208 Bulletin American Museum of Natural History. [Vol. XXVII, 



Myrmecobius, the Peramelidce, Notoryctes. 



After a very careful examination of a large series of specimens Bensley 

 concluded (1903, pp. 99-107) that Myrmecobius is more probably an aber- 

 rant Dasyurid rather than a direct descendant of the Jurassic Trituberculata, 

 and that the high number of molars and their irregularly cuspidate character 

 were a result of degenerative processes accompanying the adoption of ant- 

 eating habits. 



The systematic position of the Polyprotodont Peramelidce has been a 

 matter of some doubt. The pes being strongly syndactylous and like that 

 of the Macropodidiie, the Peramelidjie have sometimes been placed with the 

 Diprotodonts under the term " Syndactyla," in contrast with the remaining 

 Polyprotodonts which are called " Diadactyla " and (earlier) "les Eleuthero- 

 dactyles" {cf. Gervais, 1836). But Thomas (1895) observed that although 

 Ccenolestes is evidently a Diprotodont, yet it is non-syndactylous; and on 

 the other hand, among the Polyprotodonts the pes of the Didelphid Marmosa 

 pusilla is incipiently syndactylous, while that of the allied Peramys brevi- 

 caudata is quite eleutherodactylous (Bensley, 1903). Hence, in the case 

 of the Peramelidte, Thomas concluded that the syndactyly was of less 

 taxonomic and phylogenetic value than the polyprotodonty. On the other 

 hand, in its mode of placentation Perameles is nearer to the Diprotodont 

 Phascolarctos than to the typical Polyprotodonts {cf. Hill, 1898). With 

 reo-ard to these and several other characters the Peramelidje are thus more 

 or less intermediate between the two suborders and tend to support the 

 theory that the Diprotodonts have been derived from very early Polypro- 

 todonts. 



The cheek teeth of the Peramelidje are derivable much more readily 

 from the Didelphid than from the Diprotodont type. The high crowned 

 molars present a peculiar modification of the tritubercular type. In Pera- 

 meles obesula the molars parallel those of the Insectivore Myogale (Gregory, 

 in Osborn, 1907, p. 113). In Thylacomys the main cusps are rounded and 

 the enlarged metacone is displaced inwards and occupies the position of a 

 hvpocone (Bensley, 1903). In their placental arrangements the Peramelidse 

 seem to have departed from the common type less than have any other 

 IMarsupials, since the allantoic placenta, though reduced in size, is still 

 functional, whereas in Dasyurus, Macropus, etc., it is vestigial and replaced 

 by the yolk-sack placenta (Hill). Quite possibly however the allantoic 

 placenta of Perameles may be merely a progressive character in which the 

 Peramelidoe parallel the Placentals (Bensley, 1903, p. 85). The pouch opens 

 backward as in Polyprotodonts. 



The Notoryctida? or ]\Iarsupial Moles are another Polyprotodont family 



