Two procedures are described for ascertaining the relationship between the freezing-point and moisture content of a soil. Since the soil sample is dried in the process of freezing, it is necessary to estimate how much water has been frozen out of the soil at the moment when the freezing temperature is recorded.

Procedure A embodies all the precautions which appear desirable when the greatest accuracy is required. Procedure B is simple and rapid and yet accurate enough for routine estimations.

1. The principal change in the composition of adult fowls during fattening is an increase in the fat content of the body, the protein increase being subordinate in amount.

2. Within the limits studied, the efficiency of conversion of food energy to carcase energy is not affected by variation in the protein content of the ration.

3. A food mixture consisting of 5% dried skim milk of 95% Sussex ground oats proved as efficient for fattening fowls as a food mixture consisting of 20% dried skim milk, and 80% Sussex ground oats.

4. The food mixtures used gave satisfactory results in fattening when used in pellet form in a dry state.

The results obtained in the preliminary investigation were entirely confirmed, the pF at the “wilting coefficient”, as measured by the modified freezing-point method, varied from 4·0 to 4·4 (round figures), with an average of 4·2.

The variation observed bears no relation to the soil texture, neither can it be explained by uncertainties in the freezing-point determinations which have proved to be accurately reproducible. Freezing-point measurements after leaching, conductivity measurements and freezing-point determinations in saturated soil and at the moisture equivalent proved that part of the variation is due to the presence of soluble salts, the more saline soils having a higher pF at the “wilting coefficient”. When the salt content does not exceed about 500 p.p.m. the influence of the salts is hardly detectable, and the pF at the “wilting coefficient” lies between 4·0 and 4·3. Besides unavoidable errors in the wilting experiments several other factors may account for this variation. They are all the factors that have any role in the “history” of the soil. In view of these uncontrollable sources of error a variation of 0·3 pF units can be considered very small.

It can therefore be confidently concluded that in ordinary agricultural soils with a salt content of less than about 500 p.p.m. permanent wilting occurs when a critical pF value (lying between 4·0 and 4·3) is reached.

The gaseous exchange of the sheep can be measured in the apparatus described with considerable accuracy and good agreement obtained with the energy absorbed from a maintenance diet. The apparatus automatically compensates for diurnal changes in temperature and humidity, and maintains a constant rate of ventilation.

The precautions necessary to ensure that the metabolism as measured is a good sample of that of the animal when subsisting on the diet, consist of control of temperature and humidity within the chamber and training the animal so that it is free of psychic strain during the period of measurement. It was found that the sheep takes 14–20 days to adjust its energy turnover to a change in diet.

Modification in structure and operation of the Haldane-Carpenter gas analysis apparatus are described in detail.

The method of A. C. Andersen is used to compute the energy metabolism and respiratory quotient from the gaseous exchange and methane production.

Critical estimations of the recovery of sulphur, when materials such as fodder and excreta were subjected to a number of analytical procedures. The only method in this series which resulted in consistently good estimations was one involving oxidation in compressed oxygen confined in an Emerson Calorimeter Bomb.

Measurements of the gaseous metabolism of sheep in this laboratory are conducted by drawing air through a sealed chamber that contains the animal, sampling the effluent air, and analysing the sample for its carbon dioxide increase and its oxygen deficit. As the analysis is dependent upon the measured fraction of nitrogen with rare gases, and particularly in respect of the oxygen deficiency this fraction must be known rather precisely. It is of course assumed that the animal in no way changes the quantities of nitrogen and rare gases passing through the chamber, but the analysis is incomplete if any gases or vapours other than carbon dioxide and water are given off.

It is well known that considerable quantities of methane, and possibly some hydrogen, are produced by ruminants, and some forms of gas analysis apparatus include a device designed to measure the concentration of methane in the sample. The device, as used in the apparatus of Carpenter & Fox (1926), consists of a platinum spiral heated electrically, the carbon dioxide resulting from combustion being measured volumetrically in the ordinary way, together with the O2 consumed. Experience with the device in this laboratory has led to its abandonment and gravimetric methods are now employed instead.

Attention has been directed to the occurrence of acetonuria in apparently healthy and in diseased cattle.

It has been shown possible to cause an alimentary acetonuria in milch cows experimentally by feeding large quantities of grass silage of the butyric acid type. The reaction of the urine remained alkaline and the ammonia content was within a normal range, showing that there was no real acidosis. The animals seemed by no means to be ill.

As to the cause of the acetonuria, a priori, there seemed to be little doubt that in these cases the ketones resulted, in the animal body, from the incomplete oxidation of the butyric acid of the grass silage. Yet, further experiments with a lactic acid silage containing no butyric acid at all indicated that the question may be more complex.