54 NATIVE BRITISH ORCHIDACEyE 



might be sufficient, if an insect became smeared with it, to remove the poUinia. To 

 test this I inserted a dry camel's liair brush into a flower, causing it to sweep gently 

 upwards along the middle of the stigma and the face of the anther, which is just 

 above it. It came out with one whole pollinium attached to it by the middle of its 

 convex side, the ends pointing outwards. With a lens I could see a little viscid matter 

 glistening on the brush. On looking to see why only one pollinium was brought 

 away, I found the anther empty — evidently one pollinium had been previously with- 

 drawn. I did the same thing with another flower, and this time both pollinia were 

 easily removed. I repeated the experiment with flower after flower, always with the 

 same result. To make sure that this was really due to the viscid matter of the stigma, 

 I passed a clean brush over the face of the anther in several flowers, without first 

 touching the stigma. This had no eff"ect — the pollinia would not adhere to a dry 

 brush. The object of the elastic hinge now became apparent. The anther must be 

 pressed against the edge of the stigma, so that the pollinia may project far enough 

 forward to ensure that an insect, retreating backwards from the tubular flower, after 

 smearing his back with the viscid secretion of the stigma, shall also brush against 

 the pollinia. These at once adhere by their convex centres, their tips pointing forwards. 

 It is these projecting ends which come into contact with the stigma of the next flower 

 visited. This explains the curious fact that pollinia are very occasionally found 

 adhering to the stigma by their extreme tips, their curved centres standing away 

 quite clear of its surface.^ 



Although always on the watch for insects visiting the flowers, it was nine years 

 before I saw this happen. In May, 1929, Colonel G. H. Evans, F.L.S., my wife and 

 I systematically watched cut spikes of C. ensijolia, together with those of other orchids, 

 on the terrace wall of an hotel garden at Challes-les-Eaux, France. We saw the flowers 

 of ensijolia visited a number of times by a small Hymenopteron, Halictus Smeath- 

 mannellus K. ?, and the still smaller H. politus Schenck ?, which carried off the white 

 pollinia attached by their convex centres to the thorax, and not to the head as I 

 expected. These insects were so small that I saw them going to the flowers without 

 suspecting that they were big enough to touch the stigma while crawling on the lip, 

 until my wife saw one with white pollinia on its back. These little bees took no 

 notice of the flowers of C grandiflora exposed in the same vase. They are figured in 

 PI. 4. A British spike of 17 flowers had 15 pollinia removed. 



' ]ourn. Linn. Soc. Bot. xisv, 511 (March, 1922). 



