14 PROBABLE NEW SPECIES OF DENDROLAGUS. 
so widely different to those of the burrowing and nest- 
making kangaroo-rats. At present we are unable to 
point to any intermediate stepping stone from one to 
another serving to show how the passage was ac- 
complished, but on the other hand there is no inherent 
improbability in the kangaroo-rats of a dense scrub country 
taking to the trees upon which alone they could in such 
circumstances find subsistence. It is far more difficult to 
explain on the opposite hypothesis why the tree-kangaroo 
should have been created with the teeth of the kangaroo 
rats of the plains, rather than with those of the native bear 
or opossum of the trees, or with an entirely different type 
of dentition. The explanation offered by the doctrine of 
natural selection is easy and rational. Those parts of the 
structure which required modification for a different mode 
of life—namely, the limbs—were so modified: those that 
were not so modified—the teeth—remained unchanged, 
because no great change in them was necessary. 
The discovery of these peculiar wallabies, if we may so 
call them, in Queensland strengthens the opinion formed on 
other grounds as to the direction of dispersion of the 
Australasian marsupials generally, viz., that it was from 
South to North. Australia proper is even now their head- 
quarters, but it is that of a seven times decimated army. 
The remains of the kangaroo tribe buried in the Darling 
Downs drift represent probably more species in that single 
river valley than are now living in all the continent. It is, 
however, curious that Dendrolagus has not been found 
among them as yet, possibly it may be a late development 
adapted to the present conditions of our coast and of New 
Guinea. Should it, however, eventually be found fossil, as 
most probably it will be, it will serve to show how large 
a portion of the land was then covered with jungle, similar 
to those small strips to which the surviving forms are now 
limited. 
