NO. 1976. TREE8HREWS: FAMILY TUPAIIDJE— LYON. 11 



than it is in the Macroscelididas. The radius and ulna are well 

 developed and present as distinct bones, and they and the humerus 

 are siibequal in length. 



In Macroscelididae the ulna is intimately fused with the radius and 

 appears but little more than a process at the upper end of the radius. 

 The radius-ulna is much longer than the humerus. 



The carpus is composed of a scapho-lunar, cuneiform and pisiform 

 (well developed) proximally, the usual trapezium, trapezoid, magnum, 

 and unciform distally and a distinct os centrale between the two rows. 



Hind limb. — The femur is only slightly shorter than the tibia, in 

 Macroscelididse much shorter. The bone has a well marked head and 

 neck and promment greater, lesser, and third trochanters. 



The tibia is well developed, with a conspicuous ridge in front. The 

 fibula is slender, and perfectly distinct from the tibia, except at the 

 two extremities, where the two bones are in contact, the lower end of 

 the fibula having a distinct surface for articulation with the astra- 

 gulus. In Macroscelididas the fibula becomes fused with the tibia 

 slightly above the middle of the bone, and the anterior spine of the 

 tibia is much more pronounced than in Tupaiidse. 



The calcaneum is laterally compressed and narrow and relatively 

 short posteriorly, broad and rather flattened anteriorly. Its poste- 

 rior portion is relatively much shorter than m the Macroscelididae. 

 The trochlear surface of the astragulus is relatively wider and much 

 shallower in Tupaiidse than in Macroscelididse. The remaining 

 bones of the tarsus, cuboid, navicular, and the three cuneiform bones 

 do not show any special peculiarities in Tupaiidse. In Macrosceli- 

 didse they are all much elongated. 



The metatarsals are without special peculiarities. The second, 

 third, fourth, and fifth are all essentially subequal, but among them- 

 selves in order of length they stand third, fourth, second, fifth. The 

 first metatarsal is distinctly shorter than the others, but is by no 

 means a short bone like the first metatarsal of Macroscelididse, in 

 which family the entire first digit is shorter than the second, third, 

 fourth, and fifth metatarsals,- which are laterally compressed and 

 much elongated. In Tupaiidse at the base of the fifth metatarsal is a 

 large unciform process lacking in Macroscelididse. 



There does not seem to be cny essential differences between the 

 skeletons of Twpaia, Tana, and Vrogale. Urogale has relatively 

 heavier and thicker bones than has Tana, especially seen in those of 

 legs and feet, and has a higher and much better developed spine on 

 the axis than has either Tana or Twpaia. 



It is not probable that the skeletons of Dendrogale and Anathana 

 differ essentially from those of Twpaia or Tana. An examination of 

 a skeleton of Ptilocercus, however, would probably reveal differences 

 from the other genera, in correlation with the pronounced cranial and 



