626 DR. R. BROOM ON 



specimen to the British Museum, recognised that it was a near ally 

 of Ornithosuchus woodwardi, though about 2g times as large. 

 Mr. Boulenger discusses at some length the question whether the 

 specimen is merely an older example of 0. woodivardi or a new 

 species, and comes to the conclusion that there are no other 

 differences than those that might be accounted for by difference 

 of age. My comparison of the specimens has led me to con- 

 clude that the animals, though allied, are at least distinct 

 species, and I have therefore much pleasure in naming the form 

 after the discoverer. 



The specimen shows most of the skull. The jugal is probably 

 perfect but is partly hidden by matrix, and the quadrato-jvigal 

 is also nearly perfect. The squamosal and postorbital are much 

 crushed, but the side view of the back half of the skull can be 

 restored with moderate accuracy. Most of the vipper side of the 

 skull is preserved, but partly broken and not well displayed. A 

 fairly complete snout with maxilla and premaxilla of what may 

 perhaps be a second individual is also preserved. And as we have 

 the remains of the last maxillary tooth in the first specimen, and 

 the corresponding tooth in the snout-fragment, we can make a 

 fairly complete restoration of the skull. When this is done, the 

 very marked differences between it and the skull of Ornithosuchus 

 tvoodioardi are apparent. The arches are seen to be massive 

 instead of slender bars, and the snout relatively much more 

 powerful. The fact of the two animals being distinct is further 

 confirmed by the fact that the British Museum has recently 

 obtained from Elgin two new specimens — the one exactly 

 corresponding in size to the type of Ornithosuchus woodivardi, 

 and the other, so far as can be made out, agreeing in size 

 with the large form. In the new specimen of 0. taylori part 

 of the palate is preserved, and it agrees essentially with that of 

 0. ivoodwardi. Thei-e are two similar openings in the back part, 

 and the pterygoid sends forward a similar but relatively larger 

 anterior process, 



Hebpetosuchus granti Kewton, 



This small Pseudosuchian from Elgin is less satisfactorily pre- 

 served than Ornit/wsuchus woodwardi, but there is sufficient to 

 show that it is a fairly near ally. The skull differs considerably 

 in its proportions, but probably the essential structure is similar. 

 The palate is narrower, and the anterior process of the pterygoid 

 is relatively broader, and there is only a single opening in the 

 pterygoid region. The shoulder-girdle and anterior limb are 

 difterently proportioned, being much more slender, and indicate 

 that the habits of the two genera were different. An interesting 

 point revealed by this specimen is the fact that the manus has 

 five well-developed digits. In the carpus there are at least three 

 elements. The five metacarpals are well pi^eserved, but most of 

 the phalanges are missing. 



