' ^ 



FERTILIZATION OF CEPHALANTHERA 



5i:i 



I passed a clean bnr,h over the face ol; the anther in several flowers, without 

 first touching the stigma. This had no effect— the pollinia would not adhere 

 to a dry brush. The obj(^ct oi: the elastic hinge now became appareut. The 

 anther must he pressed against the back of the stigma, so that the pollinia 

 may project far enough forward to ensure that an insect, retreating backwards 

 from a flower, after smearing himself with the viscid matter of the stigma, 

 shall also brush against the pollinia. These at once adhere by their convex 



centres, their ends pointing forwards. It is these projecting ends wliich 

 come into contact with the stigma of the next flower visited. This explains 



the curious fact that pollinia are occasionally found adhering to the stigma 

 bv their extreme tips, their curved centres standing away quite clear of its 



surface. 



In Apri 

 at Hyeros, where it is rare 



ndifiora {tlie species \\lucli DarAvin investigated) 

 On inserting a brush into the flower, it came out 



with tlie oTcater part of tlie pollinia attached. 



^y 



secretion is 



sufficiently copious to remove the pollinia, but there is this 



diflteVenco. If the flower is visited before the pollinia have become anchored 



to the edge of tho stigma, they can be removed entire, but if already anchored 



ly their ujjper portions can be withdrawn. 



only 



'ft 



necessarily always self-fertilized. To my surprise, for at that time I believed 



self-fenilization alone to be possible, I 



difi 



humble-bee, Bomhm lucorum, in Surrey*. In my eagerness to secure the 

 bee for identification, I caught it at once. As it bore no polhnui I regarded 

 the visit as a purely accidental occurrence. I now think that, had I reframcd 

 from catchin<.- it, I might have witnessed the actual removal of pollinia by a 



humble-bee. 



iifi 



J 



always self-feitilizcd. 



I 



e^Lamined numbers of flowers at Vence, but only found two from winch the 

 pollinia had been withdrawn. At Mantes, near Paris, I also found two- 

 flowers from which one pollinium, and one from which both pollinia, had 



been removed. t ^ ^ 



Our third species, the beautiful rose-red teplialanthera riihra, 1 found 



at Vence in May 1920. The pollinia were readily removed with a brusli,. 



¥< 



M 



As I happened to have a dead bee 



of 



iust the richt size [Osmia leiana S), caught sleeping in a tlower oi >:,erapias 

 lonaipetalaiG-^ivd. Chron. 1920, p. 70), I resolved to see how it would answer 



wirh w.^— 'v^- - i c^ - . . ^ ^ r\ 



On 



wiru C. nd>ra. Grasping it with my forceps I ins<n-ted it into a flower 

 ^vit1ldrawing it so that it lightly brushed against the stigma and amber. ^^■ 

 the verv fii'^t trial both pollinia were removed with ease. 



Since this paper was read, I 



di/l 



near the same spot where Sombus Incorwn was .ee.i to visit tlie ilow.rs on Juno 7th, 191.), 

 f-oni two flowers of wlilclibotli the polUnia had been removed. There was no polle,, on the 



stigmas. 



