n8 
Annals of the Transvaal Museum 
and 4 , p. 46) are much more curved than in our form. In Thecodontosaurus 
Browni the femur has a length of 24 cm. and the lower end of the trochanter 
quartus lies 11 cm. below the proximal end. In our form these measurements 
are 49*5 cm. and 21-5 cm. respectively, and it easily follows that the trochanter 
quartus lies relatively higher in our form than in Thecodontosaurus Browni „ 
The trochanter major lies also relatively higher in our form. 
Although our tibia is shorter than that of Gryponyx africanus, its head is 
absolutely broader. The relations of the head of the tibia in Gryponyx afri- 
canus are altogether different from those in our form. The shape of pubis and 
ischium is also quite different. 
A comparison of the ischium of the present form with the well-preserved 
one of Teratosaurus suevicus H. v. Meyer ( 5 ) shows that, where the last is 
twice as long as broad, the ischium of our form is more than three times as 
long as broad. Casually connected with this greater relative breadth is the 
fact that the articulation surfaces of the ischium of Teratosaurus suevicus 
make much greater angles with each other and with the shaft of the bone 
than in our form. In the present form the articulatio iliaca makes an angle 
of less than 25 0 with the long axis of the ischium. The articulatio pubica 
makes an angle of about 40° with the long axis of the ischium and of about 
6o° with the articulatio iliaca. In Teratosaurus suevicus these different angles 
are respectively 55 0 , 75 0 and 130°. 
If we compare the fibula of the present form with that of Plateosaurus 
Quenstedti, we are at once struck by the great difference between the two. 
The proximal end of our fibula has a large posterior process, while the pos- 
terior border of the upper end of the fibula of Plateosaurus Quenstedti is evenly 
rounded. The shaft in our form becomes a narrow rod in the middle, while 
in the fibula of Plateosaurus Quenstedti it is proximally very broad and distally 
only slightly narrower. The medial view of the distal end of the fibula of 
Plateosaurus Quenstedti does not show the medial ridge of our form. The 
anterior ridge of our form is broad distally, but it does not bear a smooth 
surface as in Plateosaurus Quenstedti. Our fibula also differs from Gresslyo- 
saurus cf. Plieningeri through its large proximal posterior elongation. 
It appears, therefore, that the present form is new and belongs to a new 
genus. The form is allied to Aetonyx and Massospondylus and is, therefore, 
an Anchisaurid. I propose to call this new form Dromicosaurus gracilis n. g. 
et sp. ( 8 pofui< 6 <; = quickly walking). The slender leg must have enabled the 
animal to go quicker than, for example, Eucnemesaurus and also implies a 
more slender form. 
Massospondylus Browni Seeley. 
The remains which are being described under this name have been dis- 
covered on the farm St Fort (Letsoanastad No. 528 of 1905 map of Bethlehem 
District) by its owner, Mr H. Walker, who very generously presented them 
to our institution. They consist of a vertebral column, front and hind legs, 
pectoral and pelvic girdle, all of one animal. The exact spot of the occurrence 
is in the Red Beds on the boundary between Clifton and St Fort immediately 
to the north of an old road through a Nek crossing this boundary, the Nek 
lying to the north of a hill which projects into the big bend of the Little 
Caledon River on St Fort. The spot lies about ten feet below the thick banks 
of the Cave Sandstone. A good search was made for the skull and a portion 
of the neck, which could not have been removed by weathering, but nothing 
further was found. 
