To give an idea of ordinary farm irrigation, I will describe the conditions existing on a farm that I inspected. The system employed was undoubtedly injurious to the crops. The orange trees were growing in the furrows, and were naturally in an unhealthy condition. The young shoots were dying off, and the leaves turning yellow, altogether showing signs of improper irrigation. The water should have been kept well away from the base of the trees, so as to induce the roots to spread. It is a rule amongst advanced irrigators never to irrigate within the area covered by the shade of the tree at noon; but the irrigation of fruit trees on that farm is limited to this area. Consequently, all the roots are confined to a restricted space close to the trunk of the tree, and, should no irrigation be possible during a drought, many of the trees would undoubtedly perish. Then, again, it is the custom on that farm to irrigate twice a week, using comparatively small quantities of water for orchard irrigation this is a fundamental mistake. The small crop of oranges on the mature trees was ample proof that the system of cultivation was radically wrong. It is also a great mistake to plant fruit trees in the water-logged soil near the furrows. have seen several instances of diseased trees caused by this, and one of the leading farmers of the Rustenburg district told me he had noticed the same thing. A friend of mine in Kimberley once expressed his surprise that, although he irrigated his trees every day, he never got any fruit. That was a natural result of overirrigation, and if scientific irrigation were better understood there would be fewer disappointments.

I

Now I will give an example of better methods. It is in the Barberton district, and the farmer in this case has one of the healthiest looking orange orchards that I have seen in the Transvaal. He grows nothing but trees in his orchard, and there are neither grass nor weeds. The surface-soil is broken into a loose tilth which prevents the sub-soil drying up, and when he irrigates, the water is allowed to soak into the ground between the trees. This he only does twice a year during the winter, and after irrigating the surface is well cultivated. There were no symptoms of root-rot and no leaves turning yellow. The farmer knew that the roots extended over twenty-five feet from the tree, and it was those roots he wanted to water. I saw another notable instance on a farm in the Rustenburg district where the water supply had practically ceased; but there again the surface was covered with a fine tilth, and under the branches, where the plough and harrow could not be used, the soil was broken up with a fork. The correct practice is to apply heavy irrigations at as long intervals as possible, depending on the nature of the soil and the kind of tree, and the distance from the trunk at which the water should be applied varies with the size of the tree. This perennial variation induces the roots to spread themselves out towards the moistened soil, producing a larger area from which the tender roots can draw nourishment and moisture. If copious irrigations are used, the trees become deep-rooted and will safely withstand any ordinary drought.

I have seen many farms where the field furrows have been laid out on far too steep a gradient-so steep that an appreciable quantity of soil is being washed away at each irrigation. This, of course, is quite wrong, for it not only carries away some of the best soil, but also cuts up the fields very badly. Then, too, the beds, or acres (as they are sometimes called by the Dutch farmers), are not properly prepared. It is not enough to plough them and turn in the water: they should be carefully levelled, so that the water will spread gradually over the surface. Where the furrows and beds have a gentle slope the water has time to sink into the soil, which it does not do to the same extent when rushing over the ground. One has only to examine the lower edge of the fields to see the waste of water and land that is constantly occurring.

One of the most important crops in the Transvaal is tobacco, and there again, in my opinion, the system practised is harmful. All the plants are grown in the furrows, and in the early stages the young plants are often entirely submerged during the process of irrigation, and the leaves covered with a deposit of silt, which prevents them from performing their natural functions. Little or no attempt at cultivation or inter-tillage is practised, and naturally the ground becomes hard and baked, thus supplying the conditions favourable to evaporation. I have seen young tobacco seedlings, as tender as water-cress, struggling through the caked surface of a dry furrow. Then they irrigate to soften the ground, and water is applied too frequently. With proper deep irrigation it would be quite unnecessary to irrigate every two or three days, as is the usual practice at present. The young plants undoubtedly require an abundant supply of moisture; but the larger plants, if well rooted, as they would be with deep irrigation, ought seldom to require watering more than once in ten or fourteen days. It is a common thing to see the lower leaves attacked by a white mould. On some farms it amounts to from 20 per cent. to 30 per cent. of the leaves, and those that are badly affected can only be used for Kaffir tobacco. One of the best known growers in this Colony agrees with me that the mould is largely due to the system of irrigation practised. He also agrees with me that the tobacco plants should be grown between, and not in, the furrows; but he was afraid it would take more water and more work. I pointed out that by this system fewer irrigations would be necessary, and, consequently, less work leading water. I admit it would require more care, but the tobacco planter would be amply repaid by a better crop of leaves. It is the opinion of experts that a lighter and thinner leaf would be grown if the tobacco plants were placed together more closely in the rows; but this would cause more shade, and some growers are afraid there would be an increase of the white mould. This would probably be avoided by a better system of irrigation, but such questions will only be definitely settled by expert observation on experiment irrigation farms.

The amount of tobacco of all sorts imported into the Transvaal during 1904 was £202,575, and in 1905 it increased to

£243,702. Now, if this valuable local trade is ever to be retained in our own hands, the tobacco growers must produce larger quantities of improved leaves of uniform quality. It is quality, and not weight, that is going to pay in the future.

Many people in South Africa appear to think that irrigation consists entirely of the works necessary for storing, or diverting, water and delivering it to the farmers; but while I fully realise the importance of the work that has to be done by Irrigation Engineers, I am also aware that real development does not rest solely with them. The economical and scientfic use of the water by farmers is a very important factor, and that is the point to which I would like to give special emphasis. The methods employed to-day are little better than those employed by the early settlers. It has been the same in the United States, but the go-ahead people there are alive to the importance of the proper use of water, and the question is being investigated by experts. It has been found that the area of land that could be irrigated by the rational use of water might be doubled or trebled, and it is easy to see what an advantage that would be to this country. It would mean that the cost per acre for water would be reduced to one-half or one-third. This saving of water is obtained by only using the amount required to get the best results. Different crops require different treatment, and improper irrigation is injurious to any crop; for instance, oats require more water than any other grain, cotton needs very little, too much water spoils tobacco, turnips, parsnips and carrots should never be flooded, and flooding lucerne during the first six months of its life is pretty certain to check the growth.

Much valuable information was given by Dr. J. Widstoe in a paper read before the Twelfth National Irrigation Congress in the United States of America. He said that on a typical western soil. 5 inches of water produced 33 bushels of wheat per acre; 10 inches of water produced 40 bushels. Adding more than 20 inches of water to the field did not appreciably increase the yield of grain. In the case of the wheat plant, then, the increase of the amount of water up to about 15 inches increases the yield of grain, but a further application tends to diminish the yield. When it is considered that the depth of water applied to the wheat, over a very large area of irrigated country, amounts to 30 or more inches annually, it will be understood what a loss in wheat alone occurs year after year through the misuse of water.

With oats the variation is somewhat similar. Five inches of water in one set of experiments yielded about 58 bushels of oats per acre; 10 inches yielded about the same amount, though the increased quantity of water increased the weight of the straw. With 15 inches 70 bushels were obtained; with 20 inches 86 bushels; and with 30 inches 82 bushels. Covering the land with more than 30 inches of water diminished the yield of oats decidedly. inches is therefore the best amount of water for oats; yet throughout the irrigated West 30 inches or more are generally used in the

About 20

production of oats-and thus again the wasteful use of water is emphasised.

It is generally true with all the ordinary crops grown in the Western States, that increasing the amount of water increases the yield up to a certain point, after which an increase in the water causes a decrease in the yield. Not all the crops are alike in this respect, however. Some crops, because of their nature-leaf surface, root system, etc.-find 10 inches of water about right for the season's growth. Other plants, because of their different natures, find 15 inches or 20, or 25 to 30 inches the best. Now, the farmer in an irrigated district should know the water requirements of the different plants that he grows as thoroughly as he knows the soil of his farm, his water right, or any other matters upon which his success as a farmer depends. Not all plants decrease in yield after a certain amount of water has been applied. Potatoes appear to be a crop, the yield of which increases continually if water is applied, up to the limit of the practical application of water. To illustrate-in one set of experiments, 7 inches of water produced 160 bushels of potatoes; 15 inches 233 bushels; 30 inches 274 bushels; and 71 inches 315 bushels. This illustrates the necessity for the farmer to thoroughly understand the nature of the plants with which he is dealing.

It may be noted, in reviewing the yields of wheat, oats, and potatoes just considered, that the value of the first few inches of water applied is much greater than that of the later applications. For instance :-5 inches of water produced about 33 bushels of wheat, or about 6.6 bushels per inch; 15 inches of water produced about 40 bushels of wheat, or about 3.2 bushels per inch of water; while 20 inches of water also produced 40 bushels of wheat, or only two bushels per inch. The value of the first 3 inches of water applied to wheat, therefore, is more that three times as much as the value of the last 5 inches, in a total depth of 20 inches. Similar results may be observed in the case of oats. Five inches of water produced 58 bushels, or 12 bushels per inch; while 20 inches of water produced 86 bushels of oats, which is less than 5 bushels per inch of water. The difference is certainly very striking. Even in the case of potatoes, the yield of which increased steadily with the increase of irrigation water, the same fact holds. Seven and one-half inches of water produced 160 bushels, or about 22 bushels per inch; while 30 inches of water produced 274 bushels, which is only about nine bushels per inch. Corn, alfalfa (lucerne), the various grasses, sugar beets, vegetables and all other crops show similar results; namely, that the value of water is highest when it is used sparingly and carefully; that the value of water is lowest when it is applied liberally and carelessly. With this generalisation in mind, note how these results may be viewed in their relation to the increase of the irrigated area.

According to the investigations of the Department of Agriculture, under the direction of Dr. Mead and his associates, 30

inches of water, or more, are used in the majority of places in the irrigated districts for the production of crops. Let us apply the varying value of water as just explained, to the economical, or rational, use of water. If the 30 acre-inches be spread over 6 acres of wheat, so that the whole area of 6 acres will be covered with water to a depth of 5 inches, each acre will yield 32 bushels of grain, or a total of 195 bushels. If the same amount of water be spread over 4 acres, that is, to a depth of 7 inches, the total yield of grain will be 165 bushels. Spread over 3 acres, to a depth of 10 inches, the same amount of water will yield 118 bushels. Spread over 2 acres, to a depth of 15 inches, the total yield will be 95 bushels, and spread over 1 acre, to a depth of 30 inches, the yield will be 42 bushels. It may thus be seen that, in the case of wheat the total amount of grain produced by 30 acre-inches of water may be increased from 42 bushels to 195 bushels by spreading the water over more or less ground. Certainly the nearly five-fold increase of grain thus made possible, will more than pay the farmer for the labour of handling six acres of land instead of one; and of higher importance is the fact that, by using the water rationally, the irrigated wheat area may be profitably increased four or five times without building another reservoir or canal.

These figures show how enormously a farmer can increase the area of land irrigated by a given quantity of water if he understands its scientific use. It is a subject worth studying, for on it depends to a large extent the agricultural prosperity of the country. Farm irrigation in South Africa has not received the attention it deserves, and I am looking forward to the time when the Transvaal will have Government experiment irrigation stations in various districts, under the superintendence of expert engineers and agriculturalists. They will be attended with the most valuable results.