CEPHALANTIIERA OR EPIPACTTS 73 



known, Individual hybrids incline sometimes to one parent, sometimes 

 to the other. 



His statement {I. c. 137) that C. cvcullata had only once been 

 found conlirms the supposition that it is a hybrid — had it been an 

 indigenous species it would hardly have been so excessively rare, 

 unless on the verge of extinction. The following points in his 

 description appear to indicate the iniiuence of Limodomm — Caulis 



validus, vagiucU tres membranaceie, amplissimse, ore oblicjuo 



acutiusculai, rostellum obtuse 4-5 dentatum (apparently the upper 

 edge of the stigma is referred to). It may be added that Dr. Wett- 

 stein's figures show considerable similarity between tlie lips of 

 C. cucullata and Limodortim. 



If this supposition is correct, the author's arguments based on 

 C. cucullata are put out of court, and his plea for the inclusion of 

 Limodorum in the genus Epipactis falls to the ground. 



It is interesting to note the circumstance under which Dr. 

 Wettstein's paper was written. He had just been studying a hybrid 

 between 0. grandijlora and jEJ. ruhiginosa, and the need of choice of 

 a generic name for it led him to make an exhaustive study of tlie two 

 genera. He had evidently a strong bias against the probability of 

 the occurrence of bi-generic hybrids, for he argues that the very 

 existence of this hybrid pleaded for the union of the two genera in 

 one. I might adduce, he says, as a new proof of the correctness of 

 my view, that in most cases the occurrence of bi-generic hybrids 

 ought to suggest the homogeneity of the genera concerned. He was 

 thus handicapped by a preconceived idea. In botanical investigations 

 an open mind is essential. A biassed mind cannot exercise unbiassed 

 judicial functions — the judge is at heart an advocate. The link 

 which in Dr. Wettstein's opinion joined, in the one case Cejjhalan- 

 thera and EpipactiSy in the other Epipactis and Limodorum, was in 

 each case a hybrid. His prejudice against bi-generic hybrids pre- 

 vented him from giving due weight to the unanswerable evidence of 

 former Avriters as to the differences between the genera in question, 

 and to lay undue stress on the occurrence of a spur in C. cucullata, 

 which might have been, and in all probability actually was, due to 

 hybridity. 



A careful study of Dr. Wettstein's otherwise very able paper leads 

 to the conclusion that Ceplialanthera, Epipactis, and Limodorum are 

 generically distinct, for his arguments entirely fail to shake the 

 position taken up by earlier writers. 



Ceplialanthera cucullata is, in all probability, a hybrid between 

 C. grandijlora and Limodorum abort imim. When Reichenbach 

 wrote his description he was not sure whether the flowers were white 

 or rosy, and naively states that in his figure he has shown them as 

 white, but that this can easily be altered should they eventually 

 prove to be rose. In his second supplement, however {I. c. 181), he 

 announces that they are ' luride alba,' as in C. grandijlora. 



P.S. — Since writing the above I have come across a footnote by 

 Dr. G. R. v. Beck to his paper Erwiderung auf Dr. Wettstein's 



