494 Mr. R. T. Gunther on a 



seems slenderer, and the subangular process of the mandible 

 is more pronounced than in U. spelcBUS. 



It is when we come to the distinctions between ^nr jaw 

 and those of U. huiTibilis and U. arctos that difficulties of 

 finding constant taxonomic characters arise. Several Pleis- 

 tocene jaws in the British Museum agree with tlie Oxford 

 jaw in important features. Nearest to it comes the above- 

 mentioned jaw, No. 22029, from the Lower Pleistocene of 

 Grays, in Essex, labelled " U. horribilis, Ord.'' (PL II. fig. 4). 

 The development of the subangular process is identical in both 

 (PI. III. figs. 10 & 11). Similar, too, is the depth and rough 

 sculpturing of the fossa, in which the strong masseter muscle 

 was inserted. The Oxford jaw belonged to a more powerful 

 animal, which is indicated by the size of the condyle, 

 which is 6'5 mm. larger in diameter than in the Grays jaw. 

 The diameters are 24'5 mm. and 18 mm. (PI. III. figs. 7 & 8). 

 The Grays jaw measures 5 mm. less than the Oxford jaw 

 from the top of the condyle to the ventral side of the 

 angle. 



A second jaw, that must be taken into account by reason 

 of the very perfect preservation of the proximal end, is 

 M. 2507, from the Welsh submerged forest in Whittsand. 

 Bay, near St. Davids (PI. II. fig. I). This has been referred 

 to "" U. arctos.^^ Though otherwise imperfect, this jaw has 

 the coronoid process and condylar region complete, and of 

 similar proportions to those of the Oxford jaw, but the 

 subangular process is less pronounced (PI. III. fig. 9, s.). 

 The following measurements are striking : — 



n e ;i • Whitesand I'aw. 



Oxford jaw. (M. 2507). 



Condyle to inside of symphysis . . 227 mm. 227 mm. 



„ „ near side of ma 137 „ - 137 „ 



„ „ top of coronoid process. — 94 „ 



„ ,, ventral side of angle . . 65 „ 54 „ 



Depth of mandible between m2 and 



^3 61 „ 54 „ 



Although obviously from a slighter-built animal, we 

 suggest the proportions of this jaw as suital)le for the 

 restoration of the contour of the damaged coronoid of the 

 Magdalen College specimen (PI. II. figs. I & 2). The sub- 

 angular process is less strikingly developed. 



The jaw has also been compared with that of the type of 

 U, fossilis, Goldfuss, from Muggendorf, Franconia (Nov. 

 Act. Acad. Cses. Leop. Car. x. pt. ii. p. 259, pi. xx. c), the 

 measurements of which are given above in the table. This 

 evidently belonged to an animal that was smaller in all 



