182 EDENTATA. 
founded on individual variations. Thus Gray laid great weight on certain cranial 
characters, especially on the form of the angle of the mandible, which he found to be deep 
and truncated in some specimens, and slender and produced in others*. ‘This he regarded 
as a constant character of specific value, observing that he had never seen any lower 
jaws which seemed to him to pass by intermediate gradations from one of these forms 
to another. A comparison of Gray’s types and of other specimens leads me to a very 
different conclusion. As was well remarked by the late Mr. H. J. Turner, “ the skulls 
of the Three-toed Sloths vary greatly, and all present a coarse, rough-hewn appearance 
which must detract from our confidence in little differences in detail”. The form of 
the mandibular angle, instead of being constant, differs in almost every specimen 
examined; and the lower jaws of the types of Gray’s Arctopithecus marmoratus, 
A. boliviensis, A. problematicus, and A. blainvillet appear to me to form an inter- 
grading series in which it is impossible to find any tangible characters. 
In 1871 Gray described two new species of Sloths from Central America, under 
the names of Arctopithecus castaneiceps and A. griseus. The former is still only known 
by the type specimen, not in very good condition, in the British Museum; and I am 
led provisionally to accept its specific distinction on account of the peculiar character of 
the hair of its face, which rather resembles that of the very distinct bradypus torquatus 
than that of any of the more nearly allied forms. As far as can be judged from the 
stuffed skin, the facial fur of B. castaneiceps is nowhere fine and closely lying as in 
B. tridactylus and its allies, or eregt and velvety as in B. cuculliger, but is crisp 
and recurved like that of the reminder of the body. In the skull the mandibular 
angle is considerably produced. 
Arctopithecus griseus, on the other hand, I cannot distinguish from the well-known 
Bradypus infuscatus of Wagler, if, indeed, the latter is any thing more than a northern 
race of B. tridactylust. Thus I would provisionally recognize two species of Central- 
American Three-toed Sloths, which may be thus characterized :— 
1. B. infuscatus. Fur of face fine, silky, and close-lying. Colour grey, varying 
in depth, and more or less mottled with white. Saddle-mark of male orange, 
mottled with black. Length of head and body eighteen to twenty inches, 
2. B. castaneiceps. Fur of face crisp, coarse, and recurved, like that of the rest 
of the body. Colour grey, strongly tinged with rufous about the head. 
Saddle-mark orange, mottled with black. Length about eighteen inches. 
* P.Z.S. 1849, pp. 65-73, pls. Mamm. x. & xi.; P.Z.8. 1871, pp. 428-449, pls. xxxv.—xxxvil. 
+ P.Z.8. 1851, p. 209. 
+ On this point I have not the materials for any certain conclusion. I may observe, however, that the only 
constant difference which I have been able to find between the two forms is that the saddle-mark of the male 
is black and white in B. tridactylus, and black and orange in B. infuscatus. All the other characters which 
have been described appear to be variable; and I may further remark that of two specimens from Para in the 
Berlin Museum one has the saddle-mark very pale, while it is deep orange in the other. 
