II.] PALEONTOLOGY OF MARSUPIALS. 51 



nental Arctogaea after the Secondary epoch, a very serious error 

 might be committed. And although it is improbable that any 

 marsupials of an Australian type ever existed in Europe or North 

 America, there is no reason why they should not have occurred 

 in south-eastern Asia. 



The extinct dasyurids of the Patagonian Miocene have been 

 already mentioned, and these, together with another S. American 

 group, are more fully noticed in the next chapter. With regard 

 to the opossums, it will suffice to state that while they are 

 unknown in the aforesaid Patagonian deposits, certain species 

 occur in the middle Oligocene White River beds of the United 

 States, and others in the lower, middle, and upper Oligocene 1 beds 

 of Europe. Although the number of their incisor teeth is unknown, 

 there is little doubt that the European Oligocene opossums 2 (which 

 have been very generally separated as Pcrathtrium\ should 

 be included in the existing genus Didelphys. Remains of 

 these animals have been obtained from the upper Oligocene of 

 Cournon in France, from the middle Oligocene beds of Hordwell 

 in Hampshire, from the equivalent deposits of Debruge in 

 Vaucluse, and of Montmartre near Paris, and likewise from 

 the Quercy Phosphorites in the south of France. With the 

 exception of a peculiar South American group of diprotodonts, 

 the remaining fossil mammals which can be referred to the 

 Marsupialia are mainly if not exclusively confined to the Second- 

 ary period ; all being of small dimensions, and many of them 

 exceedingly minute. While many of them evidently died out 

 without leaving any existing descendants, one group seems to 

 have been the ancestral type from which the existing Dasyurida 

 have originated. Among the former, we have the family Tricono- 

 dontidce, as represented by the genus Triconodon of the upper 

 Jurassic of England and also by nearly allied forms from the 

 corresponding rocks of the United States. In this family the 



1 It may be well to mention that the beds of St Gerand-le-Puy, in France, 

 which many writers reckon as lower Miocene, are here classed as upper Oligo- 

 cene. See Lydekker, Cat. Foss. Mamm. Brit. Mus. Pt. ,iv. p. xvii. 



2 The existing Didelphyida differ from the Dasyurida in the presence of four, 

 in place of three, pairs of incisor teeth in the lower jaw, and of five pairs in the 

 upper jaw instead of four. 



42 



