A NEW EUROPEAN EPIPACTIS 105 



If Mueller/ were originally latifolia, and the stigma began to 

 turn backwards with a view to self-fertilisation, an early result would 

 have been that the rostelluni would be withdrawn beyond the reach of 

 visiting insects, and cross-fertilisation would immediately cease. 

 This would be disadvantageous to the plant, and, if it occurred before 

 self-fertilisation were fully assured, might cause its extinction. Varia- 

 tion, however, to be maintained and increased, cannot be harmful to 

 the plant, or natural selection automatically puts an end to it. 



Even supposing that the stigma did move backward, like a 

 roller-blind, over the clinandrum, so that the pollinia were deposited 

 direct on its viscid surface, these would lie on their sides, as in 

 latifolia. In Muelleri they stand erect on their bases, a very much 

 smaller area thus coming into contact with the stigma. To bring 

 this about there must have been a concurrent modification of the 

 anther also. How can Ave conceive that the anther should have been 

 modified so as to cause the pollinia to stand on their bases, when it 

 would clearly be more advantageous for them to continue to lie on 

 their sides ? Moreover, full self-fertilisation would have been attained 

 as soon as the stigma had moved far enough backwards to receive 

 the whole of the pollinia. Why should the backward movement 

 continue till part of the stigma is thrust well beneath the base of the 

 anther, where it is out of action ? If Dr. Schlechter's suggestion is 

 correct, latifolia, in changing into Muelleri, has sacrificed an 

 extremely efficient system of cross-pollination for an inferior method, 

 and has further carried its modification so far that its self-pollina- 

 tion is less effective than it would have been if the variation had 

 been arrested at an earlier stage. From a biological point of view 

 the hypothesis that Muelleri is a self-fertilising form of latifolia 

 appears to present insuperable difficulties. 



If Muelleri is a variety of latifolia, we might reasonably expect 

 to find intermediate forms. I could find nothing of the kind at 

 Thorenc, except one plant which had the short ovate leaves of Camus' 

 viridMora , and may possibly have been a hybrid with it, though, 

 in view of the extensive leaf-variation in latifolia, I put it down 

 to a similar cause. Midler mentions an intermediate specimen from 

 Driberg, apparently the only one found. This absence of inter- 

 mediate forms is in strong contrast with E. latifolia var. viridijlora, 

 in which they are said to be so numerous and variable that it is 

 often difficult to decide whether a particular plant belongs to the 

 type or the variety. 



Even assuming that Muelleri arose by variation from latifolia, 

 the marked differences it now exhibits, if permanent, appear suffi- 

 cient to justify its claim to specific rank. Dr. Schlechter has seen 

 my figures of the column, which he said in every way showed the 

 same characters which had already been pointed out to him in the 

 case of German specimens. They therefore appear to be constant in 

 such widely separated areas as Germany and the South of France. 

 It seems scarcely in the interests of exact science to regard two plants, 

 showing marked and constant differences in the essential parts of the 

 flower, as belonging to one and the same species. Is it not rather a 

 case of essential differences masked by a superficial outward re- 

 semblance ? 



Journal of Botany. — Vol. 59. [Apbil, 1921.] i 



