It has long been a matter of common observation that many plants, with their stamens and pistils in separate flowers, whether monoecious or dioecious, depend upon insects for their fertilization; the insects, in their visits from flower to flower in search of honey, evidently bring the pollen from the anthers of the staminate to the stigmas of the pistillate flowers, as every gardener knows who has tried to keep his varieties of melons and other plants of the same family in a condition of purity. Darwin, in his work " The Fertilization of Orchids " (London, 1862), showed that many perfect flowers, with their pistils and stamens in close proximity, are so constructed that the pollen can never without extrinsic aid reach the pistil of its own flower, or serve to fertilize that of any other flower. He showed that there is a complete provision that the flowers of many orchids should never be fertilized by their own' pollen - in other words, that in-and-in breeding is impossible - and arrived at the conclusion that cross fertilization is necessary among flowers of the same species, in order to promote the greater vigor of the whole, and to prevent the perpetuation and fixation of individual peculiarities.

This subject has since received the attention of many close observers, and many interesting facts have been established. It has been found that in perfect or hermaphrodite flowers, which were supposed to be especially arranged for self-fertilization, the stamens and pistils, while close together, are really so placed with reference to one another that self-fertilization is almost impossible, and unless insects come to their aid no seed will be produced. It is another interesting point that flowers needing the aid of insects are supplied with an attraction for them in the form of nectar. In most flowers the pollen is a fine dust which readily adheres to whatever it touches, and the contrivances to prevent the pollen of a flower from reaching its own pistil are no more varied and interesting than those which insure that a bee or other insect shall come in contact with this pollen and carry it off upon its body. To illustrate this, an engraving (from Gray) of a section of the flower of a common iris or flower-de-luce will serve.

In this there are three recurved outer petals and three erect inner ones, united below in a tube; the pistil has three styles, which are broad, petal-like, and notched at the top; each style bears just below the notch a stigma, which is a thin plate projecting like a little shelf, the upper side only of which is stigmatic, or capable of receiving pollen; below each stigma is an anther, as seen in the engraving. Here the stamen and pistil are not only in the same flower, but in actual contact; yet while everything appears favorable for the pollen to reach the stigma, it cannot do so of itself, for the anther or pollen case has the openings through which it discharges pollen turned outward, and the stigma is above and its receptive portion turned away from the anther. When an insect visits this flower, it can only get at the honey by crawling in under the petal-like style, and in so doing, as well as in making its exit, it must rub against the anther and become dusted with pollen; when the insect thus charged with pollen goes to another flower, it must, in its attempts to get under the style to reach the honey, dust some of its pollen upon the upper surface of the shelf-like stigma.

To explain the many ways in which this plan is varied would require a volume; the curious modifications of it in only a portion of a single family, the orchidaceae, fill a book of over 350 pages. The common barberry has irritable stamens, and the laurel (Kalmia) has its stamens bent like a spring, the anthers being caught in notches in the corolla, but when disturbed the bent filaments spring toward the pistil with force. These were both regarded as admirable contrivances for bringing the pollen in contact with the stigma, but closer observation has shown that these sensitive stamens scatter their pollen rather on the insect which irritates them than on the pistil of their own flower. In order to insure cross fertilization in many cases, the stamens discharge their pollen before the pistil is sufficiently mature to receive it, or vice versa. Dichogamous is the term adopted to express this unequal perfection of stamens and pistils, of which the common plantain (plantago) affords an illustration. Flowers alike in all other respects are often dimorphous as to their stamens and pistils; primroses in the garden and the wild bluets (Houstonia) so common in spring, furnish examples. In these flowers the corolla has a long tube and an expanded limb.

In some of the flowers the stamens appear at the throat of the corolla, while the style is very short; in others the style projects, and the stamens are low down in the tube. In either case self-fertilization is not likely to occur, but an insect with a long proboscis has an ample opportunity to effect cross fertilization. There are also trimorphous flowers, in which there are three kinds of stamens and styles as to length, admirably adapted to secure service from the different kinds of insects that visit them. Sometimes the insects suffer from the performance of this service; the pollen of the milkweeds (asclepias) is in waxy, pear-shaped masses, joined in pairs with an adhesive attachment to the little stalk which unites them. Did not the legs of the insects in search of nectar come in contact with this adhesive attachment, and by its means draw the pollen masses from their pouches, the pollen could never reach the stigma; it often happens that bees have their legs so encumbered by these adhering pollen masses, that when they reach the hive they are unable to climb upon the comb, but fall down and perish on the bottom of the hive.

Some interesting experiments show that in many flowers where there is evident provision for self-fertilization, the stigma can hardly be impregnated by pollen from the same flower, while it will readily accept that produced by another flower, and that the agency of insects is absolutely essential to the production of seed in quite perfect flowers. - Excellent memoirs on this subject have appeared in "Nature," the "American Journal of Science and Arts," and the "American Naturalist." A useful resume for young people is given in Prof. Gray's "How Plants Behave" (New York, 1872). The work of Darwin already referred to is admirable as showing how such investigations should be conducted.

Section of Iris.