302 Annals of the South African Museum. 



From the structure of the manus it is quite manifest that it could 

 rest on the ground and support the weight of the body. In my 

 opinion the foot did not rest, as believed by von Huene, on the palm, 

 but was quite as digitigrade as the hind foot. My reasons for this 

 conclusion are that it is impossible to fit the metacarpals together 

 satisfactorily so that they could lie on a flat surface, whereas if they 

 are arranged in digitigrade fashion they fit accurately, the ends 

 resting on one plane. 



When the foot is placed on the ground the 1st large digit probably 

 bears none of the weight, though possibly the side of the toe may 

 touch the ground. The large claw curves inwards, and may also 

 touch the ground, though it bears none of the weight. The balls of 

 the 2nd, 3rd, 4th, and 5th toes, with probably a palmar pad and the 

 short 4th and 5th digits carry all the weight. The 2nd digit is 

 doubtless considerably extended, so that it makes an angle of about 

 80 with the metacarpal, and the claw is flexed so that it may just 

 touch the ground. The 3rd digit is less extended, and probably to a 

 considerable extent rests on the ground. The penultimate phalanx 

 is more extended and the short claw flexed. The 4th digit has no 

 claw and rests flat on the ground. The short 5th digit, which has a 

 stout proximal phalanx and probably a single rudimentary distal one, 

 also rests on the ground. 



When the animal is standing on its hind feet the digits are 

 probably more flexed, and when catching its prey all three claws 

 can be brought to work together by flexion, but for satisfactory 

 prehension it is necessary for the two hands to work together. 

 Probably the animal caught its prey with its hands and killed and 

 tore it up by blows from one or other of the hind feet. 



The femur is imperfect, all the portion above the lower trochanter 

 being lost. The lower half is considerably stouter in relation to its 

 length than in either Gryponyx, Nassospondylus carmatus, or Plateo- 

 saurus. It is, as preserved, somewhat distorted, but the measure- 

 ments are as follows : 



Distal end to lower end of trochanter 155 mm. 



Width of distal end as crushed 114 ,, 



Probable width of uncrushed 90 , , 



Greatest width of narrowest part of shaft 45 ,, 



The proximal end of the left tibia is preserved in good condition 

 It measures 102 mm. by 52 mm. 



