MUTATIONS IN VIOLA. 187 



color of the corolla. This blueness of the corolla was one of 

 the facts which, taken along with the massed rootstocks, leaves 

 and flowers, convinced me that the plants could not be referred 

 to V. emarginata, although the locality was one in which that 

 species might be looked for. 



A year after having published it I transferred a clump of it 

 to my garden, and the one self sown seeding from it had corollas 

 almost purple, so that I have since then suspected the whole 

 thing of being a mutate, of which V. emarginata is the parent, 

 notwithstanding that the characters of it are good enough for a 

 proper species. 



V. ABERKANS, Greene, Proc. Philad. Acad, for 1903, p. 683. 

 This offspring of V. fimbriatula is the first violet observed by me 

 which at the very outset impressed me as indubitably a mutate. 

 In the midst of a colony of the true V. fimbriatula near Wash- 

 ington there grew one plant differing from all the others not 

 only by its cordate long-petioled leaves with no hint of the 

 dentation, but also by the fact that its caudex was multicipitous. 

 I transferred the plant to my garden. It flourished there for 

 three years. Seedlings from self-sown seed sprang up around it 

 the second year, others the third year, seven or eight of them 

 in all. Out of those one was a revert to V. fimbrtatula, a perfect 

 revert, without shadow of approach to its true parent, while the 

 other six or seven were as precisely true to the parent. I noted 

 at the time some characters of calyx and corolla, and wrote them 

 down, but the manuscript is lost. 



After all this, the violet came in to me, in the dry, from Mr. 

 Witmer Stone, for my opinion as to what it should be, and I 

 gave him my manuscript name for it, and my view of its origin. 

 He afterwards expressed his own — a zoologist's — opinion of it. 

 It is not now rare in the TJ. S. Herbarium, and has come in 

 from various localities ; and here, notwithstanding my own 

 exclusive right to the name, I find several sheets have been 

 labelled by Mr. House " V. aberrans (Stone) House " ! 



V. SECBDENS, Greene, Pitt. v. 131. At the place of publica- 

 tion this has been sufficiently indicated as, in my opinion, a 

 mutate ; bearing just that relation to V. subagittata of the West, 

 which V. aberrans bears to V. fimbriatula of the East. 



