In contrast with the abundant data available on the composition of cow's milk, information on ewe's milk is relatively sparse. Among the more extensive analyses which have been published are those

1. Observations were made during the autumn and spring of 1959–60, 1960–61, and 1961–62 on pregnant and lactating Clun ewes of mixed ages grazing under natural conditions. The pattern of food intake during pregnancy and lactation, the output achieved in the form of weight of lambs produced at parturition, milk yield, and bodyweight change of the ewes themselves was investigated. The interrelationships between these factors, as influenced by the feeding of supplementary concentrate, either for 7 weeks before or 7 weeks after lambing, or during both periods, was elucidated.2. Daily dry-matter intake was found to be of the order of 1·5–2·0 kg. for this type of ewe, but there was considerable individual variation (coefficient of variation about 20%). Intake did not increase concurrently with the demands of the foetus during the later stages of pregnancy, and even became depressed as parturition approached. Immediately after lambing there was a spectacular increase in intake, which, after rising to a peak, was maintained at a high level during the lactation. Ewes receiving supplementary concentrates showed an increase in food intake over those unsupplemented, but this increase was less than the amount of dry matter fed in the supplement, even on rather bare pasture.3. During the early stages of pregnancy there was no apparent difference in the weight gains of singleand twin-bearing ewes, but during the last 8 weeks prior to lambing twin-bearing ewes made greater weight gains. In relation to weights taken just after parturition ewes producing twins had lost most weight. Unsupplemented ewes lost more weight than those receiving supplements. Lack of supplementation prior to lambing only limited production in the case of twin-bearing ewes.

This study has been directed towards finding a reliable and simple method of estimating the food intake of ewes grazing under natural conditions in the field, and relating the findings to the production of live-weight gain, milk and lamb. The method evolved for measuring the food intake of the grazing ewe is described. Part II deals with the measurement of production, and the relationship of food intake to productive output together with the factors influencing these relationships.The best approach to the problem of estimating the food intake of grazing animals has been the study of faecal output and, since the work of Garrigus (1934), several such studies have been made.To measure food intake in absolute terms, using a wide range of foods, not only must faecal output be determined but also the food/faeces ratio or digestibility of each food. This latter determination has presented a difficult problem. Evidence from the present study suggests that when the primary consideration is the elucidation of differences between animals on the same pasture in relation to production, as distinct from the problem of pasture evaluation, the lack of a very accurate measure of food/faeces ratio is less of a handicap than it might appear.

A METHOD of estimating grazing intake is essential for studies of feed utilisation in ewes under natural conditions. Since the ewe is a close and selective grazer, methods involving the cutting of comparable grazed and ungrazed areas of the pasture are unlikely to allow reliable estimation of pasture intake. It seems likely that only by a study of the faeces produced can reasonable estimation of individual intakes be made. The first requisite for such estimations is an accurate method of estimating faeces output. Sheep, with their relatively lower and drier faeces output, present fewer problems in total faeces collection than do cattle, which produce large quantities of faeces of a low dry-matter content. The collection of faeces, free from urine, presents a special difficulty in female animals.The estimation of intake with the indirect " faecal index" methods generally involves the estimation of faecal output with indigestible marker materials, and the estimation of the concentration of faecal components, the relationship of which to feed/faeces ratio has previously been determined by means of controlled feeding trials indoors. Intake estimates derived by this combined method are subject to substantial errors, both in the estimation of faecal output (Lambourne, 1957) and in the estimation of feed/faeces ratio (Raymond and Minson, 1958). Even with recent refinements as described by Brisson and Pigden (1957) and Corbett, Greenhalgh and McDonald (1958) the error involved in the estimation of faecal output is unlikely to be reduced to a level comparable with that possible by means of total faeces collection.As a preliminary to studies on feed utilisation by grazing ewes the development of new techniques was therefore found necessary by which (a) total faeces collections could be made, and (b) dry matter intake estimated from them. The equipment which has now been developed for total collection from the ewe is described in this paper; the method of estimating dry matter intake from faeces output has been briefly reported (Owen, 1960) and the results obtained in these studies will be presented more fully later.

This paper reports a continuation of previous work on the mineral constituents of ewe's milk, in particular copper, iron, manganese and zinc. Milk taken at weekly intervals from four groups of Clun Forest and Suffolk x Chin Forest ewes was studied over lactations varying from 6 to 12 weeks. The copper content showed a marked decline as the lactation advanced, while there was no consistent trend in iron content. Maximum, minimum and mean values for iron and copper are given for the sheep in each group.Manganese was determined in the milk from two groups and zinc in. that from one group of ewes. There were significant differences in weekly manganese content in both groups, but no consistent trend as lactation advanced. Zinc content showed no significant differences from week to week, but the milk of one ewe was consistently very much lower in zinc than that of the other five sheep.