1910.] The Ccenolestoids, Polydolops. 211 



from the type represented in Microbiotherium tortor (Sinclair, 1901, pi. Ixii), 

 which, as in the Ctenolestoids, has a large rounded metaconule in the 

 position of a hypocone. The lower molars of the Ccenolestoids are a modi- 

 fication of the tuberculo-sectorial type, the trigonid and talonid being very 

 clearly defined. In the presumably most primitive genus Halmarhiphus 

 the normal cusps of the tuberculo-sectorial crown are very clearly definetl 

 (Sinclair, /. c, p. 420). All the molars in Coenolestes and Halmarhiphus 

 have a prominent antero-external cingulum as in Didelphis. Halmarhiphus 

 and Coenolestes occasionally retain the antemolar dental formula of Didelphis 

 (Sinclair). 



The fourth upper premolar and first lower molar among the Csenolestoids 

 vary greatly in character. In the less specialized genera Coenolestes, Hal- 

 marhiphus, and Garzonia they are simple in form; in the larger form Abde- 

 rites the first lower molar has a high piercing sectorial blade, fluted anteriorly 

 and analogous to that in the Plagiaulacidae. Sinclair has adduced evidence 

 (/. c, pp. 417-418) to show that this sectorial molar has been derived from 

 the tuberculo-sectorial type of Halmarhiphus. 



Accordingly there is seen to be considerable though not positive evidence 

 that the Ccenolestoids have been derived from the Didelphoid type, and this 

 is strengthened by the similarities pointed out by Miss Dederer (1909) 

 between the skull of C.ienolestes and those of the smaller Dasvuridse. On 

 the other hand, in certain characters of the dentition they are prototypal to 

 the Diprotodont type and especially the smaller phalange rs. 



The pes, however, as stated above, is entirely eleutherodactylous and 

 shows no trace of the syndactyly so characteristic of the Diprotodontia. 

 This may be an adaptation to terrestrial and ciu\sorial habits as in the 

 smaller Dasyui'es; but the detailed characters of the skull show no striking 

 Diprotodont characters and the writer is inclined to regard Coenolestes and 

 its allies as an independent suborder, an offshoot of primitive Polyprotodonts, 

 which has paralleled the Diprotodonts in certain characters of the dentition. 



The so called " Multituberculates" of Patagonia. Are they not highhj m.odi- 



fied C<enolestoids? 



The most serious theoretical difficulty in the derivation of the Csenolestoid 

 dentition from the tritubercular and tuberculo-sectorial type lies in the fact 

 that, according to Ameghino's figures (1903, pp. 31-192), the forerunners 

 of the Caenolestoidea in the far older Notostyloi)s Beds ( ? Basal Eocene), 

 have molar teeth which are of the multituberculate rather than of the 

 tuberculo-sectorial type. It may be well then to inquire whether the genera 

 Polydolops, Pliodolops, etc., are related to the Csenolestoids, or, as Ameghino 

 believes, to the ^Nlidtituberculates. 



