712 



THE AMERICAN NATURALIST [Vol. XLVII 



basis as something more than mere collections of facts. 

 The purpose of the present paper is to set forth a few of 

 the more important of these evolutionary principles, with 

 their significance in the general process of evolution, and 

 to suggest a possible explanation for the fixation of char- 

 acter which shall be more satisfactory than that proposed 

 by the selection theory. 



In our search for such principles of conservatism, it is 

 primarily apparent that in the main those features which 

 are slow to change in one family are slow to change in 

 others also, and that consequently there are certain rather 

 definite categories of characters which throughout all 

 animals and plants show an inherent tendency to be con- 

 servative and slow to change, and others which are funda- 

 mentally plastic and variable. The conservative cate- 

 gories are, in general, those of number, relative position 

 and general plan, characters usually of little functional 

 significance; whereas the commonly variable features are 

 those of more importance for survival and include such 

 distinctions as size, shape, color and texture. The essen- 

 tial difference between these groups of categories is not 

 at all in their absolute degree of conservatism or plas- 

 ticity, but rather in their general tendency to become 

 fixed or to remain plastic. Number, position and plan 

 are not always constant, by any means, but they tend to 

 become so during the course of evolution, whereas size, 

 shape, color and other commonly variable characters are 

 almost always changeable and rarely become stereotyped. 



The conservatism of number is everywhere apparent. 

 The two great groups of radially symmetrical animals, 

 the coelenterates and the eehinoderms, are constructed 

 (with a few exceptions) on the plans of six and of five, 

 respectively. Insects, on the other hand, display almost 

 invariably a scheme of three or its multiples in the number 

 of body regions, segments, appendages and many other 

 structures. Among fishes the number of gills, of visceral 

 arches, of fins and of fin rays varies little throughout 

 large families ; and in the higher vertebrates, the number 



