violets of the Atlantic Coast. 87 



PassiiiGj over for the present the consideration of the Linna^an 

 species, Viola pedata, which differs in root-structure from the other 

 members of the group, we shall find that V. palmata, also de- 

 scribed by Linnaeus, nui}^ be fairly regarded as the type of its 

 class, since it is the aggregate from which most of the remaining 

 species have been separated. With sagittata and possibly dentata, 

 V. ptahnata constitutes what we ma}^ call the heterophyllous type 

 of stemloss violets, or those in which the earliest leaves differ in 

 shape from the later appearing ones. In pahnata onl}^ the first 

 two or three leaves, which are cordate in outline and rather 

 small, are entire, the remainder being usually lobed to a greater 

 or less extent. In the majority of forms there are three main 

 divisions, of which the central one is the largest, the lateral lobes 

 being occasionally cut-toothed or still more deeply divided. The 

 general contour of the leaf is ovate or oblong, the length some- 

 what exceeding the breadth, the base never cucullate or inrolled 

 as in ohliqua, our common round-leaved violet. With a view to 

 ascertaining how closely these two species might approach each 

 other in leaf-forms, I set out several specimens in close proxim- 

 ity one fall. Tlie following summer the leaves of palmnta were 

 scarcely at all lobed, but they preserved their characteristic 

 outline, and were quite clearly distinguishable from the allied 

 species. Similar observations have been made b}^ others who 

 have had the plants under cultivation. But this is not the only 

 distinguishing character of V.palmata; it grows almost invari- 

 ably in rich, snaded woodlands, and, as Schweinitz has observed,* 

 never occurs in swamps or bogs, where ohliqua is most common. 

 Dr. Gray qnce reduced p(dinata to varietal rank in the fifth edi- 

 tion of the Manual,t but he afterward restored it to its former 

 ])lace,| a conclusion in which every other botanist of the century 

 has concurred. The species of Muhlenberg and Schweinitz here 

 referred to palinata are merely forms exliibiting slightly unusual 

 degrees of lobation. Le Conte's V. septemloha, however, belongs 

 to a different category. It is apparently confined to brackish 

 meadows along the coast from Staten Island to the Gulf States, 

 and I had always considered it a good illustration of varietal dif- 

 ferences induced by local influences, but on a recent excursion 

 with Dr. Britton to the home of the plant I became thoroughly 

 convinced as to its specific validity. The leaves are quite gla- 



*Ain. .Touni. Sci., 5, 54, 1822. tGray, Man. Ed., 5, 7S, 18G7. 



JCoult, Bot. Gaz., 11, 254, 1886. 



