GKEGORY — ON CiENOLESTES 
113 
tibia is very long, the femur being only 63.5 per cent as long as the 
tibia, i.e., nearly as long as in the saltatorial Macropus giganteus in 
which the femur falls to 57 per cent of the length of the tibia. This 
contrasts with Perameles nasuta in which the femur is 94.3 per cent of 
the length of the tibia, and still more with SarcophiluSj Trichosurus, 
Phascolomys and Phascolarctos^ in which the percentage rises rapidly 
from 100 to 134. 
Other indications of cursorial or partly saltatorial modes of locomo- 
tion may be found in the large size of the lumbar vertebrae, which have 
well-developed, forwardly-directed parapophyses, in the length of the 
tail and large size of the caudal centra and chevrons, in the fact that 
the ilium is nearly parallel with the sacral portion of the column as in 
the kangaroos. There is a distinct so-called center of motion located 
in an anticlinal vertebra, in the posterior dorsal region, as in Dasyuridse, 
Macropodidse and other terrestrial or leaping forms, and in contrast 
with the primitive arboreal Didelphidse in which there is practically 
no anticlinal vertebra. 
In brief, the evidence for cursorial and probably in part saltatorial 
habits seems fairly conclusive, but it is also barely possible that these 
small and very active animals run up the trunks and branches of trees. 
B, Heritage 
That Ccenolestes has been derived eventually from Didelphis-\ike 
arboreal forms, in accordance with the views of Dollo and Bensley with 
regard to marsupials as a whole, is evidenced by the occurrence of 
many primitive marsupial characters in the locomotor apparatus, 
such as are found in the Didelphidse in association with arboreal habits. 
Even the hands and feet retain evidences of ultimate derivation from 
a Didelphis-\ike prototype. In the pes perhaps the closest resemblances 
are to be found in Phascogahj but although the underlying pattern is 
similar, a significant difference is seen in the larger size of the volar pads 
in Ccenolestes, which is possibly associated with a scratching or scraping 
action of the feet. 
In the postcranial skeleton of Ccenolestes Osgood (p. 98) notes a 
number of peculiar or unique characters which perhaps collectively 
indicate a long separation from other families of marsupials. Even the 
resemblances to others are distributed in such a way and are accom- 
panied by so many differences that Ccenolestes cannot easily be made to 
fit in either with the Polyprotodontia or the Diprotodontia. On the 
whole, the postcranial skeleton suggests a nearer relationship with 
JOUENAL OP MAMMALOGY, VOL. 3, KO. 2 
