A NEW EUEOPEAN EPIPACTIS 103 



of the stigma, and if removed leave two circles of pollen on the stigma, 

 the diameter of each free from pollen, for they are cleft in two 

 at the base, though undivided at the apex. It will thus he seen 

 not only that every one of the reproductive organs in Miiller's plant 

 differs from the corresponding one in latifolia, but that they are 

 arranged and co-ordinated on a different plan. Self-fertilization 

 appears to be inevitable, but in latifolia the pollinia are transported 

 by insects. 



Is Midler's plant the true E. viridiflora Rchb., or are continental 

 botanists right in regarding the latter as a subspecies, race, or variety 

 of latifolia ? Reichenbach's description (Flor. Germ. Excurs. p. 134) 

 is as follows : — 



" E. viridiflora (Sen p.) Hoffm. Folds elliptico-acuminatis 

 amplexicaulibus, labio coroato-ovato acuto piano, petala sepalaque 

 lanceolata ovario oblongo longiora aequante. Serapias latifolia 

 b. sylvestris Pers. Serap. litifolia Fl. dan. 811. Pedalis, vaginae 

 arctae praecedentis [E. I at folia'], folia omnesque reliquae partes 

 magis elongatae tenuioresque ut ortum e locis umbrosis indicent ; flores 

 virides plus vel minus rubicundi." 



English botanists will recognize the source from which Babington 

 drew his description of E. media Fries. It will be seen that, with 

 the exception of the longer acuminate leaves, and the " labio piano," 

 there is no character given to distinguish the plant from latifolia. 

 The younger Reichenbach was unable to accept his father's view as 

 to the specific rank of E. viridiflora, though he must have con- 

 sidered its claim thereto very carefully. His Icones — a wonderful 

 monograph of European orchids — show that he was interested in the 

 reproductive organs and frequently figured them. He closely studied 

 viridiflora, for he wrote a much fuller description of it than his 

 father, but he says not a word about the absence of a rostellum or the 

 unique position of the stigma (both of which characters are obvious 

 at a glance, and in marked contrast with latifolia), nor does he 

 mention the different shape of the anther and of the pollinia. His 

 silence is capable of only one reasonable explanation ; his varians 

 (viridiflora) was not Miiller's plant, and did not exhibit its charac- 

 ters. Continental authors are therefore right in regarding E. viridi- 

 flora Rchb. as a form of latifolia. Without suspecting it, Midler 

 was dealing with a hitherto unnamed species, and was mistaken in 

 identifying it with E. viridiflora Rchb., a very natural error, for 

 Reichenbach's description, as far as it went, agreed very well with 

 his plant, and it was clear that the latter did not belong to any 

 other known species of Epipactis. It would therefore seem to him 

 that it must be viridiflora. 



Ascherson and Graebner (Syn. Mitt. eur. Flor. p. 862) are the 

 only authors, as far as I know, who mention that viridiflora has no 

 rostellum, and is self -fertilised. I have ascertained that they did so 

 solely on the strength of Miiller's paper, and not from study of the 

 living plant. 



It therefore seems best to restrict the name viridiflora to the 

 variety of latifolia to which it was originally given and is now almost 

 universally applied, and to give the name E. Muelleri to the plant 

 brought to light by his researches. 



