346 THE JOURNAL OF BOTANY 



purple coloration in the flowers of H. latifolia and its allies is 

 usually attributed to the influence of deep shade. As against 

 this, it is a noteworthy fact that the flowers in our dune Helle- 

 borine viridiflora have no purple tinge at all, although they are 

 fully exposed to sunshine, growing, as the plant does, on the open 

 dunes. 



The flowers seem to contain a much smaller quantity of nectar 

 than do those of H. latifolia and other species of Helleborine 

 which we have seen. This, in conjunction with the greenish, less 

 open character of the flowers, and an early and rapid swelling of 

 the ovary, seems rather suggestive of self-fertilization. We may 

 mention that Mr. Eolfe independently suggested that the plant 

 looks like a self-fertilizing form. This is a point which calls for 

 further investigation. We have made no observations as to 

 insect visitors, but we may mention that on one occasion we 

 found, on a single plant, the dead bodies of two different species 

 of Diptera, the heads of which had, in some fashion, got en- 

 trapped in the throat of the flowers. One belonged to an insect- 

 eating genus, and the other was not identified. 



H. viridiflora occurs widely scattered on the sand-dunes from 

 Hall Eoad, in South Lancashire (v.-c. 59) to South Shore, in 

 West Lancashire v.-c. 60), and in some places it occurs in fair 

 quantity. Its special habitat is the low secondary dunes built up 

 by Salix rejpens, and it does not usually occur, either in the wetter 

 part of the dune valleys, or on the more elevated and drier 

 marram-clad dunes. Last summer it was in flower on June 20th, 

 and its flowering period was practically over by the end of July, 

 as in the first week of August we found it difficult, as above 

 stated, to collect flowering specimens. 



We have not troubled to get details as to the Continental dis- 

 tribution of Helleborine viridiflora, but have consulted a few of 

 the Floras relating to coastal areas. 



In Lloyd and Foucaud's Fl. de I'Ouest de France, H. viridiflora 

 is stated to occur in sandy woods on the coast. Buchenau, in Fl. 

 Ostfries. Inseln (1881), p. 134, records Epipactis latifolia as 

 occurring on " bewachsenen Dunen " in the various islands, and 

 on Norderney in " Dunenthaler in der Mitte der Inseln." E. lati- 

 folia is recorded on the inner dunes of the coast of Holland ; and 

 in the Dutch islands of Ameland, &c, it is found in " duin- 

 pannen " with Salix repens (Holkema, De Plant. Neder. Noord- 

 zeeelanden). It is possible that some of these dune plants may 

 also prove to be H. viridiflora. It is certain, however, that 

 H. viridiflora is by no means a littoral plant, as according to 

 Eouy it is disseminated over a wide area in the interior of 

 France. 



We desire to express our indebtedness to Mr. E. A. Eolfe ; also 

 to Messrs. Albert Wilson, H. Adair, C. E. Salmon, and C. E. 

 Eitchings, for sending us fresh specimens of various species of 

 Helleborine ; and also to Mr. A. Bennett for references to some of 

 the literature cited. 



