A452 DR. RB. BROOM ON THE EARLY CONDITION OF THE 
passing forward, terminates in a sharp point. The “prescapula” 
is well developed, and here, as in Dasywrus, a large part of it has 
no pre-existing cartilaginous basis. In the figure a dotted line 
indicates the anterior border of the cartilaginous scapula. 
The coracoid (co.) is small, and is directed mainly forwards 
from its attachment with the scapula. 
The clavicle (¢/.) is rudimentary, and has no connection, except 
perhaps by ligament, with either the acromion or the sternum. 
The rudimentary clavicle of the Bandicoot was, I believe, first 
discovered by Hill, who records its occurrence in his second 
paper on the Placentation of Perameles (4). 
Comparative Observations. 
While the shoulder-girdle of Dasyurus in its early stage agrees 
in the main with that of the Diprotodonts, it shows one or two 
interesting points of difference. In Pseudochirus we find the 
coracoid forming with the sternum a distinct joint, as in Dasyurus ; 
but in Zrichosurus and Petrogale the cartilage of the coracoid is 
structurally continuous with that of the sternum. And while in 
Dasyurus, as in Pseudochirus, the shoulder-blade becomes free 
by the direct separation of the coracoid from the sternum, in 
Trichosurus, at least, the separation is brought about by the 
degeneration of a portion of the coracoid. In the Diprotodonts, 
in the early stage, the coracoid from its attachment with the 
scapula passes mainly inwards and backwards; in Dasyurus it 
passes inwards and slightly forwards. It seems probable that 
though the coracoid in Dasywrus is less developed than in some 
of the Diprotodonts, it is in a much less specialized condition, 
more nearly approximate to that of the primitive marsupials. 
If we compare the early Dasyurus girdle with that of the 
Anomodont (fig. 4), the most striking difference is seen to be 
due to the great reduction of the coracoidal portion of the girdle. 
In the Anomodont, and almost certainly in the early Theriodont 
from which Dasyurus is descended, we find the scapula fixed by 
a powerful coracoid (co.) and precoracoid (p.co.). In Dasywrus 
the space between the first rib and the clavicle is very short, but 
is almost fully occupied by the cartilage which supports the 
scapula. Whether this cartilage represents the coracoid or the 
precoracoid of the ancestor, or both, is a question which might 
be argued at some length. In the paper above referred to (2) I 
advanced some reasons for believing that it represents the 
