I24 THE AMERICAN NATURALIST. [VoL. XXXV. 
HUXLEY. DOLLo. 
Placentalia Marsupialia Placentalia Marsupialia 
| 
Terrestrial Arboreal Terrestrial Arboreal 
Metatheria Metatheria Eutheria Eutheria 
| / oS 
a Eutheria 
j (primitive Placentalia) 
/ | 
Metatheria Metatheria 
as an index of a placental origin of marsupials, the other as to its 
value as a factor of differentiation of marsupials and placentals. 
For the solution of the former we have not farto seek. We 
may at once accept Dollo’s statement that the ancestors of the 
Marsupialia were, in the possession of a modified foot, too 
specialized to have given rise to the Placentalia, because it is 
impossible to. make any other assumption. In doing so, how- 
ever, we do not in any way compromise Huxley’s view; for it 
must be remembered that Huxley derived the Placentalia, not 
from Marsupialia, but. from Metatheria. The prehensile foot 
he regarded as a special marsupial character. On the other 
hand, we cannot accept the evidence of arboreal ancestry as 
indicating a placental origin of marsupials. It scarcely needs 
to be observed that the assumption by non-placental Meta- 
theria of an arboreal habit would account for, and inas satisfac- 
tory a way, the arboreal modifications of the foot which Dollo 
has demonstrated throughout the existing forms. 
. The latter problem yields much more satisfactory results. 
It is on this account, in fact, that Dollo's results have been 
considered in detail If the former existence of an arboreal 
habit throughout the Marsupialia were susceptible of decisive 
proof,’ we would have a fairly reliable index of the cause of the 
2 ee Oe p evidence of arboreal ancestry as suggestive rather 
- Dollo has demonstrated conclusively that the Australian 
Marsupialia are of arboreal derivation, but a study of their characters, notably 
th * 
s Pp des y teeth and feet, convinces me that they have arisen by a rapid adap- 
iation from opossum-like ancestors. If such is the case, they are not at 
