86 



University of California. 



[ Vol. 8 



centra to form chevron-bones, the form of the elements in the 

 pelvic arch, and the specialization of the humerus in at least 

 two species, it differs from the previously known genera of this 

 order. 



So far as approximation to a primitive type is concerned, the 

 probable existence of a larger number of cervical intereentra than 

 appear in other forms of this family is about the only really 

 primitive character recognized. The separation of the radius 

 and ulna, and the tibia and fibula, and the greater development 

 of the pelvic arch are characters which, with the separate zygo- 

 pophysial facets, indicate less complete accommodation of the 

 Shastasaurian skeleton to purely aquatic conditions than we find 

 in the Jurassic Ichthyosaurus. These characters are found also in 

 the Triassic Mixosaurus, excepting those of the pelvic girdle, 

 which has not yet been described from that genus ; and to them 

 should be added, as a common characteristic, the thickened spines 

 of the upper arches. Others, notably the modification of the 

 dorsal rib articulation, the formation of chevron-bones, and the 

 shortening of the humerus, are specializations not found in this 

 genus. The closest affinities of Shastasaurus are evidently with 

 Mixosaurus, from which it is, however, sharply separated by the 

 characters just mentioned, unless the structure of the forms in 

 that genus be very different from that which has been ascribed 

 to them. 



The primitive characteristics found in the epipodial segments 

 of the limbs and in the zygapophyses would probably be con- 

 sidered as sufficient to indicate somewhat greater age for this 

 genus than the lower Jura with its longipinnate forms, in which a 

 very small gap may exist between the epipodial bones. The spec- 

 ializations in the rib articulation, chevrons, and humerus, indicate 

 a distinct line of descent which started with a form like Mixo- 

 saurus, and which in the upper Trias still retained some of its 

 early characteristics, but had already gone a considerable distance 

 along several lines of specialization that were never followed in the 

 other genera. So far as we now know this line ended in the late 

 Trias. Through the kindness of Professor W. C. Knight of the 

 University of Wyoming, and Dr. Charles R. Eastman of Harvard 

 University, the writer has been enabled to examine a series of 



