146 A. Smith Woodward — Stereosterimm from Brazil. 



importance of the structure of the tarsus, which exhibits a separate 

 tarsal bone supporting each of the five digits, and made Stereosternum 

 the type of a new reptilian order, termed Proganosauria. Other 

 specimens wanting the greater part of the neck and tail were 

 acquired by the British Museum, and these were noticed by Mr. 

 Lydekker in 1889,^ described by Professor Seeley in 1892.^ It is 

 now evident that the animal is related in some undetermined way to 

 the ancestry of the Plesiosauria. 



The new specimen of Stereosternum is shown in PI. V of about 

 half the natural size, and is curled up on the slab of limestone so 

 that the end of the snout touches or overlaps the slender extremity 

 of the tail. The head and neck are only indicated in impression ; 

 but it is clear that the head is of elongate-triangular form, with 

 extremely delicate bones, and if the tip of the snout be truly 

 represented by the end of the impression near the recurved tail, 

 the length of the head is much less than that of the neck. Although 

 the hindermost cervical vertebrge are proved to be short by one 

 of the British Museum specimens, the impression seems to show that 

 the centra of those nearer the head are considerably longer than 

 deep. The skeleton is sharply bent and imperfect in the region 

 of the pectoral arch; but the number of vertebrge between the neck 

 and the sacrum must be about twenty, as already determined in 

 a British Museum specimen, while the series of ribs is well shown 

 on one side. The remains of the hind-limbs correspond with 

 those of previous specimens. Assuming that an obscure fragment 

 of the pelvis overlaps the transverse processes of the two hinder 

 sacral vertebrae, the tail comprises not less than sixty vertebrte, 

 of which at least the foremost seven bear robust transverse 

 processes gradual!}' diminishing in size backwards. The centra of 

 the first twelve caudals are compai-atively short and stout and 

 exposed from above or below; the following thirty-five vertebrae, 

 which are more elongated, are shown from the side, with the neural 

 arch projecting within the curve of the tail, and the delicate loose 

 chevrons lying parallel with the long axis of the centra on the 

 outer side of the curve ; the remaining vertebrge, to the number 

 of about fourteen, are very slender and indicated only by an obscure 

 impression. As a whole the tail is thus somewhat more than twice 

 as long as the trunk, occupying slightly less than three-fifths of the 

 entire length of the animal. 



A detached tail of a precisely similar character, discovered in the 

 same stratum and locality as the portions of trunk of Stereosternum 

 in the British Museum (i.e. Ipanema), is interesting as showing not 

 merely the skeleton but also an impression of the soft parts. This 

 specimen (Brit. Mus. No. R. 538) is 0-13 m. in length and displays 

 forty vertebrge in regular series, those towards the hinder extremity 



1 R. Lydekker, " Catal. Foss. Eept. Brit. Mus.," pt. ii (1889), p. 302. 

 Though the locality is not mentioned in the Catalogue, the dealer's lahels state that 

 these specimens were obtained from Ipanema. 



2 n. G. Seeley, "The Mesosauria of South Africa " : Quart. Journ. Geol. Soc. 

 vol. xlviii (1892), pp. 598-602, figs. 2-4. 



