PROGRESS IN VERTEBRATE PALEONTOLOGY 449 



mammalian groups previously regarded by most naturalists as 

 widely sundered. This linking-up of the proboscideans with the 

 sea-cows is indeed most important, while the additional evidence 

 in favour of a connection between the zeuglodonts and the 

 creodont Carnivora is scarcely less so. The discoveries, as the 

 title of the volume indicates, are, however, by no means restricted 

 to the mammalian class ; and the identification in the Egyptian 

 Eocene of a giant tortoise related to the modern species of 

 the Mascarene Islands, of a southern type of freshwater 

 pleurodiran tortoise, of a gigantic snake, and, above all, of an 

 apparently ostrich-like bird, is of the very highest importance, 

 alike from a phylogenetic and from a distributional point of 

 view. 



As regards the geographical distribution of animals in past 

 times, the Eg}^ptian discoveries will indeed render it necessary 

 to recast many of the prevalent theories. On this occasion it 

 must suffice to mention that Dr. Andrews is in favour of the 

 view that Africa was connected by land with South America 

 during the late Cretaceous, and possibly even in the early 

 Tertiary period. It is added that if the existence of this land- 

 bridge be admitted, the Tertiary carnivores of Patagonia, gene- 

 rally known as sparassodonts, may be the descendants of the 

 Eocene creodonts of Africa. 



The suggestion conveyed in the last sentence at once brings us 

 face to face with a problem which looms large in the year's work. 

 Dr. Andrews, if he does not actually class them as creodonts, is 

 evidently of opinion that the Patagonian sparassodonts are very 

 closely allied to that group, from which, in his opinion, they are 

 derived. 



On the other hand, Mr. J. W. Sinclair, of Princeton University, 

 in an elaborate and sumptuously illustrated memoir — forming 

 part 3 of the fourth volume of Reports of the Princeton University 

 Expeditions to Patagonia, 1896-9 (Princeton, 1906) — definitely 

 classes these debatable carnivores as marsupials. Indeed, he 

 even goes so far as to place some of them (such as Prothylacinus 

 and Cladosictis) in the same family as the existing Tasmanian 

 thylacine. To discuss in detail the arguments in favour of this 

 view is obviously out of the question in the present article ; but 

 it may be mentioned that the sparassodonts lack the palatal 

 vacuities and likewise the epipubic bones (vestigial in the 

 thylacine) of modern marsupials, while, according to the obser- 



