I need hardly add that in the case of all steep parts of a plantation all manure should be, if not buried deeply, at least covered with soil after the digging of a trench large enough to contain the manure. On the plantations on the Nilgiri Hills the manure is put into pits 2½ feet long, 1 foot 6 inches wide, and 1 foot deep on the lower side of the pit, which of course would make the side of the pit on the upper side of them much more than one foot in depth. The trenches or pits are dug across the slope and in front of each coffee tree, and in the line (i.e., not in the centre of each set of four plants). These pits are not filled up to the brim, but the manure is placed in the bottom of them, and is then covered with soil, so that the pit is about one-half filled up. The soil taken out is heaped in a curve above the pit so as to prevent heavy rain washing down into the pit. When more manure is required to be added—say bone-meal—it is scattered on the soil in the pit, or the top soil in it is scraped off and the manure scattered and then covered up.

I now propose to consider our manurial resources in detail, and shall begin with the first stay of all agriculture, farmyard manure, as to the value of which for coffee I have never met with any difference of opinion. But there are many objections to relying on farmyard manure, or, at least, to applying it on a large scale, as, if the planter keeps many cattle of his own, he runs great risk of his herd being invaded by disease, and the difficulty and expense of feeding a large number of cattle is very considerable. In some cases it is possible to hire cattle from the natives, and this is done occasionally, and at the rate of 15 rupees a month for 100 head, but here again risk from disease is often incurred, and if it broke out, the natives would withdraw their cattle. The question then naturally arises whether, considering the great cost and trouble attendant on manufacturing cattle manure on a large scale, we cannot find some substitute that would diminish the quantity now required. And here it is important to ask what farmyard manure consists of. It consists, then, of the excreta of animals, and the vegetable matter used as litter. From a chemical point of view it mainly provides, in addition to the organic matter, in a slowly-acting form, lime, nitrogen, potash, and phosphoric acid, and from a physical point of view it furnishes a padding to maintain the texture of the soil, or, in other words, to keep it in a loose and friable condition. And with reference to this last very important point, I may remind the reader that Sir John Lawes has well pointed out that "All our experiments tend to show that it is the physical condition of the soil, its capacity for absorbing and retaining moisture, its permeability to roots, and its capacity for absorbing and radiating heat, that is of more importance than its strictly-speaking chemical composition." Now as regards the chemical aspect of the manurial question, if we assume, as we have every reason to do from the small quantity of potash required, and its supply from decomposing stones in the land, that the potash does not require to be taken into account, we shall find that our nitrogen and phosphoric acid can be far more cheaply supplied by fish, or by a mixture of bone-meal and oil-cake than by farmyard manure, and should it be found that potash does require to be added, we could obtain it more cheaply from ashes or kainit. Then in order to provide the padding that farmyard manure supplies, and to furnish nitrogen in a slowly-acting form, we could collect dry leaves, twigs from jungle trees, ferns, and any other available vegetable matter, form them into a compost with some earth, or jungle top soil, and apply the mixture to the land. With such a compost as I have suggested, bone-meal or fish-manure in small quantity might be mixed, and we should then have a very good substitute for all the chemical and physical advantages to be derived from the very best kind of farmyard manure. But there is another way of arriving at the same end, which is open to many planters, and that is by collecting top soil from the fringe of jungle commonly left round the plantation, or from the uncultivated jungle of the estate, or from adjacent pieces of jungle land. And such pieces of land varying from ten to twenty acres can commonly be purchased, and can be used to supply top soil. This, of course, has in it much vegetable matter in various stages of decay, and a mixture of it with a small quantity of bone-meal would form a manure superior, as I shall afterwards show when I come to treat of top soil, to farmyard manure chemically, and superior to it from a physical point of view. To such local manurial resources I would call particular attention, as planters have hitherto relied far too exclusively on cattle manure, and imported manures, such as bones, fish, and oil-cake, and it is evident that we could dispense with much of all these manures if we made a full use of the resources I have recommended. In concluding my remarks on cattle manure I may observe that it is both costly to supply and to apply to the land. It is difficult, of course, to make exact calculations on the subject, as the facilities for supplying litter vary so much, but generally speaking it costs from 70 to 80 rupees an acre if we manure at about the rate of a third of a bushel per tree.

I now turn to a consideration of the value of jungle top soil, a manure to which I have only lately given particular attention, though I was, of course, well aware of its value in a general way, and may begin by stating that two samples of what we were using on my estates have been analyzed by Dr. Voelcker, the object being partly to ascertain the value of the soil and partly to compare its cost with the cost of cattle manure. After estimating the cost of making cattle manure, and calculating as closely as possible the cost of obtaining and applying jungle top soil from land adjacent to the plantation, it was found that in the case of the best sample of top soil, obtained by removing only four or five inches of the soil, it paid nearly twice as well to use it as a manurial agent as it would to use cattle manure, and I may add that three tons of the soil contain the same manurial matter as two tons of ordinary well-made English farmyard manure. In the case of the second sample analyzed, and which I was sure from the character of the land must be of inferior quality, it was found that 2¼ tons of the soil would contain as much manurial value as one ton of farmyard manure, and that the cost of using the two materials would be about the same.

I had also analyzed at the same time a sample of a kind of decayed pink-coloured rock, as I had found that coffee had thriven well in the pink soil which had evidently been formed from the rock in question, but the manurial value was so small that Dr. Voelcker thought that it might merely be of use in improving the physical condition of the soil. I however applied it to some backward coffee, and also applied some of the best top soil to a contiguous piece of backward coffee, and was much surprised to find that the pink soil, to which little direct manurial value was attached by Dr. Voelcker, showed results superior to the best top soil applied alongside of it, and I am now applying it on a large scale. This soil, I may mention, is applied by the natives to the surface of their vegetable beds. They do not attach any manurial value to it, but apply it to keep the vegetables cool, as the soil has quite a remarkable effect in keeping itself cool while the adjacent soil is quite hot, and I have now applied it to the flower beds near my house, and also to the walks around the bungalow. This pink decayed rock is sometimes streaked with a white decayed rock, which the natives call jadi mannu, and sometimes the latter so much preponderates that it looks nearly white. I am told by the natives that if you mix the red and white earth together and apply the mixture to the surface of the land it will never get dry.[[54]]

In concluding my remarks on soil applications, I may observe that if top soil costs the same price as cattle manure, the former is to be preferred for four reasons. It is much more easily handled and applied; it is a better substance for mixing with other manures, such as bonedust or ashes, for instance; it has a better physical effect on the soil; and is nearly free from weed seeds which abound in cattle manure.

I may add that I have since made a calculation with the object of seeing how, by the addition of manures to the kemmannu soil, I could make a mixture which would have all the fertilizing ingredients of farmyard manure in addition to the advantages possessed by the soil, and which I have just enumerated. I find that if to 83 parts of the soil I added 1 part of bonedust, 12 parts of castor cake, 2 parts of potash salt, and 2 parts of lime, I should make up a compost equal to good English farmyard manure, and at but a slightly increased cost, which would be more than covered by the special physical and other advantages arising from the use of kemmannu.

The pulp of the coffee is very apt to be carelessly treated, and it is important to remember that Mr. Hughes, in his "Report on Ceylon Coffee, Soils, and Manures," estimates that, if properly preserved, two tons of pulp are equal to one ton of good farmyard manure. But it must not be washed, as it often is by being run into a pulp pit with water, or nearly all its valuable constituents would be lost. It should be mixed, he tells us, with cattle dung, or, if that is not procurable, with liberal supplies of lime, and he also suggests that it should be put under cover day by day. We have adopted on my property a plan which I think in these climates is the cheapest and best. A layer of top soil is placed in the road alongside of the coffee where we desire to use the manure; then each day's pulp is carted direct to the plantation and scattered over the top soil, and more top soil added, till we have a layer as thick as we find convenient, but of course not so thick as to prevent carts passing over it to other parts of the plantation. On these layers of pulp and top soil lime or bonedust may be sprinkled.

Dry fallen leaves is another local resource which should by no means be neglected, and they are commonly used for littering the cattle sheds. Such leaves are about equal to cattle dung. A sample of those we use was analyzed by Dr. Voelcker, and the result gave 1 per cent. of phosphate of lime, 1 per cent. of ammonia, and ¾ per cent. of potash.

Green twigs[[55]] cut from jungle trees are of considerable manurial value, and the natives seem well aware of the value of the different kinds. A sample of the following six kinds which are most approved of by the natives—namely, Japel, Nairal, Ubble, Gowl, Mutty and Hunchotee, was analyzed by Dr. Voelcker, and the result gave ¼ per cent. phosphate of lime, ¾ per cent. of potash, 1 per cent. of lime, and ¾ per cent. of nitrogen.

Ferns are of considerable manurial value, and are rich in potash, and they should be used to litter the cattle sheds.