170 ON FOSSIL BIRDS FROM MADAGASCAR AND EGYPT. [Jail. 19, 



channelled by the groove for the extensor tendons (text-fig. 15, A, 

 e.g.) ; this groove is closed at its lower end by a strong ridge 

 running upwards and outwards towards the outer border, and 

 forming a prominent rugose surface above tlie outer condyle. 

 There was no bony bridge over the tendons, but a slight ridge 

 along the inner side of the groove marks the insertion of a strong 

 tendinous sling (see text-fig. 15, A, t.). 



Comparing this specimen with the tibio-tarsi of the otlier 

 Ratites, it may be said that to some extent it combines tlie 

 characters of several of them. In the depth of the extensor 

 groove it approximates to Bhea and ^Epyoi-nis. The oblique ridge 

 which blocks the lower end of the groove occurs also in jStruthio, 

 though in this case the groove itself is much shallower. In 

 ^lEjiyornis, on the other hand, the ridge is entirely absent and the 

 groove runs nearly down to the condyles. The form of the outer 

 condyle and its relations to the post-condylar process are similar to 

 what is seen in Rhea and to some extent in Struthio also. 



It seems clear that this bird difliers widely from the other 

 Ratites, and certainly from any of the Oarinatne, and I propose to 

 name it Eremopezus eoccenus. 



The dimensions of the type specimen are : — 



cm. 



Width of distal articular end 4'8 



Width of lower end of shaft (at highest 



point preserved) 3'5 



Width from front to back of the outer 

 condyle 3"5 



The occurrence in this region and in beds of Eocene age of 

 a Ratite bird is a matter of considerable intei'est ; but until 

 suflicient remains are found to give a more definite idea of 

 the relationships of this form, it will be unwise to draw any 

 very wide conclusions. At the same time, it may be remarked 

 that the existence of a true Ratite so long ago as the Eocene 

 makes it at least possible that some of the main groups of 

 the Ratitse may have had a common ancestry, and are not the 

 results of separate retrogressive modifications leading to the loss 

 of flight, with the various correlated changes. In the present 

 case the relationship between the Struthiones and the^pyornithes, 

 referred to by Burckhardt *, is natm^ally suggested ; and the 

 occurrence of Sti'uthious birds [Struthio and Hypselornis) in the 

 Pliocene of the Siwalik Hills may possibly indicate some remote 

 connection between the present form and the Bromceics-Casuarius 

 group. There is, however, another possibility that must not be 

 lost sight of, and that is, that after all Eremopezus may be merely 

 another instance of reti'ogressive modification leading to loss of 

 flight and increase of bulk in a group of Carinate biixls, such as 

 has occurred in the case of the Stereornithes, the Gastornithes, and 



* " TJeber ^pyornis," Palasont. Abhaiidlungen, N. F., vol. ii. p. 145 (Jena, 1893). 



