300 TILE JOUKNAl. OF IJOTANV 



and ao-aiii on the lltli, I found a spike from the lowest flower of which 

 one poliinium had been taken, the other, in one case, had descended 

 on the stigma in the usual way. On May 17th in the late afternoon, 

 after rain, I found three spikes with both pollinia and two spikes 

 with one poliinium, cleanly removed from one flower in each spike. 

 This could only have been due to the agency of insects, and shows 

 that their visits are not so rare as might be supposed. 



Orchis lactea Poir. This species has been confused with 

 O. trUleiitata Scop., but, although the similarity of the flowers 

 suggests descent from a common ancestor, the habit and appearance 

 of the plant are so different that one suspects that the authors who 

 rep-ard it as a variety of trldeiitata have never seen it growing, or 

 compared living specimens of the two plants. As far as I have 

 observed, they appear to grow on different geological formations, 

 lactea occurring on the schist, whilst tridenfata is abundant in some 

 localities on limestone. I found a colony of lactea at Bonnes, near 

 Hyeres, growing in loose gritty soil consisting of disintegrated 

 schistose rocks. On many of the plants a small white crab-like 

 spider was lying in wait amongst the flowers, its protective colouring 

 rendering it inconspicuous. No bees were about, but it was a cloudy 

 afternoon. 1 found on the ground, at the foot of a spike, a spider 

 which had grip})ed a bee b}^ tlie neck and had evidently fallen with 

 its victim from tlie flower-spike. I put it in my vasculum, and on 

 reaching home found that it still held the bee, man}' times larger 

 than itself, in its jaws, alwa^^s carrying it round to the opposite side 

 of the spike to escape observation. It was identified at the Paris 

 Museum as a young Thomisus onastus, and the bee as A2)is mellijica. 

 From this it appears probable that O. lactea is fertilised, at least in 

 part, by the common hive-bee. The flowers were found to have been 

 well visited, many pollinia having been removed and abundant pollen 

 deposited on many stigmas. 



Cephalan'ihera rubra Rich. With reference to my paper on 

 the fertiUsation of Cephalanthera (Journ. Linn. Soc. xlv. 511), I 

 watched this plant in a wood at Vence, wdiere it was frequent, without 

 seeing any insect visit the flowers. I also at different times exposed 

 cut flowers in various likely places, alwa3's witliout success, until the 

 following happy accident occurred. On June 11th, 1922, I gathered 

 several spikes in an open wood near Challes-les-Eaux (Savoie). My 

 wife carried them in her hand, and, wdien passing the same place on 

 our way back, a red humble-bee came to them and visited three 

 flowers. At 8.15 a.m. on the 17th, a very wet morning, the same 

 kind of bee came to some spikes of G. rubra in a mixed bunch of 

 flowers at the open window of my room, and was so engrossed in the 

 third flower he visited that I caught him with a pill-box. He was 

 identified at the Paris Museum as Bomhiis arjrorum F. 



C. GRAXDiFLOEA S. F. Gray. At Mantes, near Paris, in May 1921, 

 I found two flowers from which one poliinium, and one from which 

 both pollinia, had been removed. Neai- Horsley, Surrey, on June 21st, 

 1921, I found two "flowers from each of which both pollinia had been 

 withdrawn, near the place where I saw this species visited by Bombus 

 lucorum on June 17th, 1919. At Vence on May 9th, 1922, I found 



