157 



speak, comparable to DE CANDOLLE'S system 15 . 



Tims having completed our circuit of Mt. Fuji, let us consider the 

 beautiful vegetation of the mountain. However different be our starting- 

 point, after having completed our circuit and combined the different views, 

 the correct idea thus acquired of the mountain is always the same. So it is 

 with the natural system. No matter what system we take for a first con- 

 sideration, after having considered all relations in all views, the result should 

 be always the same. 



For the framework in constructing the dynamic system, I prefer 

 ENGLER'S system to others, as it is the one most widely used by systematizers. 

 For the present, I shall content myself with forming the system of the 

 Augiosperms, as that class is the one with which I am most familiar, 

 although I believe that systems for the other classes can be formed in the 

 same manner. Now, to construct a dynamic system, arrange the series and 

 families in the same order as in the system taken for a framework, and on 

 both sides of a series or a family, put into such order as you like the several 

 series or families respectively that are known to have some relations to the 

 middle series or family, bearing in mind that the relative positions between 

 the series or families thus arranged laterally and those in the middle vary 

 with criteria. Next, put many short lines on both sides of the same series 

 or families in the middle, a little more distant than the series or families 

 already placed, keeping in view the fact that the lines denote series or families 

 whose relations to the middle member are as yet unknown to us, though such 

 surely exist, according to the participation theory. In the full extent of the 

 latter theory, all the series and families, as many as are in the system, must 

 necessarily be related in equal or different degree according as we consider 

 the matter from the standpoint of universality or from that of particularity. 

 The true method, therefore, in the above system - construction, is to arraiige all 

 series or families other than the middle one on both sides of the latter. 



1) For the sake of convenience, I have here metaphorically compared different systems to 

 different views of the mountain. Yet, speaking more correctly, a static system such as 

 ENGLEE'S is something like a mosaic picture of the mountain, one part of which is taken 

 from one view, and another part of which is taken from another ; while the real natural system 

 is, as it were, comparable to the mountain itself. 



