440 Wortman — Studies of Eocene Mammalia in the 



limb is carried in the fore and aft plane of the body, the outer 

 edge of the sole of the foot touches the ground and the 

 plantar surfaces are directed more or less inwards. This sub- 

 ject will be again referred to on a subsequent page. The remain- 

 ing bones of the tarsus do not present any characters of 

 especial interest. They resemble the corresponding bones of the 

 fox with the exception of the internal cuneiform, which is 

 large and functional for the support of the hallux. It has 

 the same form as that of Herpestes but is proportionally broader 

 and heavier, which would indicate a less degree of reduction 

 for the hallux than in this species. 



There are only four metatarsals of the foot preserved, that 

 for the hallux being missing. When these bones are carefully 

 articulated, they show that the toes were more spreading than 

 those of Herpestes and somewhat less than those of the otter. 

 They entirely lack that highly compressed form and closely 

 interlocking arrangement seen in the fox. The distal ends are 

 characteristically broad, with hemispherical heads and well- 

 developed plantar keels. They resemble the corresponding 

 bones of Cynodictis very closely in these particulars, and differ 

 greatly from those of the fox, in which there is a distinctive 

 " square-cut" appearance. The third metatarsal is slightly the 

 longest and strongest bone of the series ; the fourth is a little 

 smaller and a trifle shorter ; the second and fifth were appar- 

 ently of about equal length, but the shaft of the second is con- 

 siderably the stouter of the two. Their proportional length 

 is much less than those of Cynodictis, in which the hind foot 

 had already begun to assume the more compressed, elongated 

 form of the modern dogs. 



If we take the length of the ischium from the posterior 

 border of the acetabulum, as a more or less fixed length and 

 compare the length of the metatarsals with it, the hind foot of 

 Vulpavus will be found to be as short as it is in Herpestes. 

 Compared with this measurement, the following proportions 

 of the different segments of the hind limb in several species 

 are given herewith : 



Length of Length LeDgth Length 



Ischium. of Mt. IV. of Tibia. of Femur. 



Herpestes J 00 117 274 282 



Vulpavus 100 117 314 326 



Cynodictis 100 154 368 363 



Vidpes 100 220 490 435 



That is to say, taking the ischial length as 100, the meta- 

 tarsal length in Herpestes is 117, the tibial length 274 and the 

 femoral length 282 per cent of the ischial length, etc. I have 

 been compelled to estimate the ischial length in Cynodictis, 

 which when correctly ascertained may give slightly different 

 percentages from those above recorded, but it may be assumed 



