260 THE AMERICAN NATURALIST. (VoL. XXXV. 
in Australia. Concerning this he remarks as follows (p. 188, 
Summary): * The only way in which it seems possible to 
account at once for the presence of forms such as Prothy- 
lacinus in the Patagonian Tertiary beds, and the absence of 
any of the Didelphyidze in Australia, is to suppose that on the 
South-American side the connection between the Antarctic 
land and what is now Patagonia was lost at a time compara- 
tively soon after the early polyprotodonts had passed across, 
and during which the Didelphyidz were being developed per- 
haps in the more northern part of South America." 
In the writer's opinion, the difficulty of explaining the 
absence of the Didelphyidae from Australia is only an apparent 
one, which is due to the interpretation of this family as a 
modern derived group. Reasons have already been given for 
the elevation of the family to an ancestral position, and if the 
case is susceptible of proof, no other explanation is necessary 
to account for their absence from the Australian region than 
the assumption that they were formerly present and subse- 
quently disappeared. If we imagine the family to have origi- 
nally gained access to the region, it is not difficult to conceive 
that, in establishing the foundations of an extremely compre- 
hensive adaptive radiation, especially under the favorable con- 
ditions of absolute freedom from competition, they should 
have thrown aside their distinctive didelphyid characters. 
And more especially is this conceivable when we realize that 
the differences of structure separating the Didelphyidz from 
the most primitive of the Australian forms (insectivorous 
Dasyuridze) are very slight.! 
Concerning the direction from which the ancestral Didel- 
phyidz may have entered the Australian region, there is at 
least some justification for the view that it was from the north- 
ward. Lydekker (96, p. 57) has cailed attention to the fact 
that there are marked indications of a faunal affinity of North 
America and Asia, which points to an interchange of forms 
between the two continents. This suggests a connection of 
the Oligocene opossums of Europe with those of North 
America, and, as Lydekker has pointed out, there may have 
1 Cf. Thomas ('88, p. 31 5}. 
