BRITISH FOSSIL REPTILES. 101 



formed by the odontoid epiphysis of the axis, which nearly equals 

 the body of the same vertebra in size. In the Lacertians, the 

 body of the atlas is a thin anruilar piece of bone, which forms 

 merely the circmnference of the articular cup for the occipital 

 tubercle ; the greater part of which articulates, as in the Croco- 

 diles and Turtles, with the odontoid epiphysis of the axis. The 

 consideration of these facts leads to the suggestion of another 

 hypothesis of the analogies of the anterior vertebrae in the Ich- 

 tliyosaurus. What has been described as the atlas anchylosed 

 with the axis, may be the true odontoid process of the axis ex • 

 hibiting the same excessive development and anterior concavity 

 adapted to the occiput, as in the Lacertian Sauria. The first sub- 

 vertebral wedge-bone, according to that view, would then repre- 

 sent the body of the atlas, as Mr. Conybeare indeed seems to 

 have regarded it, but reduced to a still more atrophied condition 

 than in the Crocodile or Turtle. Buthereanother difficulty pre- 

 sents itself: admitting the first subvertebral bone to be the 

 atrophied body of the atlas, what then it may be asked are the 

 second and third subvertebral bones ? I confess that in this, as in 

 some other of the problems of morphology, I see only a choice of 

 hypotheses of which none are free from objection. This at least 

 is certain, that the subvertebral, cervical, wedge-shaped ossicles, 

 which hitherto have been observed only in the Ichthyosaurus, 

 are most admirably adapted, as Sir Philip Egerton has well 

 pointed out, with the anchylosed condition of the atlas and axis, 

 to ensure the fixation of the head which is essential, in an active 

 predatory animal, to its swift and agile movements through the 

 water. 



The costal processes or ribs commence in the Ichthyosaurus at 

 the axis or second cervical vertebra, and are continued through 

 the anterior two thirds of the caudal region of the spine. Those of 

 the cervical and anterior part of the thoracic regions are slightly 

 bifurcate at their proximal extremity, and are articulated partly 

 with a tubercle on the centrum, which represents the inferior 

 transverse process, and partly to the outside of the base of the 

 neurapophysis : as they become placed further back, the two 

 heads become gradually blended into a single expanded proxi- 

 mal extremity, which at length becomes a simple convex tu- 

 bercle in the ribs of the caudal region. The ribs quickly in- 

 crease in length, which is greatest at the middle of the thoracic 

 abdominal cavity : from this point they become gradually ab- 

 breviated to the sacral vertebrae, and then suddenly contract 

 into short and straight appendages, which progressively dimi- 

 nish until they finally disappear. Those ribs, which are bifur- 

 cate or bilobed at their upper extremity, are traversed by a longi- 

 tudinal impression extending from the -aw^q of bifurcation along 



