306 E. S. Riggs — B rachiosaurus altithorax. 



its greater length, its subovate outline, the lateral direction of 

 its gleuoid surface and the proximity of the same to the surface 

 for sternal articulation. The femur is considerably longer, but 

 making allowance for distortion the two could not be generic- 

 allv distinguished. The sacra differ in the presence of small 

 cavities in the centra of this form. The anterior caudals of 

 both are biconcave, their diapophyses similar, but the neural 

 spine in Camarasaurus, according to Cope's measurement, is 

 fully twice as long as that of this specimen. The essential 

 difference in the vertebrae lies in the thoracic centra. Those 

 of Brachiosaurus, as has been pointed out, range from 39 to 

 43 centimeters in length and from 37 to 35 centimeters in 

 breadth. A " lumbar" vertebrae of Camarasaurus is described 

 by Cope as being IT centimeters in length and 42 in breadth. 

 This difference alone w T ould seem to warrant generic distinction. 

 If Atlantosaurus be regarded as a valid genus, there is not 

 enough in common between its type specimen and this one to 

 determine their relationships. Amphicoelias need hardly be 

 considered in this connection. 



The further description of the sacral and iDresacral vertebrae 

 of Brachiosaurus will be taken up in another paper. It is to 

 be hoped that their removal from the matrix and careful study 

 will establish the position of this genus. 



Field Columbian Museum, Chicago. 



