HELLEBORINE VIRIDIFLORA IN BRITAIN 345 



species is very similar, if not identical, in shape and size ; but in 

 H. viridiflora there is never a linear median hunch as in^ H. 

 violacea. In the colour of the stem and foliage the fresh yellowish- 

 green of the dune plant contrasts strongly with the colour of 

 H. violacea, which is of a grey-green, suffused with purple. Apart 

 from other distinctions already mentioned, H. viricliflora may be 

 readily separated from all forms of H. latifolia by the smaller 

 size and different colour of the flowers, and especially by the form 

 of the label. This in H. viricliflora is as long as broad, and 

 gradually tapers to an acute point, which is frequently not at all 

 recurved ; whereas in H. latifolia the label is always broader than 

 long, and terminates in a recurved apiculus. From H. media 

 Druce, which seems to cover a number of -forms, some of which 

 are, perhaps, not always separable from H. latifolia, H. viridiflora 

 is, perhaps, less easily distinguishable. We have not had the 

 advantage of fresh examples of H. media for comparison, but 

 basing our comparisons on authentic dried material and descrip- 

 tions, H. viricliflora is separable from H. media in the following 

 respects : Its flowers are smaller, although arranged in a similar 

 lax raceme. In habit it is usually even more slender and wiry- 

 looking than H. media, with narrower leaves, and the flowers, 

 greener and less conspicuous, are devoid of that purple tinge, 

 some trace of which is usually present in H. media. In this latter 

 the germen is slender in comparison with the thickness of the 

 flower-buds and flowers, and tapers gradually into the pedicel. 

 In our examples of H. viricliflora the germen and the flower buds 

 are nearly equal in size, both as regards length and thickness. 

 The germen of the unopened flowers is as long as, and often 

 somewhat exceeds, the length of the flower {i.e., the length of the 

 sepals), and approximates thereto in thickness; it also rapidly 

 narrows into the short pedicel. In this respect the following 

 measurements may be cited as average ones for the dune plant : — 

 Uppermost flower-buds : germen (not including pedicel) 8 mm. 

 long, bud (from top of germen to tips of sepals) 7 mm. Fully 

 open flower: germen 10 mm. long; flower 9-10 mm. Lowest 

 flower of raceme : germen 12-13 mm. long x 6 mm. in diameter. 

 The dune plant, by reason of the lax, few-flowered racemes of 

 small greenish little-expanded flowers, with the relatively large 

 and thick germens, has a distinct facies of its own. _ 



In our examples, H. viridiflora, as compared with its nearest 

 congeners, has a comparatively small amount of pubescence ; and 

 this is particularly noticeable in the case of the germen, which is 

 practically glabrous. In H. latifolia, media, and violacea, also, 

 the upper surface of the nerves of the leaves is covered with 

 minute but distinct asperities. These are very faint and hardly 

 visible on the leaves of H. viricliflora. 



We are satisfied that H. viridiflora, as we know it from the 

 Lancashire dunes, is quite distinct and separable from all the other 

 British species of the genus. On the Continent it seems to be 

 sometimes confused with greenish-flowered examples of H. latifolia 

 and the var. virescens of H. atropurjmrea. Any deficiency of 



