THE BRITISH PANSIES. 



177 



sand-dune Pansy, which formerly I had suspected of being merely a 

 "state " directly induced by its environment. 



Further evidence was needed, however. It was necessary to cultivate 

 the plants under observation and control. I am greatly indebted to Mr. 

 W. Hales for his kindness and care in growing several forms from seed 

 supplied to him. Others have been grown in a garden at Chesterfield for 



Fig. 62.— Viola variata Jord. 



1, Flower ; 2, leaf from middle of stem ; 3, leaf from base of stem ; 4, stipule ; 

 5, form of stipule often found on upper leaves. (Natural size.) 



a number of generations. The result of the cultivation-experiments is 

 that although the plants may be more luxuriant than in the wild con- 

 dition, the characters upon which the species are founded remain perfectly 

 constant and unchanged. I think Mr. Claridge Druce, M.A., F.L.S., will 

 allow me to state that he has grown seeds of Viola obtusifolia Jord., 

 determined by me, with exactly similar results. Moreover, certain forms 

 believed to be hybrids have been found. They only occur here and there 



NiO /x . \ ■: i } ■ ! 



. % ■ XW t> * 



I / 



Fig. 63. — V. caxtiaxa Dbabhle. 



1, Flower ; 2, intermediate and upper leaves ; 3, lower leaf ; 4, stipule. 

 (Natural size.). 



singly or a few together, and only when accompanied by two distinct and 

 well recognized forms. These supposed hybrids are not themselves 

 referable to any of the named forms, and they show characters quite 

 unlike those found in the plants believed to be distinct species. It is 

 impossible to enter here into a discussion of these characters, but they 

 will be described fully elsewhere. The evidence for a hybrid origin is 

 very strong and supports materially the argument for distinctness of the 

 parent forms. 



