarterly Journal of the Geological Society of London, vol. 25.djvu/519

 of an annular fragment of a former fang around the fang of an effective tooth, in one of the alveoli, and the protrusion of a small immature tooth from a very wide alveolus demonstrate that, as regards the manner of succession of its teeth, this Kimmeridge crocodile agreed with those of our day.

Minute Structure of the Teeth. — This so agrees substantially with that of living crocodiles that any description of it is superfluous. Vertebroe. — All the vertebrae which have as yet been brought into view are amphicoelian ; both articular faces are moderately but decidedly hollow. The centrum is cylindrical. Its middle is constricted, laterally compressed, and overhung by the expanded articular ends, which have a nearly circular contour. In two of these vertebrae the posterior articular face has a rather longer vertical than transverse diameter, due, I think, to squeezing. The neurapophyses, laterally compressed and thin, are suturally attached to the upper surface of the centrum in nearly its whole length, reaching, however, nearer to the anterior than to the posterior margin. The suture descends slightly on the side of the centrum. The neural canal is indented by a median groove, deepest at its middle, as is common in some crocodiles.

One of the vertebrae, imbedded in the same mass of stone as the ramus of the lower jaw in the British Museum, has a costal facet on the anterior margin of the transverse process (diapophysis), near its root, and none on the centrum, which makes it correspond to the fifth or sixth thoracic vertebra of living crocodiles, from which, however, it differs in the absence of a hypapophysis, as well as in its hollow articular faces. Beyond this costal facet the process has been unfortunately broken off ; its section is trihedral ; but the right transverse process is 3.5 inches long. The posterior border of the neural spine, the only part of it visible, is .5 inch thick, and 2.5 inches long, and it ends abruptly as if squarely truncated. The postzygapophyses are very mutilated, they project backwards from the neurapophysis, their lower ends nearly meeting in the level of the crown of the neural canal, and their upper ends diverging. Their surfaces look outwards and downwards. In another vertebral arch with the spine and transverse processes, which, from the great length of these latter, together with the absence of a costal facet from their anterior margin, I presume to be part of a posterior thoracic vertebra, the prezygapophyses project from the front of the neurapophyses just above the plane of the anterior margin of the diapophyses. The size and form of their surface are hidden ; but their anterior margin is part of a very wide curve, and it has a more nearly upward direction than in the more anterior vertebra.

VOL. XXV. PART I. 2 E