were 'selected' only in the sense that they were picked out as resembling each other; they were not 'selected' in the sense in which the term is used by the breeder of pure-bred stock. It has been noted above that the ewes selected in the United Provinces failed; this failure was not, however, so marked with the Bikanir ewes, and the inference is that the Bikanir ewes were less mongrel in their constitution than those of the United Provinces. GOAT BREEDING. 159. Our conclusions in the preceding paragraph apply also to goats. There are obvious directions in which the goat might be improved both as a milking and a hairproducing animal. In many parts of this, as of other countries, the goat is the poor man's cow; and there can be little doubt that, like the cow, its milk yield could be raised with little difficulty by selection. Since many breeds, or types, of goat have been recognised in different parts of the country, there is reason to suppose that, in the case of this animal, resort to crossing might also result in considerable improvement in the milk yield. Very little attention has been given to the subject and, in view of the hardy character of the animal, and the great need for increasing the milk supply, the possibilities of the improvement of milking strains should be explored. With the object of improving the quality of the hair, suggestions have been made from time to time that the Angora goat should be tried for crossing. So far as we have been able to discover, this has not been done in India, at least on any considerable scale. This breed has been used with highly satisfactory results in South Africa and elsewhere and we recommend trials of it in this country. It may be found possible to grade up to the pure Angora strain. In that case, improvement would be easy and straightforward; but, if the same difficulties arise as in the case of sheep, a beginning must be made with a selected strain of Indian she-goat. Fortunately, in some hill districts, there is reason to believe that relatively pure bred goats are to be found. On the plains, they may be expected to be as mongrel in constitution as sheep, and would have to be dealt with in the same way. OF TOTAL NUMBER CATTLE BUFFALOES. Cattle and Buffaloes 160. We now turn to the main subject of this chapter. In no part of the evidence we received was the disposition to AND generalise more marked than in the replies relating to cattle, their quality and their management. The great importance of cattle, the miserable condition of so many of them, the obvious difficulties encountered by stockowners, all tended to induce general statements from which it was not always an easy matter in cross-examination to disentangle facts. So many factors have to be taken into account in considering the improvement of cattle that we feel a preliminary review of the position is called for, before we proceed to formulate our own conclusions. With this object we have examined the livestock census statistics for each province, as well as certain other figures which assist in explaining the difficulties of the Indian stockowner. We recognise that there are imperfections in the published figures, but we believe they are sufficiently reliable to deserve careful analysis by those responsible for cattle improvement. For obvious reasons, we cannot treat them in the detail that they deserve. We attempt no more than to discover certain broad facts concerning the distribution of cattle and the main factors responsible for differences in the total number of cattle found in different parts of the country. Thirty years ago, the cattle census figures, which were then very imperfect, accounted for fewer than ninety millions of cattle and buffaloes in British India. Ten years later, some 115 millions were enumerated. Improvements in the methods of collecting statistics were introduced about this time and there was a steady increase in the numbers, until, by the year 1914-15, 147 millions were recorded. It is clear that, up to this time, although an increase may actually have taken place, the main reason for the higher figures recorded was the expansion and improvement of statistical work. In the last ten years, the fluctuations have been no more than are to be expected in a country where the livestock depend largely on the produce of widely varying seasons. The number of cattle recorded in 1919-20 was 146 millions and, in 1924-25, 151 millions. DISTRIBUTION OF CATTLE 161. From the census returns, it is obvious that the distribution of cattle is very unequal. Inequalities depending on the nature of the land to be tilled, the extent of irrigation from wells, the amount of scrub and jungle, the rural population and the size of the holdings are indeed to be expected, but, when all allowances have been made for such causes, the census figures show that surprising differences exist between provinces. The provinces vary so widely in extent and the total number of cattle is so large, that a valid comparison of the livestock position is not easily made, unless the figures are reduced to a common denominator. Of the various common denominators open to us, we believe that, for India as a whole, the best basis of comparison is "per 100 acres of net sown area" which is the basis adopted in the "Agricultural Statistics of India." The statement given below shows, for each of the major provinces and for British India as a whole, the total numbers of cattle, the relation between that number and the human population, and the number of ordinary cattle and of buffaloes maintained for every one hundred acres of net sown area. An estimate of the total area of natural grazing land in each province is also included, This has been arrived at in a manner which is explained subsequently. As for livestock, this grazing area is stated per 100 acres of net sown area :— Distribution of cattle and buffaloes in the major provinces of India and in British India in 1924-25 The great difference between province and province in the cattle population, especially in the number of ordinary cattle, is very remarkable, and an examination of the estimated areas of natural grazing land shows that the figures under this head throw no light on the reasons for it. In paragraph 164 below, we separate the cattle into six different classes and we shall then refer again to their numbers. Before we do so, it is necessary that we should explain the method we have adopted in estimating the grazing area, and should also discuss the quality and the value to the ordinary cultivator of the great so-called grazing areas nominally at the disposal of cattle in India. It will also be necessary to refer to the value of the statistical returns from which the livestock figures are derived. The estimate of the natural grazing land available in each province has been arrived at by adding to the area of forest land open to cattle grazing, three-fourths of the area of culturable waste and one-fourth of the area of unculturable waste found in each province. It should, however, be noted that no inconsiderable part of the fodder which the cattle pick up for themselves is to be found not on the areas which we have termed natural grazing lands in the statement given above, but on cultivated fields after the crop has been harvested, and on current fallows. Indeed, in many parts of the MO Y 286-12 country, the weeds growing on cultivated land, the grasses on field borders and along water channels, the cultivated plants which spring up from seed falling before harvest, and the stubble of crops, furnish the main grazing available to cattle. CHARACTER OF GRAZING LANDS IN INDIA. 162. Agriculturists unacquainted with conditions in India would at once challenge the basis of the comparison we have made in the above statement; their attention would be attracted to the vast areas of uncultivated land found throughout this country, and, on the analogy of the experience of stockowners in temperate climates, they would advance the opinion that since grazings form the natural food of cattle, the grazing area, rather than the tillage area, should be used in comparing the numbers of cattle existing in each province. The vast uncultivated tracts of India do, undoubtedly, in the aggregate, afford a large amount of grazing, and, in certain brief periods of the year, there is an abundance of grass; but, except on the common grazing land near villages, where the early grass is devoured by starving animals, and the later growths never get a chance to develop into anything to which the term "pasture" could be applied, the growth of grasses is extremely rapid; and they quickly become unpalatable to stock. Were these grasses to be cut and stored either as hay or silage, at a suitable stage, great quantities of cattle food could be provided; but for many reasons, which need not for the moment be given, they are rarely stored for use; and when the dry months of the year set in, the class of fodder to be found on grazing land is very poor. The long spells of dry weather hinder the development of the smaller grasses and prevent the formation of a close sward. The hard and dry grasses can indeed keep life in the cattle which have access to them, but very rarely can they keep cattle in good condition. As compared with grazing lands in temperate climates, Indian grazing lands are of very little use to the cultivators of tillage land, even when they are readily accessible, not because the grasses are bad but because they are only abundant and of high feeding value for a very short period in the year. At the time of year when the grower of field crops most needs fodder to supplement his own supplies, the fodder to be found on natural grazing land has little value for stock feeding. But, poor as grazing lands generally are from the point of view of economic stock husbandry, exceptions are to be found, and it is unquestionable that grazings influence both the numbers and the management of the livestock of the country. It is in certain of the grazing areas that the better cattle of the country are bred, and the conversion of the best grazing land of the past into tillage land to meet the needs of a growing population has probably increased the difficulty of maintaining the quality of cattle. The natural grazing lands of India are, as we have seen, to be found within the areas classed for statistical purposes as "forests," ""culturable waste" and not available for cultivation." A large proportion of the land which is classed as forest is open to grazing, either throughout the year or for that part of it when grazings are of most value. We have assumed that the whole of these open forests are available for cattle. Though grazing fees are charged, they are very low, and do not exclude cattle. We have not made a deduction from the total area of forests open to grazing, as we have in other cases, because, although we recognise that much of the area so classed can produce little or no grass, and much is inaccessible, we think that the average quality of the open forest areas is substantially better than that of the average land included under "culturable waste," and very much higher than that of land "not available for cultivation." Though the whole of the land falling within the category "culturable waste" is open to grazing, there is much of it which, though nominally "grazing area," does not produce any useful herbage. In some districts, there are extensive areas of laterite soils yielding nothing of value; in others, prickly pear, bushes of many worthless kinds such as lantana, and coarse herbaceous weeds cover much of the ground. In assuming that three-fourths of the culturable waste is available as grazing, we believe that we have adopted a full figure. The grazing value of land "not available for cultivation " is still more difficult to assess. A very large proportion of it, because rocky and distant, is quite inaccessible to cattle, or it may be absolutely barren from absence of soil in the hills or of water in the plains. Its general character is such that we do not think it likely that more than one-fourth can be classed as grazing land. For British India as a whole, we are of opinion that the total of 209 millions of acres which we have arrived at by the method explained above may be regarded as an approximation which over-estimates rather than under-estimates the extent of the natural grazing land accessible to cattle. As the quality of grazing land varies widely, we recognise that the totals arrived at in this way for each province cannot be regarded as strictly comparable. The figures supply, indeed, but a very rough index of provincial resources in the matter of grazing land, but we think them sufficiently near the mark to prove that the extent of natural grazing land available does not explain the differences which exist in the number of cattle maintained in the different provinces. VALUE ΤΟ BE ATTACHED ΤΟ STA 163. These differences are, indeed, so remarkable that they necessitate some reference to the trustworthiness of the statistical figures. In the permanently settled provinces TISTICAL FIGURES. of Bengal and Bihar and Orissa, where no village accounts are kept, and where the task of collecting the original livestock figures falls, not on a village accountant, but on a village watchman, the figures are admittedly much less reliable than in provinces where they are collected by revenue officials. Even in provinces where the village accountant is employed, we have had evidence that the work of enumeration is sometimes very imperfectly done. On the other hand, it may be pointed out that the census has now been taken MO Y 286-12a |