172 University of California Publications in Geology [Vol. 10 



more like the broader skulls than are the three slender specimens. No. 

 644 belongs apparently to a younger individual than those having the 

 slender skulls, judging from the closure of the coronal and parieto- 

 squamosal sutures. These sutures are hardly, if at all, discernible 

 in the latter specimens. It may be possible that the sutures have 

 closed for other reasons than age alone. 



An objection to the separation of a new subspecies may be raised 

 by the apparent inconstancy in Mylodon harlani of the variations 

 given as characteristic of M. h. tenuiceps. No. 21160 Univ. Calif. 

 Coll. Palae. has been cited as an example in which a small palatal 

 index is associated with a broad cranium, while in Brown's specimen 

 from Nebraska a small cranial index is associated with a compara- 

 tively broad palate posterior to the fifth tooth. It is also worth noting 

 that in M. garmani, although the cranial index is very small, the 

 palate is very broad. In establishing the subspecies the writer has 

 been impressed by the lack of continuity in the gradations exhibited 

 by the Rancho La Brea series when the two characters under consid- 

 eration are taken in combination, rather than by the variability of 

 the characters taken separately. 



The writer would not hesitate to include the Nebraska skull de- 

 scribed by Brown in this subspecies, and thus render a new name 

 unnecessary, were it not for the following considerations : 



1. Tlie differences between M. h. tenuiceps and M. harlani con- 

 tained in the Museum of History, Science and Art of Los Angeles are 

 based on a comparison between individuals in which the coronal and 

 parieto-squamosal sutures have closed. In the Nebraska skull the 

 coronal suture is still distinctly to be seen. Another apparently char- 

 acteristic feature of a more advanced age, which is fairly constant in 

 the Rancho La Brea series, is the degree of pitting of the palate. In 

 the typical slender skulls this is distinctly more developed than in the 

 Nebraska specimen. In no. 695 the palate is too badly worn to indi- 

 cate this with certainty. 



2. In the three, possibly four skulls, which are considered to rep- 

 resent M. h. tenuiceps, the palate posterior to the fifth tooth is nar- 

 rowed decidedly. This portion of the palate in Brown's specimen is 

 more as in the broader types of skulls from Rancho La Brea. 



The amount of variation exhibited by the fourth inferior tooth 

 in specimens of the Museum of History, Science and Art of Los 

 Angeles support the previous contention that it is unsafe to base any 

 primary specific characters on this tooth. It must be admitted, how- 



