1888.] ANATOMY OF THE MESOSUCHIA. 425 



dilian basilar piece + the two pleurocentra + the pars odontoidea. 

 As bearing on this it is not without significance that the lateral 

 surface of the atlas in Ichtkyosaurits is impressed by a diapophysial 

 and a parapophysial pit, for the double costal articulation, as occurs 

 in the axis and the other vertebra behind it. 



There remains for discussion the inverted V-like piece that caps 

 and superiorly closes the neural arch. As already said, this is 

 missing in Mr. Leeds's specimens, but of its former presence no doubt 

 may be entertained. What is its morphological import ? Cuvier's 

 view that it represents the proc. spinosus of other vertebrae was the 

 doctrine generally accepted until about 10 years ago, when P. Albrecht 

 advanced reasons for regardino; it as a vestige of a vertebra ances- 

 trally present between the atlas and the skull, but since suppressed. 

 To this he attached the name proatlas. Albrecht's principal 

 ground for this conception of the nature of the " piece superieure " 

 appears to be the emergence of the first spinal nerve in front of the 

 neural arch of the atlas, for which reason it is by some named sub- 

 occipital nerve, whereas all the other spinal nerves escape from the 

 neural canal behind or through the neurapophysis of the vertebra to 

 which they serially correspond. Au approximately vertical plane 

 laid through the point of emergence of a spinal nerve will divide the 

 neurapophysis into an anterior part bearing the prfezygapophysis, 

 and a posterior portion supporting the postzygapophysis and the 

 spinous process. The neurapophysis appears to have two roots, of 

 which the posterior may be ligamentous, and the nerve passing out 

 between these leaves the neural canal not, Albrecht says, interver- 

 tebrally as commonly taught, but vertebrally by piercing through 

 the neurapophysis, which point of exit is morphologically interver- 

 tebrally situated. Now the vertebral complex called the atlas lies 

 behind the first spinal nerve, and siace the serial correspondence of 

 the spinal nerves and vertebrse expressed in numerical order is not as 

 2 : 2 or 3 : 3, but as 2 : (2—1), or 3 : (3— 1) ; or, to express the 

 same circumstance another way, since the second and third spinal 

 nerves correspond respectively to the vertebrse next in front of them, 

 it follows that the first spinal or suboccipital nerve does not corre- 

 spond to the atlas, but to a vertebra serially in advance of this. A 

 vestige of such an anterior vertebra Albrecht discovers in Cuvier's 

 piece superieure. This he regards as representing the neural arch 

 of the ancestrally present, now suppressed, vertebra once interposed 

 between the atlas and the occiput (3 1 a). This superior element was 

 subsequently discovered by Albrecht m Hatteria (32). Dr. G. Baur 

 has found it present in Chameleo, sp. (33). Prof. O. C. Marsh has ob- 

 served its presence in Morosaurus and Brontosaurus (34). L. Dollo 

 also has noticed it in Iguanodon (35). Its presence seems always 

 associated with incomplete coalescence and synostosis of the two 

 sides of the neural arch, and with the absence of a normal spinous 

 process ; and this is not without significance, for it hints that after 

 all Cuvier's view respecting it may express the truth. The develop- 

 ment of the " piece superieure " in two halves and its discontinuity 

 from the atlautal neurapophyses are not irreconcilable with such 



