164 DB. E,. BROOM ON THE STRUCTUBE AND [June 4, 



on the same side the temporal arch and the lower jaw are missing, 

 but otherwise the skull is almost perfect. The skull is more 

 elongated than is usual in Udenodon, and differs from most 

 species in having the inter orbital region very considerably 

 wider than the parietal region. The eyes are moderately 

 small and directed more outwards than upwards. The nostrils 

 (n.) are placed far forwards and rather small. The nasal bones 

 are prominent immediately behind the nostrils. The caniniform 

 ridges are flat and slender and directed well forwards. The 

 suborbital arch is moderately round and rather feeble. The 

 frontal region is broad and moderately fiat, and is characterized 

 by a rather prominent median ridge (r.f.). Posteriorly, the 

 frontals are considerably wider than in front. The postfrontals, 

 where they join the frontals are flat and broad and to a considerable 

 extent roof over the orbits. Externally they are rather slender. 

 A distinct ridge runs from the posterior border of the postorbital 

 portion of the postfrontal bone inwards, then backwards along the 

 posterior part of the postfrontal. The parietal region is about 

 two-thirds the width of the frontal region, and is characterized 

 by the presence of two well-marked postfrontal ridges {r.jp.f.), 

 with an intervening depressed parietal portion. The squamosals 

 are large, and the anterior portions which form the temporal arches 

 are developed considerably, horizontally outwards. 



The second specimen (PI. XVI. fig. 1) referred to is an almost 

 perfect skeleton of a small Udenodon, and it is especially valuable 

 in that the bones are scarcely at all displaced. The specimen was 

 found in an impure and fairly hard shale. The skull had evidently 

 been long exposed, and is so badly weathered that very little now 

 remains of the bones of the upper side of the head. The post- 

 orbital arches are quite lost, though evidences of their positions 

 are given by the underlying matrix. The squamosal, so far as it is 

 displayed, agrees very closely with that in Udenodon gracilis, and 

 nothing in the other parts of the head seems to oppose this 

 determination. 



When the slab in which the skeleton lay was split, it w-as 

 found that this had been so arranged that the remains were 

 almost equally divided between the two sides. In the larger of 

 the two portions, which may be looked upon as the main slab, are 

 preserved the almosr. perfect left fore-limb, the impressions of a 

 number of vertebrse, a large number of ribs and impressions of ribs, 

 the sacrum and caudal region, the left ilium {il.), and the left hind- 

 limb, which unfortunately is twisted and has not been fully 

 displayed, and the right ischium (is.) and pubis {ph.). In the 

 counterpart slab is seen the head, the right fore-limb, almost all 

 the vertebrae and ribs, and the right ilium with the right hind-limb 

 extended and almost perfectly displayed. In the drawing (Plate 

 XVI.) the bones and impressions on the main slab are with the 

 skull figured and shaded in true relative position, while the bones 

 of the counterslab are in outline in proper relationship with those 

 of the main slab. 



