Flower of Sweet Pea, in its natural position.

Fig. 72. - Flower of Sweet Pea, in its natural position.

Ditto. The wings are depressed, the stamens and pistil exposed.

Fig. 73. - Ditto. The wings are depressed, the stamens and pistil exposed.

In the Sweet Pea (Figs. 72 and 73 on account of its larger size the action is still more easily visible. Fig. 72 represents a flower in the natural position. Now if the two ends of the wings be taken between the finger and thumb, and pressed down, so as to imitate the effect produced by the pressure of an insect, the keel is depressed with the wings, while the pistil and stamens are thus partly uncovered, as shown in Fig. 73. When the pressure is removed, the flower resumes its former position.

Trifolium repens (the White Clover) agrees with Lotus in its general structure, but is somewhat simpler. The wings are actually united to the keel at one point. In T. pratense the flowers are longer, and the honey is only accessible to those bees which have a very long proboscis. As in other such cases, however, Bombus terrestris obtains access to it by eating a hole through the side of the flower. According to Darwin this species is only fertilised by humble bees, but Delpino disputes this. Trifolium subterraneum has small cleistogamous flowers (MohlBot. Zeit., 1863) besides the usual ones.

Anthyllis vulneraria also agrees with Lotus in its general arrangement. The tube of the flower is, however, elongated; and in consequence, this species is only visited by bees with long tongues. In the young flower, though the pistil is in the keel, and necessarily in contact with the pollen, H. Muler has observed that the stigma is dry, and that no pollen adheres to it. . Subsequently, however, when most, or all, of the pollen has been removed, the stigma becomes sticky, and pollen adheres closely to it.

In Ononis (the Restharrow) the general arrangement is very similar. There are, however, several important differences. Ononis does not secrete honey, and consequently there is no need for the separation of the upper stamen, which in this genus is attached to the rest. Again, in Ononis all the stamens are thickened at the end; the outer ones, however, much more so than the inner ones. The inner ones, on the contrary, produce much more pollen than the others; a difference of function which is even more marked in the Lupins.

Ononis is exclusively fertilised by bees, and H. Muler has repeatedly seen male bees visiting this species in a vain search for honey.

Onobrychis sativa (the Common Sainfoin) agrees with Trifolium repens (Clover) in its general structure; but the wings are greatly reduced in size and appear to serve only in preventing the honey being reached from the side, or at least in rendering this more difficult. This species is sufficiently conspicuous, and as the honey is accessible even to insects with a short proboscis, it is much visited. When mature, the stigma projects I to 1 1/2eyond the keel, and according to H. Muler the flower has lost the power of self-fertilisation.

In Genista tinctoria the ten anthers lie in two distinct rows. While the flower is still in the bud, the four upper anthers of the outer row are already on the point of opening, while those of the inner circle have not nearly reached their full size. These four anthers now open and shed their pollen into the space at the apex of the keel, after which they shrivel up. The fifth, although it has attained its full size, remains closed. The next process is that this anther and those of the second row also open, and the pollen occupies the end of the keel between the anthers and the stigma, as in Lotus. While, however, in Lotus when the insect leaves the flower and the pressure is thus removed, the keel resumes its position, and the stamens and pistil are again protected; in Genista tinctoria, on the contrary, the flower opens once for all. The keel is at first nearly parallel to the standard (Fig. 74). This position is, however, one of tension; the keel is retained in it by the union of its upper margins, which inclose and retain the curved pistil which presses against them like a spring. The sides of the keel have near the base a projecting lobe (Fig. 76 m), which locks with one at the corresponding part of the wing. When an insect, alighting on the flower presses open the keel in search of pollen, as soon as the curved end of the pistil is set free, it springs up with a jerk, the keel, on the contrary, springs back (Figs. 75 and 76), and the pollen is ejected in a shower. It appears that the flowers do not open of themselves if insects are prevented from visiting them (Henslow, Four. Linn. Soc, v. x. p. 468).

Flower of Genista tinctoria unopened.

Fig. 74. - Flower of Genista tinctoria unopened.

Figs. 75, 76. - Ditto, opened, std, standard; v, wing; k, keel; m, projection on keel

Genista tinctoria contains no honey, and yet it is visited by several insects which do not consume pollen.

The flowers of the Common Furze (Ulex Enropceus) agree in essentials with those of the preceding species. The calyx, however, is larger, and coloured; the wings are longer in proportion and project beyond the keel. They also lock at the base with the keel, and when they are pressed downwards the flower bursts open. The Furze has, like Cytisus and the Broom, on the outer part of the staminal lobe a honey-containing tissue.

In the Laburnum, the tip of the pistil is protected from its own pollen by a ring of close hairs; when, however, the pistil has grown to the very point of the keel, these hairs shrivel and turn outwards, so as to expose the tip of the pistil, which thus comes in direct contact with the breast of any bee which may alight on the flower.