780 PROF. BALDWIN SPENCER, ON A [NOV. 20, 



to the straightness and to a certain extent the length of the 

 squamosal suture, the fossil resembles Phascolomys and Phascolarctos, 

 from both of which it differs, however, to a great extent in other 

 features. The squamosal is indeed larger than in any existing 

 or fossil marsupial yet known, and its forward extension completely 

 preveuts the alisphenoid from coming into contact with the parietal 

 as it does in most marsupials, such as Phalawjer, Trichosvras, 

 Pseudochinis, and Dasyurus ; though, on the other hand, the parietal 

 does not reach the alisphenoid in Peragale, sEpyprymnus, and 

 Phascolomys, with which in other respects the fossil has but little 

 in common. 



No intertemporal bone can be detected. 



The zygomatic arch is strong and massive, most resembling in 

 form that of the Dasyuridse both in the considerable outward sweep 

 of the hinder part and in its marked upward curvature; in the 

 former respect it most nearly resembles a large skull of a Dasyurus, 

 and in the latter that of a Sarcophilus. There is not a trace of the 

 swollen posterior part of the squamosal which is such a striking 

 feature in certain of the Phalangeridaa, such as the genera 

 Phalanyer, Trichosvrus, Pseudochirus, and Petaurvs ; but, in strong 

 contrast to what is characteristic of these forms, the arch arises low 

 down, and, as is well seen in lateral view (fig. 5, PI. L.), the upper 

 surface rises immediately so as to form a strongly marked curve : 

 at the same time the arch flattens out. laterally, bends over (tig. 1 ) 

 towards the upper surface and runs forwards to be inserted into 

 the notch in the malar process of the maxilla. The latter process 

 is well developed and passes off almost at right angles to the side 

 of the maxilla, bending abruptly backwards and at the same time 

 outwards to be attached in the characteristic way to the hinder end 

 of the arch, beneath which it runs backwards to the glenoid cavity. 

 The abrupt bend of the malar process most resembles that of 

 Phascolarctos, though in the latter the main body of the arch runs 

 parallel to the length of the skull. In the Phalangeridae the bend 

 is a more gradual one, and the sweep of the arch is not nearly so 

 pronounced as it is in this fossil form. Though the lower part of 

 the arch is broken away, it appears probable that there was a slight 

 zygomatic process at the anterior end ; and, as in Trichosurus and 

 certain other genera, only still more markedly, there is a very 

 distinct ridge (PI. L. fig. 5), which traverses almost the length of 

 the arch, and sharply marks off an outer and upper from an outer 

 and lower surface, the latter serving for the attachment of the 

 masseter muscle. 



The glenoid cavity is considerably elongated transversely, the 

 downward process which bounds it posteriorly beiug, as in the 

 Dasyuridye and Perameles, apparently independent of any structure 

 concerned with the auditoxw passage. The shape and relationships 

 of this process are most similar to those of Thylacinus, and differ 

 markedly from those which obtain in the Phalangeriduj, amongst 

 the members of which the process in question forms the anterior 

 part of a well-marked bony canal. Also, as in Dasyurus, the 



