122 THE AMERICAN NATURALIST. [Vor. XXXV. 
Dollo accordingly finds in all of the Marsupialia examined a 
more or less extensive arboreal modification of the foot, and 
from this he draws the following conclusions : 
All of the Marsupialia were at one time arboreal. 
In the possession of a modified foot they are too specialized 
to have constituted the source of the Placentalia. 
They are, on the other hand, Aighly modified Placentalia, in 
which the placenta has been lost. 
They doubtless arose from very primitive Placentalia, in which 
an imperfect placentation resulted in frequent accidents, such as 
premature birth, and it was probably to avoid these accidents 
that the placenta disappeared. 
According to this conception there would be four types of 
development in the Mammalia, as follows: (1) An oviparous 
condition in the Monotremata, and formerly in the Prototheria. 
(2) A viviparous condition, zz which the placenta has not yet 
appeared; found in none of the existing forms, but formerly 
present in the Metatheria. (3) A viviparous condition, with 
placenta, in the Placentalia, and formerly in the Eutheria. 
(4) A viviparous condition, zs which the placenta has disap- 
peared, in the Marsupialia. 
II. Huxiey’s HYPOTHESIS IN THE LiGHT OF RECENT 
RESEARCHES. 
In order to define clearly the position of Dollo and other 
investigators on the question of the relationships of the Mam- 
malia, it is necessary to notice in some detail the view of 
Huxley as expressed in his famous paper of 1880, with which 
modern speculation on this subject may be said to have begun. 
Huxley recognized three primary and, in a sense, hypo- 
thetical divisions of the Mammalia, which he designated as 
Prototheria, Metatheria, and Eutheria, these groups being 
represented at the present day by the Monotremata, Marsu- 
pialia, and Placentalia, respectively. The existing Monotre- 
mata and Marsupialia he regarded as highly modified away 
from their original types, but the most generalized of the 
existing Placentalia (Insectivora) as only slightly so. He 
