BRITISH VIOLETS 25 



V. stagnina, which is an exquisite and delicately beautiful violet ; 

 and I rather suspect that it is either a primary or (if hybrids of 

 the fen-species are fertile) a secondary cross, stagnina being the 

 subordinate parent. 



With regard to the intermediates between V. hirta and 

 V. odorata (V. sepincola Jord., V. permixta Jord., &c), it may be 

 argued, first, that such polymorphic species are likely to produce 

 a great diversity of hybrid offspring (as, indeed, I found to be the 

 case, when I lived in West Sussex), some scentless, others 

 fragrant ; and secondly, that mongrels are probably not rare, as 

 the first crosses normally bear numerous capsules, and may be 

 fertilised by one or other of the parent species. In such a case, 

 the multiplication of special names appears to be needless. 

 V. imberbis Leighton, retained as ft imberbis Henslow by Mr. 

 Williams, is reduced to a form in the Monograph. 



V. epipsila Ledeb., our most recent addition, has already been 

 found in five English and Welsh counties, as well as in Kerry ; 

 besides the type, a variety (glabrescens Asch. & Graebn.) and a 

 form (minor Gregory) are figured, and a luxuriant, w T hite-flowered 

 hybrid with V. palustris is recorded from Dartmoor. 



On p. 71 the following note occurs : — " Mr. A. J. Wilmott 

 (Journ. Bot. xlix. 289-293 (1911)) has shown that, in strict 

 accordance with the rules of synonymy, the name V. canina 

 Linn. ' should be used for the plant which has, since 1823, been 

 called V. Biviniana . . .' I propose, however, to follow the 

 arrangement adopted in the London Catalogue of British Plants, 

 ed. 10, and in Babington's Manual of British Botany, ed. 9, 

 revised by H. & J. Groves. Accordingly, I adhere to the tra- 

 ditional use of the name as applying to the Dog Violet." The 

 London Catalogue is not an authority on nomenclature, the 

 names there adopted only representing the revisers' personal 

 opinion ; again, the last edition of the Manual was, I believe, in 

 print before the results of the Vienna Congress were known : nor 

 is the traditional view uniform, our earlier post-Linnean writers 

 having acted in accordance with Mr. Wilmott's contention. In 

 fact, the use of V. canina L. for either species involves confusion 

 and obscurity ; and, how 7 ever desirable it may be to keep up the 

 names of the Species Plantarum, wherever possible, this one 

 seems to be even more objectionable than Bosa villosa L. If 

 V. ericetorum Schrader (ex Hayne Arzneygewachse) is not valid, 

 V. flavicornis Sm. may stand. 



Many other interesting questions might be discussed ; but for 

 these the book itself should be consulted. Enough has, I think, 

 been said to show that Mrs. Gregory has rendered a real service 

 to students of this attractive but difficult genus ; thus deserving 

 their best thanks and hearty congratulations. 



Edward S. Marshall. 



