43 
cylinder, but the edges do not fuse behind. Except that 
the eyeless condylar facet is larger than the ocular one 
the asymmetry is not noticeable below the neural spine. 
The centrum is overlapped by the large ill-defined anterior 
zygapophyses of the 2nd vertebra, and itself bears faint 
posterior zygapophyses at the bases of the neural arches, 
whilst on the ocular side the latter sends back a 
hook-like process which fits outside the neural arch 
of the second vertebra. This is the only trace of 
the characteristic method of articulation of the vertebree 
of the Cod. 
Second Vertebra (fig. 17, V’.2).—The neural arch is 
perforated on each side by two foramina for the roots of 
the third spinal nerve, but the bridge of bone separating 
the right pair is extremely slender. The large irregular 
anterior zygapophyses are asymmetrical, that on the eye- 
less side being much the larger. The neural spine is also 
asymmetrical as in the atlas, but the asymmetry extends 
down on to the neural arch. <A small pointed transverse 
process is present, with a long accessory rib (A.R.?) 
strongly attached to its base much lower down than the 
attachment of the first intermuscular bone. Both the 
centrum and neural arch bear post-zygapophysial facets 
not shown on the eyeless side. 
Third Vertebra (fig. 17, V.*)—Whole of the vertebra 
markedly asymmetrical, being more strongly developed 
on the eyeless side in every respect. Neural spine and 
spinal canal arch to the left. Anterior and posterior 
zygapophyses more strongly marked on ithe same side. 
The centrum not only bears a strengthening ridge (S.R.) 
which is much stronger on the eyeless side, but is itself 
more bulky on that side, so that the notochordal canal is 
eccentric in position. The transverse processes, like all 
those succeeding them, and as already described by 
