While the mango grows in humid tropical regions subject to heavy rains throughout the year, it is not successfully cultivated for its fruit under these conditions. It requires the stimulus of a dry season to fruit abundantly. To a certain extent this stimulus can be given by artificial means, but there can be no doubt that the best regions for commercial mango culture are those in which there is a well-marked dry season occurring at the proper time of year.

This is illustrated by conditions in India. Lower Bengal is a humid region in which moisture-loving tropical plants are completely at home. Mango trees in this region are ragged in appearance, with foliage of an unhealthy color, and the fruit does not ripen well. In sharp contrast, the trees at Saharanpur, on the dry plains of northern India, are vigorous and stocky in habit, with abundant foliage of rich green color. They fruit more profusely than those in the moist lowlands, and the fruit ripens perfectly. Saharanpur lies at an elevation of 1000 feet, and has an annual rainfall of about thirty-five inches. During the season when mangos are ripening, no rain falls and the air is hot and dry. Temperature of 100° F., continued throughout day and night, are common. The monsoon, or rainy season, lasts but a few months.

The total amount of rainfall is not so important as the season during which it occurs. Where the dry season coincides with the normal flowering time of the mango, good crops of fruit can be expected, but it seems doubtful whether the finer grafted mangos can be cultivated successfully in regions where there is much precipitation during the flowering season. Some of the seedling races will fruit under these conditions, but the choice Indian varieties are more exacting in their climatic requirements.

On this point G. N. Collins 1 states : "The fact that the tree may thrive in a given locality and yet fail to produce fruit should always be kept in mind. It may be considered as proven that the mango will be prolific only in regions subjected to a considerable dry season. On the moist north side of Porto Rico the trees grow luxuriantly, but they are not nearly so prolific nor is the fruit of such good quality as on the dry south side, and in the very dry region about Yauco and at Cabo Rojo the fruit seemed at its best, while its abundance was attested by the fact that fine fruit was selling as low as 12 for a cent. In Guatemala and Mexico the mango was found at its best only in regions where severe dry seasons prevailed."

Fawcett and Harris 2 report similar conditions in Jamaica. They say: "Although the mango grows freely everywhere, it is not a fruitful tree in every district; in the southern plains and the low, dry limestone hills it produces enormous crops year after year, and very often two crops a year, the main crop from May to August, and the second crop later in the year. . . . In humid districts and along the northern coast the tree is not at all fruitful, except in very dry years, and in the wet districts like Castleton it rarely fruits."

In the Botanic Garden at Rio de Janeiro, Brazil, there is a magnificent avenue of mango trees planted by the emperor Dom Joao VI more than a century ago. So far as known these trees have never matured any fruits. They blossom, and occasionally set fruits, but the latter invariably drop off before reaching maturity. J. C. Willis, former director of the garden, attributes this to the fact that they are planted on low wet ground.

1 Bull. 28, U. S. Dept. Agr.

2 Bull. of the Bot. Dept., vol. 8, 1901.

Other mango trees in the immediate vicinity but on higher ground produce fruit regularly.

Mangos can be grown successfully on soils of several different types. In Porto Rico deep sandy loam has given excellent results. On this soil the tree makes rapid growth and attains great size. The sandy soils of southern Florida have proved satisfactory. Clay, provided it is well drained, seems to be good.

In India, some of the best mango districts are situated on the great Indo-Gangetic plain, where the soil is a deep, rich alluvial loam. This may perhaps be considered the best of all mango soils. An analysis of surface soil from the mango orchards in the Saharanpur Botanic Garden shows that it contains:

C. F. Kinman1 says:

"A shallow soil underlain with stone or hardpan, although sufficiently deep to produce shrubs or other low-growing wild vegetation, will not satisfy the needs of the deep rooted mango, whose growth in such ground will be slow and its yield poor, at least after the first few years. The application of fertilizers, however, will materially decrease the depth of the soil required. . . . Mango trees are often found on very light, unfertile sand, which may be a few feet in depth, and still produce flourishing growth if the subsoil is suitable. As the mango, like most other fruit trees, thrives best on a deep loose loam with good drainage and a high percentage of humus, those who intend planting it commercially should secure, if possible, this type of soil."

1 Porto Rico Agr. Exp. Sta. Bull., 24.

Much more important than the mechanical or chemical composition, in most cases, is the drainage of the land. The mango avenue in the Botanic Garden at Rio de Janeiro illustrates this. If the subsoil is permanently wet or poorly drained, the tree cannot be expected to fruit profusely.

While the mango is more susceptible to frost than the hardier races of the avocado, mature trees have withstood temperatures below the freezing point without injury. In general it may be said that most varieties, if not in active growth at the time cold weather strikes them, will withstand 28° or 29° above zero, provided such temperatures are not of long duration. Young trees in vigorous growth may be injured seriously by a temperature of 32°. At Miami, Florida, five-year-old trees of one or two varieties were killed outright by a freeze of 26.5°. Old seedling trees have gone through temperatures lower than this without losing more than the smallest branches. The cultivated kinds show slight differences in hardiness. Observations have been made at Saharanpur and lists drawn up showing the relative susceptibility to frost of many varieties. The vagaries of the 1917 freeze in southern Florida, however, have resulted in an impression that such lists are not altogether dependable, and that much depends on local conditions, the physiological state of the tree, and other factors as yet not understood.