190 TEN YEARS^ PROGRESS IN VERTEBRATE PALEONTOLOGY 



Although Borhycena has lost so many marsupial characteristics, its 

 allies, Cladosictis and Amphiproviverra, retain many extremely impor- 

 tant ones. The alisphenoid bulla, the form and relations of the malar, 

 the basicranial region, the dental formula, the strongly inflected angle 

 of the jaw, the separate intercentrum of the atlas, the seventh cervical 

 pierced by the vertebrarterial canal, the form of the astragalus, the small 

 size of the lunar — all these and many others noted by Sinclair betoken 

 Marsupial, not Creodont, afiinities. 



In view of the many important agreements shown by the Sparasso- 

 donts with Thy I acinus, including a peculiar displacement of the ecto- 

 cuneiform beneath the cuboid, Sinclair referred the group to the family 

 Thylacinidse ; but it has been suggested by Matthew that this does not 

 necessarily imply that the Sparassodonts are either descended from or 

 ancestral to their Tasmanian allies, but that the Borhyaenidae and the 

 true Thylacinidse may represent closely related, but still parallel, lines, 

 independently derived from early northern didelphoids. This hypoth- 

 esis, especially on account of its paleogeographical implications, is 

 worthy of careful examination. 



In the same memoir Dr. Sinclair described the remains of a number 

 of Epanorthids or C^enolestoids, figured the skull of Ccenolestes, the sole 

 surviving member of this family, and showed that the Santa Cruz Hal- 

 marhiphus appears to be ancestral to Ccenolestes and also occasionally 

 retains the full didelphid tooth formula. Studies of the skull of Cceno- 

 lestes by Miss Dederer^^ and the present writer^^ indicate that we have 

 here to do not with a true Diprotodont, but with the representative of a 

 distinct suborder, the Paucituberculata of Ameghino, derived from Poly- 

 protodonts peculiar to South America and paralling in dentition the 

 smaller Australian phalangers.* 



MULTITUBERCULATA 



To our fortunate colleague, Mr. Gidley, has been accorded the privi- 

 lege of describing f a well preserved skull associated with the lower jaw, 

 some vertebrae and limb bones, of Ptiloihis, a basal Eocene survivor of 

 the Mesozoic Plagiaulacidae, a family hitherto known only from disas- 

 sociated jaws and teeth, and about whose affinities Owen, Falconer, Cope, 

 and many others had disputed almost in vain. At first glance few zoolo- 



" American Naturalist, vol. xliii. 1009, pp. 614-618. 



13 The Orders of Mammals, p. 211. 



* Broom, in a recent paper ''On the Affinities of Csenolestes" (Free. Linn. Soc. N. S. 

 Wales, 1011, vol. xxxvi. part 2, pp. 315-320), concludes that Cwnolestes is a true 

 polyprotodont, not deserving even subordinal separation. 



t Proc. U. S. Nat. Mus., vol. xxxvi, 1900, pp. 611-626, with plate 70. 



