Merriam.] 



Triassic Idi tli yopterygia . 



73 



inferior ends, and when the lower portions of the opposed sur- 

 faces are in contact, the upper ends are separated by a consider- 

 able cleft (PI. If), fig. (i). That the upper portions of the 

 opposite surfaces were normally not so near together as the lower is 

 indicated by the impossible position taken by the rib when 

 they are brought together, the shaft being then directed upward 

 instead of downward. Their roughness, as compared with the 

 lower surfaces, shows that there was more cartilage between them. 

 It has already been noted that the lateral extension of the depres- 

 sion beneath the pedicel of the upper arch may stretch over 

 the upper portion of the diapophysis. In the cervicals this 

 groove is short and covered by the spine-like, lateral projection 

 of the pedicel. In the middle dorsals, where the pedicels are not 

 greatly expanded, the groove is even longer than in the cervicals 

 and extends into the roughened upper surface of the diapophysis, 

 where it is opposed to the rib head (PI. S, fig. .1). The rough- 

 ened and grooved upper surface of the diapophysis was evidently 

 not a surface of direct articulation, but was covered by cartilage 

 which connected it with the upper portion of the rib head, 

 while the smoother lower surface articulated almost directly 

 with the portion of the rib opposite it. 



As is apparent, the whole plan of rib attachment through 

 the greater part of the dorsal region is quite different from 

 that of the typical Ichthyosauria. In Ichihyosaurus the 

 diapophysis and parapophysis are both present, are quite 

 distinct, and are of approximately equal size as far back in 

 the column as the posterior dorsal region, where they unite. 

 Also, the articular surfaces of the diapophyses which are confluent 

 with the neurapophysial surfaces in the cervical region, as in 

 SJiastasaurus , become separated from them at about the four- 

 teenth or fifteenth vertebra instead of at the thirty-fifth or a later 

 one. In SJiastasaurus true double articulation is found in 

 only a few of the most anterior vertebra?, and even there the 

 parapophysis is relatively very small. The loss of the lower 

 articular surface has been compensated for in the dorsal region, 

 next the lungs, by elongation of the diapophysis and develop- 

 ment of the method of attachment of the ribs just described. 

 This permits a rocking movement of the ribs on the vertebra?, 



