125 



quently, groups of plants whose characteristics mostly occur in correlation, are less 

 changeable ; while groups of plants whose characteristics usually occur indepen- 

 dently are more changeable. Such families as the Orchidace?e and Graminese 

 belong to the former, while such as the Logauiacere and Myoporacese approach 

 the latter class. 



What he state in 21 and 22 is, in brief, that on one hand he regards 

 characteristics in linkage - relation as those denoting the serial orders of advance- 

 ment and gives them a high systematic importance, but on the other hand 

 he puts no value on the characteristics which occur independently and without 

 regard to others. But, in this I cannot concur. In my opinion, we ought 

 to take into consideration for classification all characters without being partial 

 to any of them. 



The perianth characters to which he refers in 23 are the most important, 

 as far as present systematic botany is concerned. But when we ask why 

 they are so appreciated, no reason is forthcoming. There is nothing but con- 

 vention to support the practice, as I have already explained. 



In 24 he says : 



Hit Ansnalmie sehr weniger Fiille liegt bei den Angiospermen klar zutage, daG 

 die Formen mit Zwitterbliiten phylogenetisch alter sind, als die sonst sick gleich 

 verkaltenden mit eingeschleclitliclien Bliiten. Diese Progression tritt tuiendlick 

 oft ein nnd ist znr Gnrppenbildung katun zu verwerten. 



But why is it that tho characters of bisexual or unisexual flowers that show, 

 as he expresses it, a serial advancement can not be appreciated as having 

 high systematic value ? This is, in my opinion, only because of insisting upon 

 a static system. According to the dynamic view of a system, there can be 

 no character that can not be so appreciated. 



In 25 and 26 2) , he refers to many characters showing serial progression. 

 He puts systematic value on some of these, but not on some others. But 

 what is the reason of such partial treatment of characters? It is, as far as 

 I can judge, merely because of an unquestioning acceptance of the present 

 static system, which would of course at once be \iolated by the fair apprecia- 

 tion of aU characters. 



1) ENOLEB, A. .1. c. p. XVL 2) ENGLKK, A. 1. c. p. XVIII. 



