THE J5EAIN IX THE EDENTATA. . 391 



Order Tubultdentata. 



Family ] . Onjctcropodid(B. 



Order Squamata. 



Family 1. Manidce. 



Some confusion may arise from the use of tlie name Sq^mmata, which is also applied 

 to lizards and snakes. 



The question which naturally arises in the consideration of sucli a division into orders, 

 is whether the gain in pedantic accuracy is a sufficient justification for disturbing a well- 

 known group, which everyone miist now recognize as a mere conventional assemblage 

 with no genuine bonds of union, and th\;s adding to the number of orders. The view 

 which Oldfield Thomas takes concerning Oryeteropus — that "it has been placed there 

 [along with the American Edentates] rather on account of the inconvenience of 

 forming a special order for its recejition than because of any real relationship to them " * — ■ 

 is endorsed by Flower and many other competent authorities. If Orycteropus is more 

 nearly related to jllanis than either is to the Xenarthra, as is not at all improbable, 

 we must at present confess that we liave no evidence to warrant such a contention. 

 Such being the case, the adoption of the Nomarthra does not apj)ear to me to help 

 toward a clearer conception of tbis undoubtedly heterogeneous collection of mammals. 



It seems possible to adopt a compromise wbicb will avoid the disruption which 

 Thomas dreads, as well as the anomalous course of including within one order animals 

 which have nothing in common except certain functional modifications which are 

 obviously adaptations to the pecu.liar mode of life. 



In the light of our present knowledge, the group undoiibtedly ought to be divided 

 into three orders, as Mas Weber suggests. But the meaning of the name "Edentata " 

 is so well appreciated that it is not likely to fall into disuse, however much we might 

 wish for such a consummation. The frank recognition of this fact may save us from the 

 disturbing influence of new orders, if we retain the name " Edentata " on the distinct 

 understanding that it consists of a heterogeneous " group " of three " orders," which 

 have no necessary relationship the one to the other. 



The three existing families of Xenarthra are much more widely separated one from 

 another than is usual among mere " families," for they possess very little in common. 

 The application of the term " suborder " to these three groups might express the 

 distant relationship more effectively if it were not for the fact that, in view of the 

 existence of only one family in each suborder, the additional names would merely add 

 to the confusion t. On the whole it is desirable to adopt Max Weber's suggested 



* Oldfield Thomas, Proo. Eoy. Soc. vol. slvii. (1890) p. 248. 



t I have not discussed in this place the very marked distinctions, especially as far as the cerebral commissures 

 are concerned, which separate the three families of Xeaarthra. They have been sufficiently indicated in the body 

 of this memoir. I may, however, remark that the apparent simplicity of the commissures in the Sloths is not 

 necessarily an indication of a primitive condition, but may be merely a reversion to the archaic condition such as 

 I have found in the Bat Nyctopliihis (" The Origin of the Corpus, Callosum," this Vol., p. 47). 



