G. L. Chauveaud 1 has advanced the theory that polyembryony is a more primitive state than monoembryony, which would seem to be borne out by this observation; for it must be true that the choice mangos of India which have been propagated by grafting for centuries are less primitive in character than the semi-wild seedling races.

Inarching is an ancient method of vegetative propagation. While several writers have attempted to show that it was not known in India previous to the arrival of Europeans, and that the Jesuits at Goa were the first to apply it to the mango, others have held the belief, based on researches in the literature of ancient India, that the Hindus propagated their choice mangos by inarching for centuries before any Europeans visited the country.

This method of propagation is still preferred to all others in India and a few other countries. In the United States it has been superseded by budding.

For the production of stock plants on which to bud or graft choice varieties, seeds of any of the common mangos are used. No preference for any particular race has yet been established. It is reasonable to believe, however, that there may be important differences among seedling races in vigor of growth and perhaps in their effect on the productiveness and other characteristics of the cion. The subject has never been investigated and deserves attention.

Seeds are planted, after having the husk removed, in five- or six-inch pots of light soil or in nursery rows in the open ground. They are covered with 1 inch or 1 1/2 inches of soil. In warm weather they will germinate within two weeks, and must be watched to prevent the development of more than one shoot. Polyembryonic mangos will send up several; all but the strongest one should be destroyed. If grown in pots and intended for budding, the young plants may be set out in the field in nursery rows when they are a foot high. If destined for inarching they must be kept in pots.

1 Compt. Rend. 114, 1892.

Inarching is more successful in the hands of the tyro than budding or crown-grafting. It can be recommended when only a few plants are desired, and when the tree to be propagated is in a convenient situation. G. Marshall Woodrow thus describes inarching as it is done in India. A slice is cut from the side of a small branch on the tree it is desired to propagate, and a slice of similar size - 2 to 4 inches long and deep enough to expose the cambium - is cut from the stem of a young seedling supported at a convenient height upon a light framework of poles. The two cut surfaces are bound together with a strip of fiber from the stem of the banana, or with some other soft bandage. Well-kneaded clay is then plastered over the graft to keep out air and water. The soil in the pot must be kept moist. After six to eight weeks the cut surfaces will have united.

Inarching may be done at any time in strictly tropical climates, but the best time in the hot parts of India is the cool season. Toward the northern limits of mango cultivation the middle of the rainy season is better.

The graft is sometimes allowed to remain attached to the parent tree for too long a time, with the result that swellings, due to the constriction of the bandages, occur at the point of union. It is better to remove the grafted plant fairly early and place it in the shade for a few weeks. It is detached from the parent tree by severing the branch which has been inarched to the seedling at a point just below the point of union with the latter. This leaves the young branch from the tree it was desired to propagate growing upon a seedling; the top of the latter is cut out, and the branch from the old tree takes its place, ultimately forming the crown of the mature tree.

The age of the stock is not important. Plants three weeks to three years old have been used with success. If kept in pots too long, however, the plants become pot-bound and lose their vigor; hence it is desirable to graft them when young and get them into the open ground as soon as possible. Seeds planted in June and July make strong plants ready for inarching by November. December and January are good months in which to inarch, and such plants should be ready to set out in the field by the following July.

Inarching, as practiced in other countries, differs in no essentials from the Indian method above described.

Shield-budding is the method employed by nurserymen in Florida. In the hands of a skillful propagator who has made a careful study of this method, it gives excellent results. In inexperienced hands it usually proves altogether unsatisfactory. Particularly is experience required to enable the propagator to recognize the proper type of bud wood, and to know when the stock plants are in the proper state of vegetative activity. By careful experimenting with stock plants and budwood of different conditions of growth throughout a season or two, a good propagator should be able to bud mangos successfully; but comparatively few men have yet devoted the requisite time and study to the subject. Thus there are at present only a few propagators in the United States who can produce budded mango trees economically and in quantity.

Fig. 11. Shield budding the mango. On the left, a bud properly inserted; next, an inserted bud wrapped with a strip of waxed cloth; above the knife point, a properly cut bud; and on the right, budwood of desirable character.

Fig. 11. Shield-budding the mango. On the left, a bud properly inserted; next, an inserted bud wrapped with a strip of waxed cloth; above the knife-point, a properly cut bud; and on the right, budwood of desirable character.