CONCLUSION. 177 



recorded. The time for such an exhaustive chissification has not yet 

 arrived, and all tluit can be safely ventured upon in the present state of 

 knowledge is to indicate the main groups, and their afKnities, and await 

 future discoveries. 



Excluding the aberrant aquatic Sirenians, now regarded as of ungulate 

 ancestry, and leaving out also Toxodon and other little known extinct 

 forms, the Ungulate Mammals may then be arranged in natural groups, 

 as follows : 



CLASS MAMMALIA. 



Sub-Class MONOUELPHIA. 



Super-Order Ungulata. 



(L) Order Hyracoidea. 

 (2.) Order Proboscidea. 



Dinocerata. 



(3.) Order Amblydactyla 



Coryphodontia. 



/I \ /^ 1 r\^• 1 i. 1 ^ Mesaxonia (Perissodactyla). 

 (4.) Order Chnodaetvla < ^^ . h • i i x 



( L araxonia (Artiodactyla). 



Before proceeding to discuss the relations of the Dinocerata to allied 

 forms, it is important to first consider the relative value of the charactei's 

 they share with these allies, and with groups stillniore remote. 



The characters found in existing mammals, and, to a great extent, in 

 the extinct forms from the 'JVrtiary to the present time, are clearly of two 

 kinds ; general characters, derived from ancestral forms, and special 

 characters, acquired in adaptation to their environment. Some of the 

 latter niay be negative characters, acquired by the disuse, or loss, of 

 parts once advantageous. 



The first series of characters are of most importance, as they indicate 

 a genetic connection, perhaps remote, with the different groups that share 

 them. Special characters, on the other hand, however closely they may 

 correspond in different groups, do not necessarily indicate affinities, but 

 may have been acquired by^ adaptation to peculiar surroundings, in groups 

 quite distinct from each other. 

 23 



