98 THE AMERICAN BOTANIST 



in it. If it is a summary of all knowledge, condensed into a 

 logical series of catch-words, then it is worth the devotion of 

 many lives. - '^' 



The fossil evidence on plant origins presents a picture 

 similar to that of the beans, but much more difficult. Had 

 we a complete record of all the plants that have ever lived 

 on the face of the earth, the evolutionist must believe that all 

 sharp boundaries between orders and families and genera 

 would melt away. Perhaps all species as we now know them 

 would also grade insensibly into other species. There would 

 remain at best such distinctions as between Jordan's Drabas or 

 de Vries's Oenotheras. Perhaps we should be reduced to such 

 gradations as those of the corn varieties bred in Illinois, or 

 of Castle's hooded rats. What then would become of classifi- 

 cation? Does that mean that the use of fossil evidence in 

 classification must lead us' more and more into a mist of un- 

 certainty? Does it mean that there will be no families and 

 orders and classes when our knowledge is complete, but only 

 one uninterrupted and insensibly graded family tree ? 



By no means. Could we only know the details of the 

 plant family tree, we could mark off the orders and families 

 in a phylogenetic manner with a precision now quite impossi- 

 ble. Each line of descent would form a "natural" taxonomic 

 group. And this, I believe, is the ideal for a synthetic syslen; 

 There will always remain, of course, the difficult)- of assign- 

 ing a place to a generalized form, a common parent of two or 

 more lines of decent. And' this is equally true whether we 

 think of a parent as one individual or a group of similar in- 

 dividuals. Perhaps we shall have to develop a new kind of 

 nomenclature for generalized forms, or a new nomenclature 

 altogether. Once the problem is clearly seen, its solution can 

 be confidently expected. 



