CHAP. Ill] THE MAMMALIAN ORDER PRIMATES 39 



that the latter arise at the root of the limb common to the whole 

 Eutherian Section (or Infra-class). The Edentate Orders (Tubuli- 

 dentata and Xenarthra) are also placed close to that region 1 . But 

 the remainder are widely dispersed, and far removed from the root- 

 stock. Special note should be taken of the distance separating 

 the Artiodactyla and Perissodactyla, as well as that separating 

 either or both from the Proboscidea, for all were formerly included 

 in the heterogeneous assemblage called the Order Ungulata. 



Complicated as all this may seem, a general acquaintance with 

 the main outlines of the scheme is absolutely necessary before use 

 is made of such a simplified " linear " substitute as is given in 

 Fig. 16. 



A few remarks are needed in justification of the expression (which recurs 

 several times in the latter diagram) "stage of" such and such a "generalized" 

 animal. 



On the hypothesis of evolution, a given animal owns an ancestry including 

 forerunners which would be classed as Eeptiles possibly, or again as 

 Amphibia, or as Fishes. Moreover in each Class, a generalized ancestral 

 form is to be expected, i.e. a form which has not realized to the full the 

 possibilities of specialization within its Class. The word " stage " is placed 

 at the opening phase of each period. The ancestor at any stage would present 

 such a combination of morphological characters as would lead to its being 

 placed on the confines of two groups. 



This digression is necessitated by the protest (made with vigour by some 

 highly competent authorities 2 ) against reference to " generalized " forms. An 

 appeal in favour of specialized groups is substituted by those who thus object. 



In revising my diagram, I was careful to consider whether it were mis- 

 leading in view of that protest, the importance of which I recognize fully. 

 I decided that the diagram is still reliable for the following reasons. In the 

 first place, it does not exclude specialized forms from putting in an appear- 

 ance during the course of progress denoted by the line joining the successive 

 stages. Secondly, I think the "generalized" form must occur somewhere, 

 sooner or later, in the genealogical tree. For instance the Eeptile (without 

 prejudice as to its specialization or the reverse) is more generalized than the 

 Mammal, the Amphibian more generalized than the Reptile. If there be an 

 objection to the diagram, I think it is to be found in the rectilinear char- 

 acter of the line of ancestry as drawn. But the diagram is submitted for 



1 Judged by the single criterion of brain-formation, this judgment appears from 

 the researches of Elliot Smith (Linn. Trans, vn, Ser. 2) to need revision; for the 

 neopallial part of the brain in some Edentata is more highly developed than in 

 certain mammals commonly regarded as their superiors. 



2 Gaskell, Origin of Vertebrates, p. 497. 



