228 



tatum. The drawing I mention has since been confirmed by 

 numerous observations during a series of years. 



Dissections of the immature buds shew that the antennae are only 

 developed after the pollen gland, and all other tissue is completely 

 formed. 



When half mature buds are opened, the antennaj are represented 

 merely by a slight prominence on the surface of the column. Darwin 

 states that he is almost certain " that the disk does not separate by the 

 simple mechanical movement of the antennse," but I venture to assume 

 that his material was inadequate for him to fully demonstrate this, 

 for he certainly failed to do so. 



The hypothesis of mechanical action is moreover clearly sup- 

 ported by his own experiments, not only that on C. saccatum where 

 the column, loithout tension, had no sensibility, as it certainly would 

 not have if the action is really mechanical as I suggest. 



If the arms of a pair of spectacles having curved holders for the 

 ears are brought together, the left upright and the right horizon- 

 tally folded downwards, they will almost exactly represent the position 

 of the points of the antennae in the flower of C. tridentatum. The 

 point of securest contact between the column and the pollen masses 

 and their appendages has been seen to be situated on the margin 

 of the column on each side of the caudicle, and it is fairly evident 

 that a turgid antennaj would mechanically aiford suflScient vibration 

 to cause a disruption of the separation layer, and thus allow of the 

 expulsion of the pollen masses by contraction of the caudicle. If the 

 antenna are really possessed of a special irritability or stimulating 

 property, proof is necessary to show how it acts ; but so far none has 

 been forthcoming ; Darwin's strongest point, (viz.) the absence of 

 antennae in MonacJianthus , being quite as strong if not stronger in 

 favour of simple mechanical action. 



Until the perfume of the flower is developed it receives no visits 

 from the fertilizing insects, and some flowers that were emasculated, 

 developed no perfume, but on the third morning after opening were 

 in a shrivelled condition. On the same date flowers that were not 

 emasculated, developed perfume and were visited by insects. Some 

 flowers which developed perfume also changed their position from 

 that at which they stood at the period of antJiesis. The attractive 

 matter for the bees which visit them appears to be similar in char- 

 acter to that noted in my annual Report for 1890 as the attraction 

 in Coryanthes flowers, and is situated on the underside of the lip. 



