AAf: TKIASSTO [GHTHTOSAUBIA. r,:i 



parapo])liysis is gradually disappearing, and the articulation is becoming solely 

 tubercular or diapopliysial. Yakowlcw (1902, 1) lias suggested that the bicip- 

 ital articulation in Ichthyosauria has arisen through the splitting of single- 

 headed ribs, or by atrophy of the median region of the head. If, however, the 

 double-headed rib is derived from a single-headed form by reduction of the 

 middle portion of the head, the reduction would probably not begin, as it does 

 in the anterior dorsal region of Skastasaurus, by separation of a small and 

 apparently functionless tubercle some distance away from the main rib head 

 (fig. 57). The division would occur according to this method by gradual con- 

 striction of the rib head until a small space came to exist between two large 

 and functional heads. Though such a condition occurs in other reptilian 

 groups, it has not thus far been known to exist in the Triassic Ichthyosauria. 



In Shastasaurus and Cymbospondylus the parapophyses generally remain 

 about the same distance from the diapophysis no matter how large they are. 

 Though the diapophysis gradually increases in height, the lower border never 

 meets the parapophysis. In passing backward through the series the diapo- 

 physes gradually increase in size and the parapophyses decrease in size till the 

 latter disappear, but at the point of disappearance the parapophyses are not 

 absorbed into the diapophyses. The only case which might be considered as 

 an exception to this rule is that of a specimen of Cymbospondylus (fig. 58) in 

 which large parapophyses are present back to the twelfth centrum. On the 

 thirteenth centrum the diapophyses are much increased in size, while the 

 parapophysis has disappeared on one side and is represented only by a rudi- 

 ment on the other side. The diapophyses on the thirteenth centrum are also 

 of a peculiar form, being swting backward instead of forward. The para- 

 pophyses do not gradually approach the diapophyses on this specimen but are 

 separated from them by a nearly uniform space. 



If the double articulation has arisen from a single articulation in the forms 

 described above it must have come in most cases through addition of an entirely 

 independent rib head, and an inferior or parapophysial tubercle. This would 

 mean that in the case of the rudimentary structures in the anterior dorsal region, 

 both parapophysis and lower rib head appeared and developed to a considerable 

 size before they could have any important function; and that the local condi- 

 tions requiring a lower tubercle have progressed from the anterior cervical to 

 the anterior dorsal region in Shastasanrus, or to the posterior dorsal region in 

 Ichthyosaurus and Bapta/nodon. 



In view of what has been said above, it seems less probable that the double 

 articulation in the anterior dorsal region of the Shastasaurinae is incipient 

 than that it is rudimentary; and it is possibly in order to consider the forms 

 exhibiting this type of rib articulation as a side branch of the Ichthyosauria, 

 which early developed a tendency to reduce the dorsal ribs to a single head. 



