Farming and Gardening for Health or Disease
(The Soil and Health)
by Sir Albert Howard C.I.E., M.A.
The Diseases of Crops
Disease in crops manifests itself in a great variety of ways. Troubles due to parasitic fungi and insects are by far the most common. Many of these troubles have occurred from time to time all through the ages and are by no means confined to modern farming. In recent years attention has been paid to a number of other diseases, such as those due to eelworm, to virus, and to the loss of the power of the plant to reproduce itself. The varieties of our cultivated crops nowadays show a great tendency to run out and to become unremunerative. This weakness, which might be described as varietal-erosion or species-erosion, has to be countered by the creation of a constant stream of new varieties obtained either by plant breeding methods or by importation from other localities. Besides the many cases of running out, failure to set seed is also due to unfavourable soil conditions, the removal of which puts an end to the trouble.
The great attention now devoted to disease will be clear from the operations of the Empire Cotton Growing Corporation, a State-aided body incorporated by Royal Charter on 1st November 1921 for the development of cotton production in the Empire. Among the many activities of this Corporation is the publication of the Empire Cotton Growing Review, a feature of which are the notes on current literature. During the six years before the war, 1934-9 these abstracts of papers on cotton research cover 964 pages of print, of which no less than 223, i.e. 23 per cent, deal with the diseases of cotton. These figures roughly correspond with the way the money contributed all over the world for the production, improvement, and testing of new cottons is spent. Some quarter of the technical staff engaged in this work devote their whole time to the study of the diseases of the cotton plant.
That something must be wrong with the production of cotton throughout the Empire and indeed throughout the world is suggested by a comparison between the above alarming figures and my own experience at the Institute of Plant Industry at Indore in Central India, at which research centre cotton was the principal crop. Between the years 1924 and 1931 cotton disease at Indore was to all intents and purposes negligible. I can recall only one case of wilt on some half dozen plants in a waterlogged corner of a field in a year of exceptionally high rainfall. The cotton plant in India always impressed me as a robust grower capable of standing up well to adverse soil and weather conditions. The examples of disease I came across in my many tours always seemed to be a consequence of bad farming, all capable of elimination by improved methods of agriculture.
As my adventures in research began in the West Indies in 1899 as a mycologist, I have naturally followed very closely the subsequent work on the various diseases of crops and have always been interested in the many outbreaks of these troubles which have occurred all over the world. Since 1905 I have been in a position to grow crops myself and thus have been able to test the validity of the principles on which the conventional methods of disease control are based. Perhaps the simplest way of dealing with these experiences, observations, and resections will be crop by crop.
In perusing the following pages one thing will strike the reader forcibly. I have found it impossible to separate the disease from the growing crop. The study of plant diseases for their own sake is proving an increasingly intricate game, to which modern scientists have devoted many wasted hours. Such studies would be amusing if they were not tragic, for no disease in plant, animal, or man can properly be viewed unless it is looked on as an interference with, or, to speak more plainly, as the distortion or negation of that positive aspect of the growing organism which we call health.
Consequently it is essential to conceive of the plant, for instance, as a living and growing thing, flourishing in certain conditions but wilting or perishing in other conditions; in any discussion of plant disease the right and the wrong methods of growing the crop are not simply the background to the argument, they are its very substance: to investigate plant diseases without a first-hand experience of growing the plant is to play Hamlet without the Prince of Denmark.
While in the West Indies (1899-1902) I devoted much attention to the fungous diseases of sugar-cane, but only succeeded in writing a few routine papers on the subject, all of no particular importance. Some twenty-five years later at Indore I grew a number of excellent crops of cane and converted them into crude sugar, both of which proceedings won the approval of the local Indian population. This experience brought out one of the weaknesses in present-day research. Between the years 1899 and 1902 I could only write technical papers on the diseases of the cane, as I had no opportunity of growing the crop or of manufacturing it into sugar. I was then in the straitjacket stage of my career. It was not till a quarter of a century later in another continent that the chance came to grow sugar-cane, to the study of whose diseases I had devoted so much attention. It is safe to say that, had these periods been reversed, my papers on the fungous diseases of cane would have made very different reading.
The methods adopted in growing sugar-cane on the black cotton soils at Indore were a copy of those devised by the late Mr. George Clarke, C.I.E., at the Shahjahanpur Experiment Station and described in detail in Chapter XIV of An Agricultural Testament. The crop is planted in shallow trenches, two feet wide, four feet from centre to centre, the soil from each trench being removed to a depth of six inches and piled on the two-foot space left between each two trenches, the whole making a series of ridges as illustrated in Fig. 1.
Fig. 1. Trench System at Indore
As soon as the trenches are made in November, they are dug to a further depth of six inches and compost is thoroughly mixed with the soil of the floor of the trenches, which are then watered, cultivated when dry enough, and allowed to remain till planting time in February. In this way the soil in which the cuttings are to be planted is given time to prepare the food materials needed when growth begins. After planting and watering, the surface soil is lightly cultivated to prevent drying out. Afterwards four or five waterings are given, each followed by surface cultivation, which carry on the crop during the hot season till the break of the rains in June, when no further irrigation is needed.
When the young canes are about two feet high and are tillering vigorously, the trenches are gradually filled in, beginning about the middle of May and completing the operation by the middle of June, when the earthing up of the canes commences. This operation is completed about the middle of July (Fig. 2).
Fig. 2. Earthing up Sugar-cane at Shahjahanpur, 10th July 1919
One of the consequences of filling in the trenches and of earthing up canes grown in fertile soil is the copious development of fungi, which are plainly visible as threads of white mycelium all through the soil of the ridges and particularly round the active roots. I saw these for the first time at the Manjri sugar-cane farm near Poona about 1920 and the same thing was frequently observed at Shahjahanpur. No one suspected then that this fungous development could be explained by the fact that the sugar-cane is a mycorrhiza former and that we were observing the first stage of an important symbiosis between the fungi living on the humus in the soil and the sap of the sugar-cane. The provision of all the factors needed for this association -- humus, soil aeration, moisture, and a constant supply of fresh, active roots from the lower nodes of the canes as the earthing-up process proceeds -- explains why such good results have always followed the Shahjahanpur method of growing the cane and why the crops are so healthy. When grown on the flat under monsoon conditions, want of soil aeration and want of a constant supply of fresh roots would always be limiting factors in the full establishment of the mycorrhizal association
As at Shahjahanpur, the operation of earthing up the canes served four purposes: (1) the succession of new roots arising from the lower nodes, thoroughly combed the highly aerated and fertile soil of the ridges; (2) the conditions suitable for the constant development of the mycorrhizal association were provided; (3) the standing power of the canes during the rains was vastly improved, and (4) the excessive development of colloids in the surface soil was prevented. When this earthing up is omitted, a heavy crop of cane is liable to be levelled by the monsoon gales; crops which fall down during the rains do not ripen properly, do not give either the maximum yield of sugar or the much-prized, light-coloured product.
The operation of earthing up left deep drains between the rows of cane. It was essential, as at Shahjahanpur, to arrange that these drains were suitably connected with the ditches which carried off the surplus monsoon rainfall, so that no waterlogging of the area under cane occurred.
At Indore the Shahjahanpur results were repeated. The intensive cultivation of a suitable variety (POJ 213 and Coimbatore 213), proper soil aeration, good surface drainage, and an adequate supply of organic matter produced very fine yields of cane, free from fungous and virus diseases and exceptionally good samples of crude sugar (gur). The yields were not quite up to the Shahjahanpur standard, because it takes some years to work up the black soils to the highest pitch of fertility on account of the physical character of these heavy soils, but I am convinced that this was only a matter of perseverance. Unfortunately the time of retirement came before I could achieve the full results, but the remarkable yields obtained in the first three years left no doubt in my mind of the final result. There is no question but that the way to grow cane is the Shahjahanpur method, which should be adopted all over the world, particularly for raising the plant material.
No fungous or virus diseases were observed at Indore. The growth of cane and the ripening process were almost ideal. But not quite. It was noticed that the length of the nodes formed under irrigation during the hot season was rather short. Some factor seemed to be retarding growth during this period. At the time I put this down to the fact that the land under cane had only just been brought under irrigation and that insufficient time had been allowed to get these fields into that high state of fertility so essential when ordinary, rain-fed, black soils are converted into well-irrigated land. As a rule this takes five years in Central India. This retardation in growth during the hot season was accompanied by a very mild attack of the moth borer (Diatrea saccharalis), which lays its eggs in clusters on the under-side of the leaves and is followed by the destruction of the young shoots invaded by the caterpillars. Only a few shoots were destroyed; nothing was done to check the moth. As soon, however, as the rains broke, this pest disappeared of its own accord and no further damage occurred. Obviously some factor was operating during the hot season which altered the sap and lowered the resistance of the cane. I suspected at the time that the soil was not sufficiently fertile and did not contain sufficient humus for supplying the young growing cane with all the water it needed, and that this very minor trouble would disappear when the irrigated area was got into really good fettle. This is obviously a matter calling for detailed investigation.
At Indore the only manure used in raising the cane crop was compost. At Shahjahanpur the canes were grown on green-manure supplemented by a light dressing of cattle manure applied to the land before the green crop was sown. The only examples of organic manuring in commercial cane growing I have been able to discover are in Mauritius, where livestock are kept solely for their manure, which is used to break down cane trash into a rough form of compost. Thus at the Benares estate the residues of 140 cattle are converted into 1,500 tons of compost at a total cost of 6s. 6d. a ton. At Mon Trésor estate 5,000 tons of compost were made at a similar cost from the residues of 300 cattle and 500 sheep and goats. Further details of this organic manuring in Mauritius are to be found in a paper by G. C. Dymond reprinted in the News-Letter on Compost, No. 7, October 1943, p. 44.
In recent years another type of sugar-cane disease -- virus -- has assumed considerable importance. If virus is nothing more than a condition caused by imperfectly synthesized protein, aggravated by the use of artificials like sulphate of ammonia in place of humus, it would follow that a drastic alteration in manuring might remove the virus condition and restore health. In Natal this has been accomplished. Mr. G. C. Dymond found that when Uba canes, attacked by streak disease (a virus trouble), were manured with compost and the process was repeated for a year or two, the crop threw off the disease and grew normally. The restoration of health was accompanied by the establishment of the mycorrhizal association, which was absent in the cases of streak disease examined.
Dymond's discovery that freshly prepared compost not only restores virus- infected canes to health, but also re-establishes the mycorrhizal association, is of great importance in the future studies of cane diseases. The first step in such inquiries should be to examine the mycorrhizal status of the affected plants and then to restore it by growing cuttings of the diseased plants in heavily composted soil. In all probability the disease will disappear. Steps should then be taken to apply this knowledge on a field scale and then to see whether such crops can be infected by disease. If, as is most probable, no infection takes place, then the cause of the trouble -- bad farming -- has been established, as well as the remedy -- freshly prepared humus.
The next step will be to see how many of the fungous, insect, and virus diseases of the cane survive the Shahjahanpur methods of cane growing. This at least is certain -- the number will be few, perhaps none. In this way sugar-cane pests can be used as agricultural censors; their prevention will tune up practice; mycologists and entomologists will then become active and useful agents in development.
Intimately bound up with the prevention of cane diseases is the maintenance of the variety. As has already been pointed out (Chapter 1), the kinds of cane grown in the East have lasted for many centuries; on the modern sugar plantations a constant stream of new kinds has to be created. The prevention of this deterioration would seem to be bound up with the prevention of disease -- the maintenance without any sign of progressive deterioration in the synthesis of protein. This is accomplished in the indigenous sugar industry of India by the use of cattle manure and the restriction of the cuttings used in planting to the joint immediately below the cane tops. These are buried at harvest time and carefully kept till the new field is planted. Commercial sugar estates might copy this well-tried practice and so save the time and money expended in testing a constant stream of new canes.
In the course of my travels I have seen something of coffee cultivation -- in the West Indies, in various parts of India, and in the coffee-growing areas of Africa. I also visited in 1908 and again in 1938 the eroded areas in the centre of Ceylon which were devoted to coffee till the well-known rust fungus -- Hemileia vastatrix -- destroyed the plantations wholesale and caused them to be planted in tea. In all this two things impressed me very much: (1) the marked response of the coffee bush to forest soils rich in humus, and (2) the poor growth seen on areas suffering from erosion. On reconsidering in 1938 the original accounts of the great fungous epidemic in Ceylon some sixty years before, it appeared to me that the loss of the fertile top soil by erosion and the inadequate provision of fresh supplies of humus were ample reasons why this coffee disease had put an end to the industry. This surmise was strengthened by the establishment of the fact that coffee is a mycorrhiza former. This point is referred to in the following extract from my report dated 18th April 1938 on a visit to the tea estates in India and Ceylon:
"In view of the results obtained on the coffee estates in Kenya and Tanganyika with compost, it was expected that mycorrhiza would be found in this crop. Unfortunately my tour did not include any coffee estates where the Indore Process had been adopted. Three samples of surface roots, however, were collected.
"The first was taken from stray coffee plants growing on the roadside on unmanured land under grass at Dholai (Cachar, Assam). As was expected, Dr. Rayner found no trace of mycorrhiza in these root samples.
"Two more promising samples were collected at Talliar (High Range, Travancore), one from a nursery, the other from established coffee. In both cases the soil contained forest humus and in both Dr. Rayner found endotrophic fungous infection of the same type as that described in tea, but confined to the older roots and sporadic in distribution.
"The evidence, although incomplete and fragmentary, nevertheless points to mycorrhiza being as important a factor in coffee cultivation as it is proving in tea."
These observations were confirmed and amplified by the examination of material sent from Costa Rica by Señor Don Mariano Montealegre. There is no doubt that coffee, like tea and cacao, is a mycorrhiza former.
The fact that coffee is a mycorrhiza former is of considerable significance in the future cultivation of this crop. The humus in the soil and the sap of the plant are in intimate contact by means of this natural mechanism. Obviously, therefore, if coffee of the highest quality is to be produced and if the plants are to withstand disease, the first condition of success in coffee cultivation is the provision of properly made humus.
This naturally involves some form of mixed farming so that an ample supply of urine and dung is available on the spot. Pigs, buffaloes, and cattle will probably be the best agents for this purpose. The day, therefore, may not be far distant when the coffee estates will be partly devoted to livestock, which will automatically cancel out the present expenditure on artificial manures and insecticides, and do much to raise the yield per acre and also improve the quality -- a matter of supreme importance in this crop.
One illuminating consequence of the devastating epidemic of coffee leaf disease in Ceylon impressed me during my tours in the island in 1908 and thirty years later in 1938. The many planters I met not only had not forgotten this visitation, but were still labouring under the thraldom of fear of the parasite. When I suggested that fungous and insect diseases are the direct consequence of mistakes in crop production and should, therefore, be regarded as friendly professors of agriculture provided by Nature free of charge for our instruction, I found myself up against a solid armour-plate of fear. Disease, like erosion, were things which had to be studied by specialists and then tackled by direct action.
Under these unpromising conditions I did not pursue the subject and go on to suggest that Hemileia vastatrix would prove most useful in another way. This disease of the coffee plant might well be used not only to teach us how to grow coffee properly, but also in reference to another crop -- the tea plant. A few coffee plants, established here and there among the tea, would tell us whether the soils of Ceylon had been sufficiently restored to fertility by the anti-erosion methods undertaken, by the planting of adequate shade, and above all by the practice of systematically converting all vegetable and animal residues into humus. They could do this without any soil analyses or other laboratory tests by simply withstanding the onset of the leaf disease or by succumbing to it; where the disease appeared, we should know that the soil still lacked fertility; when it was absent, we should be able to be satisfied with the measures taken.
Such a device would be very simple. It would be efficient because it would be using Nature's own agencies in testing conditions. Why should we not make use of so excellent and so inexpensive a method? The Ceylon tea planter should look on coffee and the diseases it carries as one of his best, his most willing, and his most reliable assistants.
Although a number of insect and fungous diseases have been reported on the tea plant, nevertheless the total damage done by these pests is not excessive Nothing like the coffee leaf disease of Ceylon, which in a few years destroyed the plantations wholesale, has been reported in the case of tea. Indeed in Ceylon, as has already been stated, tea replaced coffee on the partially eroded soils, a fact which suggests that the tea bush is exceptionally hardy and robust. This view is confirmed by the behaviour of this species under cultivation. The plants are constantly plucked and so deprived of those portions of their foliage richest in food materials; every few years the bushes are heavily pruned, after which they have to re- create themselves; in China a tea plantation lasts a century or more. Only a very vigorous bush could endure such treatment for so long.
It would follow from all these considerations that the struggle between the host and the parasite might easily result in the victory of the former, if the tea plant were given a little assistance. It might then be easy to reduce the damage done by pests to something quite insignificant.
Can the tea plant itself throw any light on this question of natural resistance to disease? Has the tea bush anything to say about the assistance it needs to vanquish the various insect and fungous pests always ready to attack it? If so, its representations must be carefully studied and if possible implemented. The plant or the animal will answer most queries about its needs if the questions are properly posed. The wise farmer, planter, or gardener always deals with such responses with sympathy and respect.
The tea plant has very recently delivered a most emphatic message on the cause of disease and its prevention which is certain to interest many readers in no way connected with the tea industry. The story I have to tell began in 1933 when I interested myself in the career of Dr. C. R. Harler (who had just been retrenched when the Tocklai Research Station, maintained by the Indian Tea Association, was reorganized in that year). I consoled him for his temporary loss of employment by assuring him: (1) that retrenchment, as in his case, often falls on the best men; (2) that he could do much more for the tea industry as an independent worker with adequate scope than as a member of the obsolete organization he had just left; and (3) that a promising line of future work lay in the systematic conversion into humus of the waste products of the tea estates. He agreed. Then Providence intervened on his behalf, on behalf of the tea plant and of the tea industry. Dr. Harler was offered and accepted (August 1933) the post of Scientific Officer to the Kanan Devan Hills Produce Company in the High Range, Travancore, the property of Messrs. James Finlay & Co. Ltd., who direct the largest group of tea gardens in the world. On taking up his duties at Nullatanni near Munnar, Dr. Harler proceeded to apply the Indore Process on an estate scale. No difficulties were met with in working the method; ample supplies of vegetable wastes and cattle manure were available; the local labour took to the work and soon the General Manager of the Company, as well as the Estate Managers, became enthusiastic. It was now possible to pose the following question to the tea plant: What do you need to throw off disease and to do your best as regards the yield and quality of tea?
The second half of this question was soon answered on the Kanan Devan tea gardens, the first half had to wait till some years later. The pioneering work at Nullatanni, which was completed towards the end of 1934, was followed by the adoption of the Indore Process on the rest of the gardens -- some forty in number. Each garden made from its available vegetable and animal wastes all the manure the tea needed; no artificials were necessary; yield and quality notably improved. But the tea plant in these gardens could say nothing about its requirements to ward off disease for the simple reason that with one small exception -- the minor root trouble referred to below -- there was practically no disease to resist in these well managed properties. All that properly made compost could do was to increase the yield and improve the quality of the tea above the high standard already reached.
When the news of Dr. Harler's successful estate-scale trial at Nullatanni reached me in September 1934, it occurred to me that it might be worth while bringing the possibilities of the Indore Process to the notice of the rest of the tea industry, which is arranged in large groups controlled by a small London directorate principally recruited from the industry itself. As I had no contacts with these bodies it was necessary to make one -- preferably with some pioneer likely to be interested. I soon found the man -- Mr. James Insch, one of the then Managing Directors of Messrs. Walter Duncan & Company. A small-scale trial of the Indore Process was completed on fifty-three estates of this group in Sylhet, Cachar, the Assam Valley, the Dooars, Terai, and the Darjeeling District. By the beginning of 1935 some 2,000 tons of compost in all were made and distributed. Five years later the quantity on the Duncan group had passed the 150,000 tons a year mark. But again the tea plant on these widely distributed properties did not answer the question: What do you need to throw off disease? The reason for this was that, as on the High Range of Travancore, the amount of disease on these estates was insufficient for such a question to be posed and answered. On these properties all the Indore Process could do was. to raise the yield and improve the quality still further.
The results already referred to and the publicity they received came to the notice of many other groups of tea estates in India, Ceylon, and Africa The methods of composting which had proved so successful on the Finlay and Duncan estates were tried at many new centres. It was in the course of these widely dispersed trials that the tea plant informed us what it needed to keep insect and fungous pests in check and why it wanted this assistance.
In a few cases during this third series of trials both insect and fungous diseases did occur to an extent which reduced somewhat the yield of tea. There was just sufficient disease here and there for the query under discussion to be put to the tea plant. The question on these particular gardens was not posed deliberately, but quite by accident. While this series of trials was in progress, example after example came to my notice in which such small applications of compost as five tons to the acre were at once followed by a marked improvement in growth, in general vigour, and in resistance to disease. Although very gratifying in one sense, these results were distinctly disconcerting. If humus acts only indirectly by increasing the fertility of the soil, time will be needed for the various biological, physical, and chemical changes to take place. If the plant responds at once, as was obviously the case, some other factor besides a general improvement in soil fertility must be at work. What could this factor be? It was clearly some agency which enabled humus to effect directly and very quickly the nutrition of the plant.
In a circular letter issued on 7th October 1937 to correspondents in the tea industry I suggested that the most obvious explanation of any sudden improvement in tea observed after one moderate application of compost could only be due to the effect of humus in stimulating the mycorrhizal relationship, which I afterwards discovered had been observed in Java in the roots of this crop. It seemed to me that this association must be present and that it would enable the fungous factor in the partnership to transfer the digestion products of protein into the sap and then into the green leaf. The virtues of humus could thus be moved from soil to plant in a very short space of time. This would enable the plant not only to resist disease, but would also explain the marked improvement in the yield and quality of tea which resulted from dressings of compost. I saw all this in imagination, as it were, on 7th October 1937 as a likely hypothesis to explain the facts. What set these ideas in train was a perusal of Dr. M. C. Rayner's work on conifers at Wareham in Dorsetshire, where small additions of properly made compost had led to spectacular results most easily explained by the establishment of the mycorrhizal association. (An account of this Wareham work has since been published in 1944 in book form under the title -- Problems in Tree Nutrition -- by Messrs. Faber and Faber, London.)
At this juncture a group of tea companies which had adopted the Indore Process asked me to visit their estates in India and Ceylon. In the course of this tour, which lasted from November 1937 to February 1938, I examined the root system of a number of tea plants which had been manured with properly made compost, and found everywhere the same thing -- numerous tufts of healthy-looking roots associated with rapidly developing foliage and twigs much above the average. Both below and above ground humus was clearly leading to a marked condition of wellbeing. When the characteristic tufts of young surface roots were examined microscopically, the cortical cells were seen to be literally overrun with mycelium to a much greater extent than is the rule in a really serious infection by a parasitic fungus. Clearly the mycorrhizal relationship was very much involved: my hypothesis was abundantly confirmed: the tea plant had a message to deliver on the disease question. My hasty and imperfect observations made in the field and in the course of a very strenuous tour -- during which many estates were visited in detail and many lectures were delivered to groups of planters -- were confirmed and extended by Dr. M. C. Rayner and Dr. Ida Levisohn who examined a large number of my root samples, including a few in which artificials only were used or where the soils were completely exhausted and the garden had become derelict with perhaps only half the full complement of tea plants. In these latter cases the characteristic tufts of normal roots were not observed; development and growth were both defective; the mycorrhizal association was either absent or poorly developed. Where artificials were used on worn-out tea, infection by brownish hyphae of a Rhizactonia-like fungus (often associated with mild parasitism) was noticed. But whenever the roots of tea manured with properly made compost were critically examined, the whole of the cortical tissues of the young roots always showed abundant endotrophic mycorrhizal invasion, the mainly intra-cellular mycelium apparently belonging to one fungus. This fungus was always confined to the young roots and no invasion of old roots was observed. In the invaded cells the mycelium exhibits a regular cycle of changes from invasion to the clumping of the hyphae around the cell nuclei, digestion and disintegration of their granular contents, and the final disappearance of the products from the cells. In this way the digestion products of the proteins of the fungus pass into the cell sap and then into the green leaves.
Humus in the soil, therefore, affects the tea plant direct by means of a middleman -- the mycorrhizal association. Nature has provided an interesting piece of living machinery for joining up a fertile tea soil with the plant. Obviously we must see that this machinery is provided with the fuel it needs -- continuous dressings of properly made compost. I saw on several occasions the response of the tea plant, which had been attacked by disease, to small dressings of compost. I was amazed by the way even a single application had reduced the amount of infection and started the tea bushes well on the way to complete recovery.
The tea plant had now answered the question: What must be done to me to be saved? It is nothing less than the restitution of the manurial rights this plant enjoyed in its forest home -- regular supplies of freshly prepared compost.
One difficulty was encountered and partly overcome in this restitution of manurial rights. In some of the tea areas the gardens were so closely jammed together that it was not possible to maintain the head of cattle needed to provide the animal manure for making first-class compost. I suggested that in such cases pigs would be the easiest livestock to keep and that the cost of the pig food brought on to the gardens could be found by reducing the amount of artificial manure that would be needed. But where land was available, steps were taken to increase the head of other livestock to make the necessary animal manure.
One interesting case of introducing cattle into the tea gardens solely for their manure came to my notice from Africa. When Viscount Bledisloe returned to England from his African mission, where he had been Chairman of a Royal Commission connected with the affairs of the Rhodesias and Nyasaland, he presented me with an enlarged set of the photographs he had taken on compost making, the virtues of which he constantly brought to the notice of the various local governments with whom he came in contact. In this way he did much of the spade work which was necessary to make South Africa compost-minded. One of these photographs, taken at Messrs. J. J. Lyons & Company's estate at Mlange, showed the cattle which the tea gardens of Nyasaland were beginning to keep solely for compost making (Plate III). This, indeed, was proof positive of progress and of enterprise. If the tea gardens of Africa can go to the trouble of maintaining cattle for the sake of the urine and dung they produce, what is to prevent other plantation industries all over the world doing the same? It is impossible to farm for long without livestock. It is equally impossible to maintain the overseas plantations in an efficient condition without these living manure factories for producing two of the essentials for making humus. Like tea, all these plantation crops -- coffee, cacao, sugar-cane, cotton, sisal, maize, coconuts, bananas, citrus fruit, grapes, apples, pears, peaches, and so forth -- are mycorrhiza formers. All need the digestion products of fungous protein to maintain the power to reproduce themselves, to provide high-quality crops, and to resist the onslaught of insects and fungi.
Plate III. Livestock for making compost on a tea estate in Africa.
But cases of disease occur in tea which cannot be remedied by getting the surface soil into good fettle. The tea is a deep-rooting plant and makes great use of the lower roots to keep up the water supply during dry weather. These deep roots must, therefore, function properly. There must be no waterlogging due to stagnant water held up by impermeable layers in the subsoil. This condition invariably results in root disease duly followed by the death of the plant. The only example of such disease of any consequence I met with during my second tour in India and Ceylon was a root fungus which appeared here and there and destroyed the bushes over small areas particularly on the laterite soils of South India. The real cause of the trouble appeared to be some interference with drainage in the lower layers of the soil, which reduced the vitality of the tea and prepared the way for the parasite. Such diseases might be dealt with most easily by Swedish pillar-drains -- vertical pits, dug well below the layer under the laterite holding up the stagnant water, and afterwards filled with large stones.
At the Gandrapara estate on the flat stretches of the alluvium of the Bengal Dooars I saw one of the best examples in my experience of successful surface drainage under a high monsoon rainfall, which I was told had proved very useful in the prevention of root disease. On this fine property, very deep and narrow minor earth drains had been constructed among the tea and connected up with wider major ditches which carried off the surplus water to the natural drainage lines. The system was based on a contour survey and had been carried out by a competent engineer. The minor drains could not easily be detected, as the tea bushes on either side met above the drains, forming everywhere a continuous green table. With the combined help of the excellent top shade and this green table the heavy monsoon downfalls were converted into fine spray, which was readily absorbed by the heavily composted surface soil without any great silting up of these minor drains. I had studied surface drainage in many parts of the world, including some of the best examples Italy has to provide, and had carried out drainage schemes on the land in my own charge, but none of these came up to the Gandrapara standard. I mentioned this fact at a lecture to a group of local tea planters at Gandrapara. By chance the engineer who had designed the local scheme was present. His grateful reaction to my chance remarks will remain as one of my pleasantest recollections.
The superficial character of the conventional investigations on the diseases of tea will be clear from what has been set out above. Nothing is to be gained by starting research on any future tea disease at the wrong end. Investigation must always begin with the soil. If the mycorrhizal association is not working properly, this must be put right in the first place. The drainage of the soil round the deep roots must also be effective. In all probability the result will be the rapid disappearance of pests. Proceeding in this way, diseases can be made very useful for keeping a tea garden up to the mark as regards manuring and soil management.
Cacao (Theobroma cacao)
A good deal of time was spent by me in Grenada about 1901 on the study of the fungous diseases of cacao. Visits were also paid to a number of cacao estates in Trinidad and Dominica. The main troubles were three: die-back of the leaders on low-lying areas (caused by poor drainage), pod, and bark diseases. A new fungous pest -- the witch broom disease -- had just made its appearance in Surinam, but had not then spread to Trinidad and the other islands. It has since become a serious trouble in the West Indies.
Among the many estates visited was a small plantation in Grenada owned by the late Rev. G. W. Branch, which stood out from the rest of the island by virtue of the heavy yields of high-quality beans; the fact was ascertained that these cacao trees were always manured with farmyard manure. Although a paper was read by the owner at one of the West Indian Conferences in the early years of this century and full details of the method of manuring were given, it never struck anyone that here in a nutshell was the solution of the main problem of cacao, namely, mixed farming and the preparation of plenty of freshly prepared compost for the cacao trees. Everybody without exception who attended this meeting was labouring under the thraldom of the NPK mentality and was only able to think in terms of so many pounds to the acre of this or that artificial manure. Though many were impressed by these Grenada results, they seemed incapable of facing up to their very obvious implications. All this happened about 1901.
In 1908 in the course of a visit to Ceylon I saw these Grenada results repeated, but on a much larger scale, at the Kondesalle cacao estate near Kandy. Thirty years later -- in 1938 -- when on my tour of the tea estates of India and Ceylon I resumed my interest in cacao and re-visited Kondesalle, at which the finest cacao beans I have ever seen are being produced. I again observed no cacao diseases on this property and was not told of any by the manager or by his assistants. The trees appeared exceedingly healthy and here again, as on the small Grenada plantation, livestock -- in this case, pigs and Hissar cattle -- were kept for producing the farmyard manure applied to the cacao trees.
During this tour samples of the surface roots of cacao at Kondesalle were fixed and sent to London for examination by Dr. Rayner. The results are referred to in my report on this tour in the following words:
"Cacao. Dr. Rayner examined the surface roots of cacao from Kondesalle (Ceylon) taken from a field which had been manured with farmyard manure. Sporadic mycorrhizal infection of endotrophic (i.e. intracellular) type was present. Compost is not yet being made on this estate. It will be interesting to see whether still better results than those now yielded by farmyard manure on this fine property could not be obtained if the cattle and pig manure were first composted with the estate wastes and used in the form of humus."
It will be obvious that in both Grenada and Ceylon examples of how to grow heavy crops of high quality cacao, free from disease, have long been provided by accident, as it were. Meanwhile both these regions have been furnished with modern agricultural departments. The astounding fact is that no one in these organizations or in the planting community has understood the value or the significance of the lessons these two estates have to teach. Nevertheless, both indicate quite clearly how cacao will have to be produced in the future if the growing menace of disease is to be averted. As is well known, much of the cacao of commerce now comes from West Africa, where it is produced largely at the expense of the original stores of humus left by the forest. As in Grenada and Trinidad, these stores will not last for ever. After a time they will be used up and the day of reckoning will arrive. Indeed, this has already come.
In the West India Committee Circular of September 1944 an article appeared on the future welfare of this crop in the Gold Coast -- the world's largest exporter of cacao. It appears that the industry is face to face with a crisis "perhaps without equal in the history of any major tropical crop in the British Empire".
Two factors are responsible for this state of affairs: (1) the swollen-shoot virus disease, first reported in 1936, and (2) capsid bugs. These two pests are being investigated at the Tafo Cacao Research Station established by the local Agricultural Department in 1938. The spread of these two diseases has been so rapid as to constitute a direct menace to the whole future of the industry. In 1943 a conference of research workers was held at Tafo, presided over by the Agricultural Adviser to the Secretary of State. A programme of future research in cacao was formulated. Plans were also made for the reorganization of the Tafo Station as the West African Cacao Research Institute, for which a director has been appointed.
There seems no doubt that what is needed to place the cacao industry of the Gold Coast on a sound foundation is not more research into cacao diseases, but the introduction of livestock into the areas growing cacao and the conversion of the wastes of the animal and the plant into humus, as Messrs. J. J. Lyons & Company have done on their tea estates in Nyasaland (see above). The Gold Coast cacao industry, which began to export produce at the beginning of the century, has obviously been living for the last forty years or so on capital -- on the humus left by the original forest. This has now been used up and Nature has registered her usual protest in the form of disease. The West African cacao trees have been deprived of their manurial rights. The Kondesalle cacao estate in Ceylon indicates what should be done to put matters right. No committees, however well selected, and no amount of research, however devoted, will alter this obvious conclusion. The time has indeed come for the prodigal to return, to confess, and to start proper farming.
There is no doubt that the cacao industry all over the Empire could at once be restored by mixed farming and the systematic conversion into compost of all the vegetable and animal wastes available. The manufacturing interests in Great Britain which need a regular and reliable supply of cacao beans should at once use their influence and insist that this obvious reform be taken in hand forthwith.
One objection to this suggestion must be answered in advance. If a portion of the existing areas under cacao is devoted to mixed farming, how is the output to be maintained? The answer is: By virtue of the vastly increased yield and better quality of the beans, as well as the longer life of the trees. There is ample land in all the cacao-growing areas of the Empire for this crop and also for livestock: there is no reason why this reform should not be set in motion forthwith. Must we always wait for catastrophe before the simplest step forward can be taken? What has the agricultural research organization of the Colonies been doing to allow such a state of affairs as this Gold Coast cacao scandal to develop?
The cotton crop suffers from many insect and a few fungous diseases. It has already been mentioned that one-quarter of the space of the last pre- war issues of the Empire Growing Cotton Review was devoted to disease. The alarming significance of the figures given can only be realized when it is remembered that cotton is a distinctly robust crop that does not need very intensive methods of farming to produce fair yields of fibre. Moreover, cotton should not exhaust the land very much, as the fibre of commerce contains little more than the cellulose manufactured from the gases of the atmosphere and the water in the soil; the flowers fall after the bolls set; the leaves of the crop mostly drop before the stalks are removed; the roots remain in the ground: the seed is very useful for feeding the work cattle. Provided, therefore, a fair proportion of the cotton seed is passed through the stomachs of oxen and other animals and the old stalks find their way back to the soil in the form of humus, this crop cannot possibly wear out the land to any appreciable extent. Further, as inter-cultivation between the rows has to stop when the flowers appear, a cotton crop always enables weeds to cover the surface which, when ploughed under, help to maintain the humus content of the soil. If the incidence of disease depends on the poverty of the soil, it would seem that there must be something very wrong somewhere in the current methods of cotton growing; otherwise these diseases ought not to occur. A cotton crop, if properly looked after, ought to be very free from pests.
During the years 1924-31 I had unique opportunities for the study of this crop, because during this period I held the post of Director of the Institute of Plant Industry at Indore in Central India, at which cotton was the principal crop. Indeed, the new institute could not have been founded or maintained without the help of large grants from the Indian Central Cotton Committee, which in turn was financed by a small annual cess on each bale of raw cotton exported from India or used in the local mills. This cess was naturally passed on to the multitude of smallholders who raised the crop. If, therefore, the Indian Central Committee could do something to help these men in return for their money, this new body and its various research workers would have justified their existence.
Before taking up an investigation of the cotton crop at Indore in 1924, a survey of cotton growing in the various parts of India was undertaken. At the same time, the research work in progress on cotton in other parts of the world was critically examined.
As regards cotton growing in India, the two most important areas are: (1) the black cotton soils of the Peninsula, which are derived from the basalt; (2) the alluvium of north-west India, consisting of deposits left in a deep chasm by the rivers of the Indo-Gangetic plain. Besides these there are small areas of garden cultivation in southern India, where American types of cotton are grown intensively under irrigation and where heavy crops of good fibre are the rule.
On the black soils there are thousands of examples which indicate the direction research on this crop should take. All round the villages of the Peninsula, zones of very highly manured land, rich in organic matter, occur. These are kept in good fettle by the habits of the people: the night- soil is habitually added a little at a time to the surface of the fields. On such zones cotton does well no matter the season; the plants are well grown and remarkably free from pests; the yield of seed cotton is high. On the similar but unmanured land alongside the growth is comparatively poor; only in years of well-distributed rainfall is the yield satisfactory. But even under the most adverse conditions one is amazed to see how the cotton plant manages to survive and to produce some kind of crop. Only the very hardiest plant could produce seed under such unfavourable circumstances. The limiting factor in growth on these black soils is the development, soon after the rains set in, of a colloidal condition, which interferes with aeration and impedes percolation. This occurs on all black soils, but organic matter mitigates the condition. As these soils dry out at the end of the rains, extensive cracking occurs which aerates the soil but also damages the roots and rapidly desiccates the soil. The varieties of cotton, therefore, must possess the power of rapid ripening, otherwise the bolls could not open in time. The growth period of any successful cotton on the rain-fed, black soil areas must be short; the plant must literally burst into cotton at picking time and show no tendency to linger in yielding up its crop. Two pickings at the most are all that is possible.
On the alluvium of north-west India a somewhat similar limiting factor occurs. Here cotton is grown on irrigation, which first causes the soil particles to pack and later on to form colloids. In due course the American varieties, whose root systems, compared with those of the indigenous cottons, are superficial, show by their growth that they are not quite at home. The anthers, the most sensitive portion of the flower, sometimes fail to open and to release their pollen: the crop is unable to set a full crop of seed. But this is not all. The ripening period, particularly in the Punjab, is unduly prolonged; as many as four pickings are necessary. Moreover, the fibre often lacks strength, quality, and life. The cause of these troubles is poor soil aeration, which in these soils leads to a very mild alkali condition. This, in turn, prevents the cotton crop from absorbing sufficient water from the soil. One of the easiest methods of preventing this packing and alkali formation is to increase the bacterial population by means of dressings of humus. In this way the soil is able to re-create a sufficient supply of compound particles to restore the aeration and improve the water supply needed by the cotton.
As regards disease, insects cause more damage to the crop than do fungi: there is more insect disease on the alluvium than on the black soils. The insect diseases on the alluvium mostly affect the bolls which, as we have seen, develop but slowly. If the cotton could be made to ripen more quickly, these boll diseases might be very considerably reduced.
The direction of research work on cotton was, therefore, disclosed by a study in the field of the crop itself. The problem was how best to maintain soil aeration and percolation. This could be solved if more humus could be obtained. At the same time, there appeared to be every chance that more humus would materially reduce, by speeding up maturation, the damage done to the ripening bolls by the various boll worms. Good farming methods, therefore, including a proper balance between livestock and cotton, seemed to provide the key to the cotton problems of India. Once the soils were got into good fettle and maintained in this condition, the question of improved varieties could then be taken up with every chance of success. To hope to overcome bad farming by improving the variety in the first place was an obvious impossibility, such a research policy amounting to a contradiction in terms.
A study of the research work on cotton which had been done all over the world did nothing to modify this opinion. Cotton investigation everywhere appeared to suffer from the fragmentation of the factors, from a consequent loss of direction, from failure to define the problems to be investigated, and from a scientific approach on far too narrow a front without that balance and stability provided by adequate, first-hand farming experience. The research workers seemed to be far too busy on the periphery of the subject and to be spending their time on unimportant details. This has naturally resulted in a spate of minor papers which lead nowhere except to the cemetery so providentially furnished by the Empire Cotton Growing Review. In Africa, particularly, much time and money have been wasted in trying to overcome, by plant-breeding methods, diseases which obviously owe their origin to a combination of worn-out soil and bad farming.
Steps were therefore taken at Indore to accelerate the work on the manufacture of humus which had been begun at the Pusa Research Institute. The Indore Process was the result. It was first necessary to try it out on the cotton crop. The results are summed up in the following table.
The Increase in General Fertility at Indore
Area in acres of improved land under cotton
Average yield in lb. per acre
Yield of the best plot of the year in lb. per acre
Rainfall in inches
27.79 (distribution good)
40.98 (a year of excessive rainfall)
23.11 (distribution poor)
The figures show that, no matter what the amount and distribution of rainfall were, the application of humus soon trebled the average yield of seed cotton -- 200 lb. per acre -- obtained by the cultivators on similar land in the neighbourhood.
In preparing humus at Indore one of the chief wastes was the old stalks of cotton. Before these could be composted they had to be broken up. This was accomplished by laying them on the estate roads, where they were soon reduced by the traffic to a suitable condition for use as bedding for the work cattle prior to fermentation in the compost pits. I owe this suggestion to Sir Edward Hearle Cole, who hit upon this simple device on his Punjab estate.
The first cotton grower to apply the Indore Process was Colonel (now Sir Edward) Hearle Cole at the Coleyana Estate in the Montgomery District of the Punjab, where a compost factory on the lines of the one at the Institute of Plant Industry at Indore was established in June 1932. At this centre all available wastes have been regularly composted since the beginning; the output is now about 8,000 tons of finished humus a year. Compost has increased the yield of cotton, improved the fibre, lessened disease, and reduced the amount of irrigation water by a third. The neighbouring estates have all adopted composting; many interested visitors have seen the work in progress. One advantage to the Punjab of this work has, however, escaped attention, namely the importance of the large quantities of well grown seed, raised on fertile soil, contributed by these estates to the seed distribution schemes of the Provincial Agricultural Department. Plant breeding, to be successful, involves two things -- an improved variety plus seed for distribution grown on soil rich in humus.
The first member of an agricultural department to adopt the Indore method of composting for cotton was Mr. W. J. Jenkins, C.I.E., when Chief Agricultural Officer in Sind, who proved that humus is of the greatest value in keeping the alkali condition in check, in maintaining the health of the cotton plant, and in increasing the yield of fibre. At Sakrand, for example, no less than 1,250 cart-loads of finished humus were prepared in 1934-5 from waste materials such as cotton stalks and crop residues.
During recent years the Indore Process has been tried out on some of the cotton farms in Africa belonging to the Empire Cotton Growing Corporation. In Rhodesia, for example, interesting results have been obtained by Mr. J. E. Peat at Gatooma. These were published in the Rhodesia Herald of 17th August 1939. Compost markedly improved the fibre and increased the yield not only of cotton, but also of the rotational crop of maize. The results obtained by the pioneers in India, therefore, apply to Africa.
Why cotton reacts so markedly to humus has only recently been discovered. The story is an interesting one, which must be placed on record. In July 1938 I published a paper in the Empire Cotton Growing Review (Vol. XV, No. 3, 1938, p. 186), in which the role of the mycorrhizal relationship in the transmission of disease resistance from a fertile soil to the plant was discussed. In the last paragraph of this paper the suggestion was made that mycorrhiza "is almost certain to prove of importance to cotton and the great differences observed in Cambodia cotton in India in yield as well as in the length of the fibre, when grown on (1) garden land (rich in humus) and (2) ordinary unmanured land, might well be explained by this factor". In the following number of this Journal (Vol. XV, No. 4, 1938, p. 310) I put forward evidence which proved that cotton is a mycorrhiza former. The significance of this factor to the cotton industry was emphasized in the following words:
"As regards cotton production, experience in other crops, whose roots show the mycorrhizal relationship, points very clearly to what will be necessary. More attention will have to be paid to the well tried methods of good farming and to the restoration of soil fertility by means of humus prepared from vegetable and animal wastes. An equilibrium between the soil, the plant, and the animal can then be established and maintained. On any particular area under cotton, a fairly definite ratio between the number of livestock and the acreage of cotton will be essential. Once this is secured there will be a marked improvement in the yield, in the quality of the fibre, and in the general health of the crop. All this is necessary, if the mycorrhizal relationship is to act and if Nature's channels of sustenance between the soil and the plant are to function. Any attempt to side-track this mechanism is certain to fail.
"The research work on cotton of to-morrow will have to start from a new base line -- soil fertility. In the transition between the research of to-day and that of the future, a number of problems now under investigation will either disappear altogether or take on an entirely new complexion. A fertile soil will enable the plant to carry out the synthesis of proteins in the green leaf to perfection. In consequence the toll now taken by fungous, insect, and other diseases will at first shrink in volume and then be reduced to its normal insignificance. We shall also hear less about soil erosion in places like Nyasaland, where cotton is grown, because a fertile soil will be able to drink in the rainfall and so prevent this trouble at the source."
Confirmation of these pioneering results soon followed. In the Transactions of the British Mycological Society (Vol. XXII, 1939, p. 274) Butler mentions the occurrence of mycorrhiza as luxuriantly developed in cotton from the Sudan and also in cotton from the black soils of Gujerat (India). In the issue of Nature of 1st July 1939 Younis Sabet recorded the mycorrhizal relationship in Egypt. In the Empire Cotton Growing Review of July 1939 Dr. Rayner confirmed the existence of mycorrhiza in samples of the roots of both Cambodia and Malvi cotton collected at my suggestion for her by Mr. Y. D. Wad at Indore, Central India, from both black cotton soil and from sandy soil from Rajputana.
The problem now to be solved in cotton production and in the control of disease is the discovery of the easiest way in which the present extensive methods of agriculture can be converted into more intensive methods. This involves a great increase in livestock in the existing cotton areas and the systematic conversion of the cotton stalks into humus. In this way the yield per acre can rapidly be increased and the fibre improved. The present supplies of cotton can, therefore, be produced from about two-thirds the area now under this crop. The land so released can be used for the production of food grains and fodder crops. A balanced agriculture is the key to the prevention of the diseases of cotton.
Every point here discussed was mentioned or suggested in the section on cotton in An Agricultural Testament published in 1940. It will be interesting to observe how long it will take such bodies as the Empire Cotton Growing Corporation and the Indian Central Cotton Committee to revise their research policies and to replace their laboratory workers by farmer-scientists.
The most important cereal in the world is rice. Moreover, it is a crop remarkably free from diseases of all kinds. Rice, therefore, should take high rank among Nature's professors of agriculture. A study of its cultivation might teach us much about the prevention of disease.
But the moment we embark on such a study we find no less than three of the principles underlying Western agricultural science flatly contradicted by this ancient cultivation.
In the first place, in many of the great rice areas of the world there is no such thing as a rotation of crops. Rice follows rice year after year and century after century without a break, without even a fallow year every now and then. Moreover, there is no falling off in yield and no sign of soil exhaustion. There is, therefore, no need of a continuous rice experiment of the Broadbalk pattern for the simple reason that such age-long experiments are to be seen everywhere. To begin a new one would be to carry coals to Newcastle.
In the second place, these continuous rice crops do not need those extraneous annual applications of nitrogenous manures which are considered to be essential for all cereals. The rice fields somehow manure themselves.
In the third place, the rice crop often covers vast areas of land in one unbroken sheet, thereby providing a paradise for insect and fungous diseases. But these do not occur: on the contrary, the rice crop is generally remarkably free from diseases of all kinds.
What is the secret underlying these unexpected and unconventional results? The beginning of the solution of the riddle will, I think, be found in the nurseries in which the young rice plants are raised before transplanting. These are always on well aerated and well manured land, the manure, as a rule, being well decayed cattle manure. The result is the rice seedlings become veritable arsenals of such things as nitrogen, phosphorus, and potash, all in organic combination. Moreover, the rice plant is a mycorrhiza former and so ample provision occurs even in the seedling stage for the circulation of protein between soil, sap, and green leaf. How important this building up of the rice seedling is will be clear, when it is realized that the transplanting process from well aerated soil to mud involves a completely fresh start in a new environment. This results in a delay of many days and, therefore, in the loss of a substantial proportion of the total growing period. Nevertheless, transplanting pays, because transplanted rice always gives a better yield than broadcast rice in which, of course, there is no delay in growth. Here we have a clear and definite lesson from the long experience of the Orient, namely, the vital importance of well-nourished seedlings. This applies in particular to crops like fruit, tea, coffee, cacao, tobacco, vegetables, and so forth. In all these well begun is half done.
But how does the rice manage to manure itself? The answer is provided by the nitrogen-fixing powers of the algal film found in rice fields. This algal film does three things: it aerates the water of the rice fields; it fixes a continuous supply of nitrogen from the atmosphere; it leaves behind a useful amount of easily decomposable organic matter. Nevertheless, more organic matter is needed in the rice fields beyond that supplied by the algal film and the roots of the old crop. How markedly rice benefits from compost has been proved at Dichpali in India. The results have already been set out in Chapter V of An Agricultural Testament, pp. 80-2.
The problem now is to find more compost for the rice crop. Nature has already provided ample vegetable waste in the shape of the water hyacinth, an aquatic weed to be found in most of the rice-growing areas of the world. This water weed should be regarded as a heaven-sent gift of Providence for the rice-growing areas, as it provides not only large supplies of readily fermentable vegetable matter, but sufficient moisture for the composting process as well. All that is needed besides is a supply of cow-dung and urine earth, both of which are available locally. In Bengal, for example, the annual yield of rice could be vastly increased if only a national campaign for the composting of the water hyacinth could be set in motion. That this weed makes excellent compost has already been fully demonstrated: first at Barrackpore, near Calcutta, by Mr. E. F. Watson, O.B.E., the Superintendent of the Governor's Estates, Bengal, and later on some of the tea estates in Assam. No future rice famines in Bengal need be feared once full use is made of the vast local supplies of water hyacinth.
What is the explanation of the comparative immunity of the rice crop from disease? I think the answer is provided by the fact that rice is a mycorrhiza former and that this mechanism works not only in the rice nurseries, but also in the paddy fields themselves: nothing has interfered with this process, as artificial manures are unknown and such bad practices as over-irrigation are, from the nature of the case, impossible. Indeed, the behaviour of this crop as regards parasites supplies strong confirmation of the view that what matters most in crop production is the effective circulation of protein between soil and sap, followed by the synthesis of still more protein of the right kind in the green leaf. High quality protein will, in ordinary circumstances, always protect the plant against its enemies.