154 SMITHSONIAN MISCELLANEOUS COLLECTIONS VOL. 54 



It is to be observed that he does not treat first of herbs consecu- 

 tively, then of shrubs, then of trees. That would have been the 

 formal adopting of a piece of regular taxonomy even in his day, 

 as we have seen, already long established. There are merely 

 spicy things that are trees, others that are shrubs, and some that 

 are herbaceous; and the same is true of things alimentary and of 

 I things medicinal. This qualitative classifying is better suited to 

 . his purpose. But that even the more strictly botanical taxonomy 

 may now and then gain a point by ignoring herb, shrub, and tree as 

 fundamental divisions, comes out interestingly in Dioscorides' 

 chapter on Sambucus. ' He has two species, one of which is only 

 an herb, the other woody and almost a tree. We shall see later 

 that, after the revival of botany, at a comparatively modem 

 period, such a hold had been gained by the old distinction between 

 the herbaceous and woody that in deference to it systematists 

 almost with one accord divided Dioscorides' Sambucus into two 

 genera and separated them widely, Sambucus being located among 

 the trees, and Ebulus among the herbs; and that hardly after 

 Bauhin as late as 16232 had followed Dioscorides in writing them 

 as one genus was the botanical world of that time ready to accede 

 to a proposition so subversive of what was deemed fundamental in 

 taxonomy, i.e., that a tree and an herb could not be congeneric. 

 Subordinately to the more general and qualitatively outlined 

 divisions of the work, Dioscorides recognizes all the more familiar 

 natural families of plants; that is to say, within each Book there 

 is a line of labiate genera, another of the leguminous, another of the 

 umbelliferous, and the succession of cognates is not often interrupted 

 by the intrusion of a genus not of such natural alliance. Even 

 among the composites Anthemis, Parthenium, and Cotula are in 

 close conjunction on his pages, as are Anchusa, Lycopsis, and 

 Echium; and the succession of the representatives of five or six 

 genera of solanaceous plants is only broken by the intrusion of 

 Cardiospermum. Examples of this need not be multiplied. It 

 is propagating fable in place of history to affirm that natural families 

 were first recognized and indicated by any Linnaeus, or Adanson, 

 or Jussieu of the eighteenth century. 



The whole subject of Dioscorides as a taxonomist merits a fuller 

 development. A thorough study of his text might show that 

 classification had progressed somewhat during those three centuries 

 that had then elapsed since Theophrastus. 



» Diosc, Book iv, ch. 155. 



' C. Bauhin, Pinax, pp. 455, 456. 



