III.] MARSUPIALS. 109 



there are four pairs of upper and three of lower incisor teeth in 

 Prothylacinus, and the same is the case with the smaller Santa 

 Crucian form described as Amphiproviverra, which appears to be 

 of a distinctly dasyurid type, although not coming very near to 

 any Australian genus. v 



With regard to the Microbiotheriida, as typified by the genus 

 Microbiotherium, these, although they are not included by Dr 

 Ameghino in the order, appear to be undoubted marsupials, since 

 they have a dentition numerically the same as that of the opossums, 

 vacuities in the palate, and an inflected angle to the lower jaw. 

 From the opossums they differ by the non-production of the palate 

 behind the last molars, and in the form of the lower jaw, in which 

 the extremity is produced to a greater extent in advance of the 

 canine. In all these respects they approximate to the Dasyurid 

 genus Phascologale, from which they differ in having one pair less 

 of incisors in each jaw. The ancestors of the Australian Dasyuridce. 

 must, however, have originally had five pairs of upper and four of 

 lower incisor teeth, as the former are retained in many of the ban- 

 dicoots (Peramelida), while Myrmecobms occasionally develops four 

 pairs of these teeth in the lower jaw. It seems therefore probable 

 that the Microbiotheriidcz were minute polyprotodont marsupials 

 of an Australian type. 



There is more difficulty in arriving at any satisfactory con- 

 clusions as to the serial position of certain carnivorous mammals 

 from the Santa Cruz beds, of which a large form described as 

 Borhycena may be taken as an example. In these animals the 

 dentition approximates to a certain extent to that of the primitive 

 or creodont Carnivora of the earlier Tertiaries of the northern 

 hemisphere, although retaining the marsupial feature of four pairs 

 of molars and only three of premolars. The replacement of the 

 teeth is also fuller than in the marsupials. Dr Ameghino has 

 suggested that these animals were transitional between marsupial 

 and eutherian carnivores, and that the latter group originated in 

 South America ; but this idea is obviously untenable. A possible 

 suggestion is that they may be specialised offshoots from the 

 marsupial stock which died out without giving origin to any 

 descendants. 



A small mouse-like mammal first described in 1863 upon the 



