Fertilization of the Scarlet Runner. 257 



downwards and towards the base of the left wing. To a spec- 

 tator looking into the flower, the way, if any, to the base of 

 the petals and of the ovary is obviously down the left wing 

 and past the mouth of the spiral tube; but there is hardly 

 room, in the quiescent state of the flower, even for the pro- 

 boscis of an insect between the base of the wing and the keel. 



On following the development of the flower in the bud, it 

 appears that the peculiarity of the spiral coil of the keel, with 

 its enclosed stamens and pistil, only appears at a late stage. 

 In the earlier stage, though the anthers and the brush round 

 the style are fully formed, the length of the style, filaments, 

 and keel, and the form of the keel, are like those of other 

 papilionaceous flowers. 



The filaments of the stamens, which are, except one, united 

 and stiff at the base, are in the upper part very thin and flexi- 

 ble, and follow the windings of the spiral keel. The anthers, 

 which are small, lie in two rows entirely within the tube or 

 hollow of the keel, a little within its mouth, and within and 

 below the stigma. 



The pollen is not very abundant, and is not dry and dusty, 

 but moist and sticky. 



The style is stout, strong, and very elastic ; it is set firmly 

 on the stiff upright ovary, so that its point of attachment to 

 the ovary and base of the flower is at some distance from that 

 of the attachment of the keel to the stiff claw of the wing. 

 The stigma is at the extremity of the spiral coil, and on the 

 lower or outer side, i. e. on the side next the wings ; it is 

 sticky, and is clothed with fine hairs. In the untouched flower 

 it just protrudes out of the mouth of the tube of the keel, so 

 that its tip is just visible on looking downwards into the flower. 

 A little below it the style is clothed with stiffish hairs or bris- 

 tles, which partly encircle the style like a circular brush, but 

 which are considerably more in number on the upper or outer 

 side of the coil than on the inner or under side. This brush 

 is opposite to and in contact with the opening anthers. 



Under these circumstances it is not obvious at first sight 

 how the flower is fertilized. As regards self-fertilization, the 

 arrangement does not seem a happy one ; for the stigma is 

 outside and below the tube of the keel, whilst the anthers and 

 pollen are shut up within it. 



The plants are frequented by, indeed they swarm with, bees. 

 These are of various kinds, of which I do not know the names. 

 But, so far as I could see, the smaller or hive-bees never suc- 

 ceeded in getting what they wanted through the mouth of the 

 flower. ^ They occasionally lighted on the petals, and looked 

 in, but invariably went round to the back of the flower, and 



