Fig. 17.—Japanese Wineberry (Rubus phœnicolasius).
Leaf and panicle, 2/3; flower after pollination; ripe fruit, both slightly enlarged.
But it is in the flower itself that we find the most ingenious arrangements to encourage useful and discourage useless visitors, to assist the former to pollinate the flower, and while offering nectar to the welcome guest to deny it to the unwelcome. The first stage in this specialization is that the flower, instead of having its axis vertical, and facing the sky, is turned on its side by the curving of its stalk, and looks out horizontally. The effect of this is to cause a flying insect on approaching the flower to alight in a particular position—namely, on the lowest petal. Following on the adoption of this attitude the next stage in development is seen in the parts of the flower beginning to alter their shape and position relative to each other and often also their colour. Thus, beginning with a quite regular flower, we can arrange a series showing more and more asymmetry. The tendency is generally for the lowest petal to become enlarged and often conspicuously marked, providing a broad, convenient platform on which insects may alight, while the remainder form walls and roof, protecting the important parts within and by their shape, which is often narrowed and tubular behind, barring access to all but chosen visitors. To find a full series illustrating these transformations we do not need to go to plants widely separated in their affinities. In the Buttercup order (Ranunculaceæ) alone every gradation may be found. The flowers of the Buttercups themselves are upright and quite regular. In the Larkspur (Delphinium) the flower is turned on its side, and a puzzling combination of coloured sepals and petals—five bright blue unequal sepals and a single large purplish petal of peculiar shape with a long hollow spur behind—produces a quite irregular blossom. The process is carried farther again in the Monkshood (Aconitum), in whose well-known blue flower the sepals and petals combine to produce a strikingly irregular blossom, with the upper sepal arching over into a great hood protecting the rest of the flower. In such irregular flowers the essential parts—the pollen-producing and pollen-receiving portions, or stamens and stigma—also alter their position and form, and are so placed that an insect, visiting the flower to obtain nectar (which is generally stored at the back, well out of the way), must of necessity receive pollen on its body, and probably deposit pollen on the stigma. To describe the variety and ingenuity of these devices as found in different flowers might well occupy several chapters, and only one or two examples can be quoted here; familiar wild flowers are chosen, and the reader should examine them for himself to understand their structure. In the well-known Pea type, one great petal arches over the flower; two narrow ones stand one on either side; the remaining two stand on edge below, with their margins in contact, enclosing the stamens and pistil. An insect visiting the flower alights naturally on the keel or pair of lower petals. Pressed down by its weight, these open, often with a sudden movement like bursting, and dust the insect with pollen. Compare also the flowers of the Snapdragons (Antirrhinum) and Toadflaxes (Linaria), in which the upper and lower lips of the corolla meet like a closed mouth, which can be forced open only by a strong insect like a bee, and is safe from predatory visits of smaller fry ([Fig. 18]). In the Sages (Salvia) the corolla is tubular at the base; there is a large lobed lip on which visiting insects alight, and a hooded roof above arching over the stamens and pistil, which are placed close against it, overhanging the entrance to the corolla-tube, at the base of which the nectar is stored. The stamens, only two of which are developed, have each a hinge near the top, the part above the hinge being like a curved rod supported near its middle. These two curved rods stand normally in a vertical position, so that their lower ends partly block the entrance to the tube; the pollen is borne at their upper ends. Should a bee insert its head down the tube in search of nectar, it pushes the lower ends of the hinged rods upwards, with the result that their upper ends swing downward against the bee’s back, dusting it with pollen just at that part of its body which, if the bee should visit a rather older flower, would come in contact with the stigma, the slender stalk of which (the style) increases in length during the period of flowering, and is in consequence the more liable to be encountered.
Fig. 18.—Flower and Fruit of Linaria purpurea. 2/1.
Only one more instance can be referred to, which can be tested by the reader any summer day wherever any of our native Orchids grow. In these, the most highly specialized of all plant groups as regards pollination by insects, the general arrangement of the flower is often somewhat similar in a general sense to the last case; but here the sepals and petals which between them form the platform, tube, sides, and roof of the flower, are all separate and often differently and elaborately coloured. The essential organs are greatly modified and hardly recognizable at first. There is only one stamen, producing two clusters of pollen, which are embedded in the roof of the flower. Each possesses a slender stalk which terminates in a little sticky disc which projects from the general surface. The pollen grains are held together in a mass by fine threads, and the whole with its stalk—the pollinium—resembles a lemonade bottle in shape. The stigma is also embedded, forming a sticky surface in the roof of the flower behind the stamen. When an insect inserts its head into the flower, its forehead comes in contact with the sticky ends of the pollinia, which adhere, so that on leaving the flower the insect flies away with the pollen sticking to its forehead like two little horns. And now a remarkable thing happens. The stalks of the pollinia, drying rapidly in the air, contract unequally, and become curved, so that the pollinia bend forward into a horizontal position. When the insect visits another flower and thrusts in its head, the pollen consequently comes in contact with the sticky stigmatic surface farther down the tube, and cross-pollination is effected.
In the cases of many of these highly specialized flowers, one is no less struck with the perfection of the arrangements made for preventing self-pollination, than those adapted to securing cross-pollination. But in a few, on the contrary, self-pollination is specially arranged for.
It must be pointed out that the insects which pollinate these specialized flowers have in many cases acquired modifications in their structure corresponding to the modifications in the flowers which they frequent. In the more specialized forms, indeed, plant and animal have become entirely dependent on each other; the plants would become extinct in the absence of the special insects through whose agency they are able by pollination to produce fertile seed; and the insects would likewise die out if the flowers to whose nectar and pollen they look for food were not available.