2 20 Bulletin American Museum of Natural History. [Vol. X, 



approaches in its structure that which Marsh has figured as the 

 4th, but is somewhat more complex. Here follows another 

 interval (Fig. 5), in which would lie the vertebrae described by 

 Marsh as the 4th. (Fig. 6.) 



The estimated 12th, 13th and 14th succeed each other with a 

 regular diminution in size, and a steady reduction of the trans- 



FiLi. 0. r>] oittosau- 

 rus excels us. Sup- 

 posed 4th caudal ver- 

 tebra. After Marsh, 

 1897, Plate xxiv. Ap- 

 proximately IB nat- 

 ural size. 



Fig. 5. Catnarasauriis. Anterior caudals. First caudal and 5th, 

 8th, Qth, i2th caudals as estimated. Am. Mus. Coll. R. 222. Approxi- 

 mately 10 natural size. 



verse processes, which gradually transforms into 

 a horizontal plate. The next vertebra preserved 

 is the estimated i6th, which exhibits a very 

 marked reduction in the transverse process, and 

 this disappears entirely in the supposed 20th and 

 2 1 St, in front of which at least two vertebrae are 

 missing. (See also Note i, p. 233.) 



The missing i6th and 17th vertebras were 

 probably detached in the manner described 

 above, and the 16 or 20 vertebrae at the end of 

 the tail were also drawn away. In front of the 

 15th no tooth marks are observed. 



