Radical Leaves of Parnassia and Valeriana. 153 
constitute another point of difference (Fig. 2). These are mentioned 
by Mrs. Arber, 1 who concludes that their presence gives little support 
to the supposed Saxifragaceous affinity of Parnassia and that the 
relationship of this genus with Hypericaceae is closer than was 
supposed by Drude. Mrs. Arber accepts Drude’s view that the 
genus should be placed in a separate family—Parnassiaceae. 
Fig. 2. Parnassia palustris. Epidermis of lower surface of leaf showing 
tannin cells (dotted). X about 200. From an unpublished drawing by Mrs. 
Arber. 
1 Agnes Arber, “ On the structure of the androecium in Parnassia and its 
bearing on the affinities of the genus.” Ann. of Bot. 27, July, 1913, p. 502. 
Mrs. Arber has kindly placed at my disposal an unpublished drawing (Fig. 2) 
showing the distribution of these tannin cells on the underside of the leaf. 
Cf. Mrs. Arber’s paper for references to the original description of these cells. 
Since this note was in type, I am reminded that the late 
Mr. Townsend, in his “ Flora of Hampshire,” asks “ Can it 
[ Parnassia ] have disappeared [from the New Forest], or did 
Mr. Mill [John Stuart Mill] visit the forest before it was in flower 
and mistake the leaves of Valeriana dioica for those of P. palustris ? 
I know an instance of such a mistake being made by an acute 
botanist.” [H.S. T.] 
