'"■■/. 17, 1872J 



NA TURE 



49Q 



. ,.0 a large aperture, and were widely separated from 

 the tenth stamen. If so, it is curious to see the nectar- 

 holding cavity so often formed by the stamens here formed 

 by the vexillum. The entire freedom and wide separa- 

 tion in the tenth stamen, in a flower displaying such a 

 tendency to cohesion, is also curious. Possibly this is 

 necessary in order to preserve a sufficient aperture to give 

 access to the nectary. 



Lotus coriiiciilaties. — The flowers of this plant again, 

 though in umbels, when open assume the normal posi- 

 tion with the vexillum uppermost. 



Lupin (Howet with one iving cut off). 



The wings are free fromthe keel. Thekeelislong,pointed, 

 and united for some distance above as well as below, with 

 an aperture at the apex (see Fig. 10). Thetentii stamen is 

 free, and is separate from the others at the base ; the 

 staminal tube is stiff and enlarged at the base into a 

 cavity, which contains nectar. The pollen is moist and 

 abundant. The style is capitate and sdff, but without 

 hairs or brush. How then can the moist pollen be forced 

 out of the narrow mouth of the long pointed keel so as 

 to meet an entering insect ? In a very curious way. Five 



Fig. 13 -Lnp-n (keel). 



of the stamens, viz. those of the inner whorl, are shorter 

 than the others, and their filaments are dilated at the top. 

 These filaments are stiff, and, 1 believe, continue to grow 

 after the five anthers of the other whorl have shed their 

 pollen. The dilatation of the filament is wedge or club- 

 shaped, the broad end of the wedge being uppermost (see 

 Fig. 11). Consequently,onany pressure being applied to the 

 keel, the broad ends of these wedges, supported by their stiff 

 filaments, collect the pollen, and push it before them to and 

 out of the mouth of the keel, where it is seen to adhere to 



the bjdy of the insect which is passing down the keel. It 

 is to be observed that the shape of the dilated filament is 

 such that, whilst pollen might work past it from below 

 upwards, the broad flat upper end of the filament meeting 

 the narrowing tube of the keel can scarcely allow it to 

 pass downwards. 



Garden Lupin (common tall blue and white). — In 

 the long raceme of this plant the pedicels are nearly per- 

 pendicular in the bud, become horizontal whilst the 

 blossom is open, and rise so as to approach the perpen- 

 dicular again afterwards. The wings are attached to 

 each other below, are blunt, and are folded over at top so 



as to afford an excellent resting place. They are not 

 attached to the keel, and move downwards more easily 

 than it does. The keel is very long, very pointed, and 

 the upper edges are slightly connected with an opening at 

 the apex so as to form an approach to a tube. The apex 

 just appears between the wings. The upper edges of the 

 keel are furnished with a few hairs (see Figs. 12 and 13). 

 The filaments of the stamens are entirely joined 

 together so as to form a close fitting tube round the 

 ovary. There is no cavity within the tube for nectar, no 

 apertures into it at the base, and it is too long and too 

 close fitting for an insect to thrust its proboscis down. 



side cut off). 



There is a cavity at the back and base of the vexillum 

 in which I have not been able to find nectar. But the 

 bees, which constantly visit these flowers, certainly go to 

 this cavity for what they want, and not to the staminal 

 tube. Five of the stamens compose, I believe, what must 

 be the outer whorl, are longer in their filaments than the 

 other five, and have longer anthers (see Fig. 14). These are 

 mature, and before the flower opens have shed their pollen, 

 which remains in a moist mass towards the mouth of the 

 pointed keel. Their filamentsthenwitherand contract. The 

 other whorl of stamens are shorter, and the anthers much 



smaller, but they are later than the first whorl, and their 

 filaments grOiV and remain stitf after the filaments of the 

 first whorl have withered. They consequently, on pres- 

 sure being applied to the keel, thrust the mass of pollen 

 upwards to its mouth. The style is long, and a rmg of 

 hairs surround the stigma, of which the upper and 

 inner are the longest, and all of which arc set upwards, 

 so that on pressure being applied to the keel the hairs 

 sweep out the mass of moist pollen which the stamens 

 have thrust to the mouth (sec Figs. 15 and 16). It is quite 

 pretty to watch the little stream of bright orange pollen 



Fig. 17.— 67f.v(Gorse). 



emerging from the narrow aperture of the blue keel, and 

 between the bright blue wings. 



The shorter yellow and blue garden lupins and Lupiir.is 

 arhorcus are similarly constructed. In the latter flower 

 the folding over of the wings at the top, and the cavity 

 at the base of vexillum, are strongly marked. 



What is the use of the hairs on the edges of the keel ? 



