86 Pollard Violets of tJie Atlantic Coast. 



in other instances it is due to an over-conservative view of what 

 constitutes a species. It is a well-known fact in botany, and I 

 presume also in other branches of biology, that the species of 

 one genus differ inter se to a much less extent than those of an 

 other genus. In Lcchea, for example, we are forced to depend 

 almost solely on the appearance and structure of the radical 

 shoots springing up after the close of the flowering season, while 

 in the nearly allied genus Polygala, we have usually not only 

 well-marked floral characters, but habit and leaf arrangment to 

 guide us in making determinations. I believe that wholesale 

 reduction to a single species of a number of so-called polymor 

 phous types is a most unphilosophical and evasive method of 

 treatment and productive of immense difficulty to the critical 

 monographer. As an illustration of the simple solution pre 

 sented when one of these aggregate types is reduced to its com 

 ponent forms, I may refer to the two Eastern species of Sanicula, 

 which for many years were sources of despair to most botanists, 

 since they presented remarkable variability in habit and phyllo- 

 taxy. Mr. E. P. Bicknell, after an extended series of field ob 

 servations, discovered that there were altogether four very distinct 

 species confused under the two originals, affording not only con 

 stant characters with respect to habit of growth and geographical 

 range, but also in the fruit, which is of paramount importance 

 in the study of all Umbellifera3.* The same author has recently 

 shed light on the Eastern forms of Sisyrinchium, no satisfactory 

 disposal of which has heretofore been accomplished. f A similar 

 condition exists among the violets of the Atlantic coast, and, 

 while I by no means wish to imply that we can obtain an a.bso- 

 lutely correct systematic treatment of this or any other genus, I 

 do contend that it is possible to so arrange the species that any 

 given plant may be determined with comparative ease. The con 

 spectus of the group, which will be found at the close of this 

 paper, is merely tentative, and is offered simply as the outgrowth 

 of the field and herbarium study already referred to. 



In taking up the discussion of individual species I wish to 

 embrace the opportunity of extending thanks to Dr. N. L. Brit- 

 ton, of New York, for the loan of numerous specimens from the 

 Columbia University herbarium, and also to Messrs. H. W. Olds 

 and D. Leroy Topping, of Washington, for abundant field-notes 

 and living plants. 



*Bull. Torr. Club, 22, 351-361, 1895. flbid., 23, 130-137,189(5. 



