In the process of pollination a bee alights on the lip. There, guided by the hairs or ridges, it is led to the orifice of the spur with its store of honeyed juice. The position of this orifice, as we have seen, is at the base of the lip and of the column, so that the in sect, if of sufficient size, while bending its head to insert the proboscis into the spur, almost of necessity displaces the pollen masses. Liberated from the anthers, these adhere to the head or back of the insect by means of the sticky gland at the bottom of the caudicle. Having sipped the nectar the insect withdraws, tak ing the pollen-masses with it, and visits another flower. The two anther-cases in an orchis are erect and nearly parallel the one to the other; the pollen-masses within them are of course in the like position. Immediately, however, the pollen-masses are removed, movements take place at the base of the caudicle so as to effect the bending of this stalk, bringing the pollen-mass in a more or less horizontal position, or, as in the case of 0. pyramidalis, the two pollen-masses originally placed parallel diverge from the base like the letter V. The movements of the pollen-masses may readily be seen with the naked eye by thrusting the point of a needle into the base of the anther, when the disks adhere to the needle as they would do to the antenna of an insect, and may be with drawn. Sometimes the lip is mobile and even sensitive to touch, as are also certain processes of the column. In such cases the con tact of an insect or other body with those processes is sufficient to liberate the pollen often with elastic force, even when the anther itself is not touched.
In other orchids movements take place in different ways and in other directions. The object of these movements will be appre ciated when it is remembered that, if the pollen-masses retained the original direction they had in the anther in which they were formed, they would, when transported by the insect to another flower, merely come in contact with the anther of that flower, where of course they would be of no use ; but, owing to the di vergences and flexions above alluded to, the pollen-masses come to be so placed that, when transplanted to another flower of the same species, they come in contact with the stigma and so effect the pollination of that flower. The adaptions of orchid flowers to fertilization by insects are exceedingly numerous and in many cases are remarkably complicated.
Propagation and Growth.—The fruit of orchids is a capsule which usually splits by three lengthwise slits, forming valves that remain united above and below. The seeds, minute and innumer able, are well-adapted to wind-dissemination. In many species the seeds lose their viability after a few months, and often are slow and difficult to germinate after planting, some requiring from three months to two years. The roots of terrestrial orchids are often
bulbous and still more frequently more or less tuberous, the tubers being partly radical and partly budlike, so that propagation of new individuals by division from the parent takes place. Often there is a marked alternation in the production of vegetation and flowering shoots; sometimes the flowering shoots are not produced for several years in succession. This accounts for the profusion with which various orchids are found in flower in some seasons and for their scarcity in others.
Tropical orchids are mostly epiphytal—that is, they grow upon trees without deriving nourishment from them. They are fre quently provided with "pseudo-bulbs," large solid swellings of the stem, in the tissues of which water and nutritive materials are stored. They derive this moisture from the air by means of aerial roots, developed from the stem and bearing an outer spongy structure, or velamen, consisting of empty cells kept open by spiral thickenings in the wall; this sponge-like tissue absorbs dew and rain and passes it on to the internal tissues.
Classification.—In number of species the orchid family is ex ceeded by only two or three other families of flowering plants. Conservatively stated, it contains at least 7,50o species comprised in 450 genera; some authorities place the number of species as high as 15,00o.
The family is divided into two main groups based on the num ber of the stamens and stigmas. The first Pleonandrae, has two or rarely three fertile stamens and three functional stigmas. It contains two small genera of tropical Asia and Africa with almost regular flowers, and the large genus Cypripedium containing about 3o species in the north-temperate zone and tropical Asia and America. In Cypripedium two stamens are present, one on each side of the column instead of one only at the top, as in the group Monandrae, to which belong the remaining genera in which also only two stigmas are fertile.
The Monandrae have been subdivided into 2o tribes, the char acters of which are based on the structure of the anther and pollinia, the nature of the inflorescence, whether terminal or lateral, the vernation of the leaf and the presence or absence of a joint between blade and sheath, and the nature of the stem. The most important are the following: Ophrydeae, terrestrial orchids, mainly north temperate, includ ing the British genera Orchis, Aceras, Ophrys, Herminium, Gym nadenia and Habenaria.
Neottieae, also terrestrial, contains 13 more or less widely dis tributed tropical or subtropical subtribes, one, Cephalanthereae, which includes the British genera Cephalanthera and Ekipactis, is chiefly north temperate. The British genera Spiranthes, Listera and Neottia are also included in this tribe, as is also Vanilla, the elongated stem of which climbs by means of tendril-like aerial roots.