Nairul (Eugenia Jambolana).  This is a good shade tree.  Coffee thrives well under it, and wherever it exists, or may have sprung up accidentally in the plantation, it should be preserved, but it is not, I consider, a desirable tree to plant, as it is a slow grower and not a wide spreader.

Wartee.  This is a tree we have always preserved, but it is a slow growing tree, not at all a wide spreader, and the leaf deposit from it is not of a valuable quality, and it is, therefore, not a desirable tree to plant.

Gwoddan (Dolichos fabaeformis).  Coffee thrives well under this tree, but it has a great profusion of very hard fruits or seeds about the size of a small plum, and these, when falling from a high tree, injure the coffee berries, as may be readily supposed; the tree, too, is not a wide spreader.  It is, therefore, not a desirable tree to plant.

I may mention here that I have recently obtained a supply of seed of Albizzia Moluccana, which is the tree most approved of for shading coffee in the Island of Java, and I am informed by the superintendent of the Agri-Horticultural Society’s Gardens, Madras (from whom I obtained the seed), that one of their correspondents who tried it some years ago reports that, “It grows rapidly, and is of great utility in putting a field of coffee under a light shade such as coffee likes,” and that, “in four years the Albizzia Moluccana, planted thirty feet apart, will cover the coffee trees.”  The leaves close during the night, thus giving the coffee plants the benefit of the moonlight and dew more freely.  Each ounce of the seed contains roughly 1,200 seeds, which, with ordinary care, should give 1,000 plants, and which, when planted out thirty feet apart, should shade twenty acres.

I now proceed to consider the methods that are adopted for planting under shade in Mysore.  The first is to clear down and burn the entire forest, and then plant shade trees along with the coffee.  The second is to clear and burn the underwood, and a certain portion of the forest trees, leaving the remainder for shade, and the third is (a system which I have myself adopted in the case of land lying in ravines) to clear off and burn the entire underwood and trees of the lower part of the ravines, leaving the upper portions of them, and the remainder of the land to be cleared and planted, under the original forest trees, as in the second method mentioned.

There can be no doubt that the first-named method is the easiest.  I am aware that it has been adopted by some very experienced planters, and it has been partially adopted by myself in the case of all my land in the lower part of ravines.  I am well able to judge of the advantages and disadvantages of both systems, as I have them under observation and treatment side by side.  On the whole, I think there can be no doubt that the balance of advantage lies much in favour of land that has not had the forest cleared