In the investigation described in this third part of the work, we have studied the effect of imposing two extreme levels of diet upon the pregnant ewe during two periods, one of early and one of late, gestation. In order to build up a comprehensive picture of the effect of these treatments we have been concerned, first, with the changes that they have wrought in the tissues and organs of the mother's body; secondly, with the effect upon the growth and development of the genital tract and its major components, and lastly with the way in which the growth of the tissues and organs of the foetus have responded to the different nutritive levels. While the latter has been our primary concern, the former have been studied in some detail so that we may view the effects produced as a whole, and gain an understanding of any competition that may have occurred between maternal and foetal tissues.

The changing of treatments during the course of an experiment is discussed both from the point of view of designing trials in which changes are anticipated and of modifying trials that have fulfilled their original purpose. Cases are distinguished according to whether the interactions of original and new treatments are or are not important and recommendations are made for each case.

We are indebted to Miss J. M. S. Thom for preparing the diagrams fot publication.

1. The ventilation of two byres, one of large (1248 cu.ft.) and one of small (567 cu.ft.) cubic capacity per cow, was studied. The CO2 concentration in the atmosphere was used as a measure of the rate of air change and the dry katathermometer to determine the cooling power of the atmosphere. Temperatures and relative humidities were recorded and the actual air movements were studied by using smoke trails. The investigations were made under a variety of weather conditions and with a wide range of outlet and inlet areas.

2. It was shown that the atmospheric conditions in the large byre were almost always superior to those in the small byre, and this was undoubtedly due to the greater cubic space per cow in the large byre. Adequate ventilation was, however, readily attained in the small byre provided the outlet area was sufficiently large.

1. The importance of uniform production of herbage for crop-drying is emphasized.

2. A replicated experiment is described which compared four crops, two annuals and two leys under four manurial treatments which included a control and three applications of fertilizer (3½ cwt. ‘nitrochalk’) distributed during the season in three ways.

3. Vetches yielded 2500–3000 lb. dry matter per acre of 28–30% crude-protein content in two cuts.

Barley yielded 2500–3000 lb. dry matter per acre of 18–21% crude protein in five to six cuts.

Leys yielded 4000–5000 lb. dry matter per acre of 15–18% crude protein in four cuts in their first year, and 5000–6000 lb. dry matter per acre of 15–18% crude protein in five cuts in their second year.

An experiment using eight pens each of twenty-five pullets was conducted to investigate the suitability of sesame cake meal as a protein concentrate for laying pullets. In meal mixtures supplying equal amounts of crude protein, sesame cake meal gave slightly lower production than fish meal and earthnut cake meal, but the differences were of doubtful statistical significance. Body weights appeared to be maintained less well on sesame cake meal, but only by an unimportant amount.

The authors thank Prof. R. G. Baskett for his advice and constructive criticism, and Miss M. E. Osborne for the care and attention devoted to the management of the pullets.

Changes in the percentage of dry matter, sugars, starch and nitrogen present in Arran Banner and Majestic potato tubers during storage in clamps until July or later were studied in two seasons. In the first season, the ascorbic-acid content and the distribution of nitrogen between three fractions (insoluble, soluble coagulable, soluble non-coagulable) were also determined. In the second season, weighed samples of tubers were introduced into the clamps, so that changes in absolute amounts of the different constituents present in the tubers or sprouts could be measured. The effects of removing or retaining the soil cover on the clamps after early April were compared.

A method is described for economizing in the use of fertilizers on deficient soil by soaking the cereal seeds in nutrient solutions containing the deficient element at a sufficiently high concentration to supply a large part of the total requirements of the plants.

The result of greatest potential practical importance was obtained on oats. Soaking oats in molecular tribasic potassium phosphate brought about an increase in yield from 17 to 25 cwt./acre (46 %) on phosphate-deficient soil at the cost of 13·6 lb./acre of the potassium phosphate (or 4·6 lb. P2O6) without the addition of any phosphate to the soil. There was a correspondingly large increase in the weight of straw. Nearly 75% of the P2O5 introduced in this way was recovered from the harvested grain. Sodium phosphate appears equally efficaceous for oats. Strikingly beneficial effects were also obtained on barley and wheat.