1873.] MR. ST. G. MIVART ON THE LEMURS. 505 



principles upon which zoological classification should repose, and 

 the value to be assigned to the various kinds of anatomical resem- 

 blance. 



Until a recent date, zoological classifications reposed on similarities 

 of form and structure accepted simply and without reference to 

 genealogical considerations. 



Of late years, however, the theory of evolution (and especially 

 the Darwinian form of it) has complicated the inquiry by intro- 

 ducing the distinction between characters which may be reasonably 

 considered to be due to inheritance and others, called adaptive, 

 which may be supposed to have originated in necessary conformity 

 to the conditions of life. 



The doctrine has now been widely received that zoological classi- 

 fication should represent (as far as possible) the genealogical tree of 

 animal life, and therefore that it should repose, by preference, on 

 characters having a genetic significance, while adaptive characters 

 should be, as much as possible, eliminated. 



Four questions then naturally suggest themselves : — 



1 . Is it possible always, generally, or ever, to decide with certainty, 

 of any given set of characters, that some such characters are genetic 

 and certain others adaptive 1 



2. Is it possible now to class animals by genetic characters only? 

 and is no zoological classification to be considered satisfactory until 

 based upon such characters ? 



3. Is it desirable that animals should not be grouped together 

 into an order unless it can be supposed that they have all sprung 

 from a common ancestor, which was not also the ancestor of any 

 other group (of more or less similar size) belonging to the same 

 class ? 



4. Is it desirable that no group of animals which can be reason- 

 ably supposed so to have sprung, should be divided into two or more 

 orders ? 



As to the first question there seems to be great difficulty in 

 arriving at a satisfactory decision. 



It is true that the coexistence of a great many common characters, 

 such as, e. g., the course of the carotid arteries in all Marsupials and 

 the more or less aborted condition of certain of the digits of the pes 

 in many Marsupials, seem plainly to be due to community of de- 

 scent ; but many other structures cannot be due to such a cause, 

 and yet seem to be equally uncaused by the exigencies of life-pre- 

 servation or reproduction. As examples of these latter I may refer 

 to the osseous investment of the temporal fossa in Ghelonia, Pelo- 

 bates, and Lophiomys, the compound tooth-structure of Orycteropus 

 and Myliobatis, the coexistence of a certain form of dentition with 

 a saltatory habit in Macropus and Macroscelides, the presence of 

 but eight carpal bones in Troglodytes, Indris, and Lepilemur, and 

 the course of the vertebral artery in Auchenia and Myrmecophaga. 

 Thus characters may be due to no visible life-exigency, and yet not 

 genetic, while, on the other hand, characters may be thoroughly 

 genetic, and yet of great utility. 



