LTASSIC FORMATIONS. 
89 
a few feeble ridges at its peripheral and commonly lower third part. It is rare to find this 
element unanchylosed with the succeeding centrum (ib., fig. 5, c x)- The neural surface of 
the atlantal centrum (ib., fig. 3) is divided equally between the medial quadrate tract 
{m) for the myelon,^ and the lateral subtriangular depressions [n p) for tlie neurapophyses. 
The length or fore-and-aft diameter of the atlantal centrum is usually relatively less than 
in the trunk- or tail-vertebrae of the same individual. The neiirapophysial surface bends 
down upon the side of the centrum, forming a prominence (fig. 5, d) on its upper part, 
representing the diapophysis ; beneath this, with a non-articular interval, projects a low 
obtuse parapophysis (ib., p). 
The neurapophyses of the atlas (fig. 6, n), as far as I have been able to infer from this 
commonly mutilated or much disturbed part of the fossil skeletons, were not united 
together atop, or there developed into an exogenous spine, but retained their distinct- 
ness, like their antecedent homotypes the exoccipitals (PI. XXVI, fig. 1, 2). In the best 
preserved specimen each atlantal neurapophysis is bent back at the middle of its length, 
the upper compressed portion overlapping the fore part of the base of the neural spine 
of the axis, as shown in PI. XXIII, fig. 6. 
Of the existence of the atlantal pleurapophysis (PI. XXIII, fig. 5, pi, a), each being 
joined by a bifurcate proximal end to the di- and par-apophyses of the centrum, there is 
better evidence. Such rib was short, directed outward and backward ; and is uncon- 
nected, distally, with any haemapophysis (PI. XXI, fig. 1, pi). 
The hypapophysis of the atlas (PI. XXIII, fig. 5, hj, a, and PI. XXVII, fig. 2) is an 
irregular triangular robust ossicle, smooth and convex on its inferior and free surface, with 
the opposite articnlar surface divided into three facets. The anterior of these (ib., fig. 
2, a) is smooth and concave, completing, with the concave part of the atlantal centrum, 
the cup for the basioccipital ball; an almost flat rough tract (ib., h) next below articulates 
synchondrosaly with the corresponding rough surface of its centrum. Beneath this is a 
smaller flat roughish surface (ib., c), sloping backward from the one above, for articulation 
with the succeeding hypapophysis. Such are the complex characters of the first trunk- 
vertebra of Ichthyosaurus. 
The centrum of the second vertebra (PI. XXIII, fig. 5, cx, and PI. XXVII, figs. 
3, 4, 5), has a flattened, roughish, anterior surface (fig. 4), like the posterior one of the 
first vertebra, with which it sooner or later coalesces. The hind surface of the axis- 
centrum (fig. 5) is more deeply and entirely concave, with a sharpish circumferential 
margin. On the upper surface of the centrum (fig. 3) the myelonal surface {m) is similar 
in size and shape to that of the atlas, but is shallower. The neurapophysial surfaces 
{n, n) are less excavated, and the diapophysial productions (fig. 4, d, d) upon the sides of 
the centrum are more prominent, better defined as processes. The same may be said of 
the parapophyses (ib., p,p), which project close to the fore border of the side surface and 
show a more distinct facet for the head of the axial rib than do those of the atlas. The 
^ ‘ Spinal marrow,’ ‘ spinal cord,’ of ‘ Anthropotomy.’ 
