1. The investigation was for the purpose of ascertaining the cause of the variations previously noted by the author as occurring in mare's milk.

2. Determinations have been made of the composition of the milk of eight Shetland ponies and one Clydesdale mare throughout the lactation period, with observations on the milk of numerous thoroughbred mares. The constituents estimated were the total solids, solids-not-fat, protein, fat, lactose, total ash, CaO, P2O5, and chlorine. The CaO, P2O5 and chlorine percentages in the ash of the milk were also determined. The freezing-point and pH were recorded in a number of samples. The animals were kept under natural conditions at grass.

3. It was found that as lactation advanced the percentage of total solids, lactose and fat of the milk increased in a linear manner, whereas the percentage of solids-not-fat, protein, CaO, and P2O5 decreased linearly.

4. The regression of the total solids and the chlorine was parabolic in character as also was that of the chlorine-lactose index.

1. The means and standard deviations of analytical data obtained from fifty-four samples of mixed milk (alternate a.m. and p.m.) and whey from the College herd of Dairy Shorthorns are presented in Table I. The samples were taken at intervals during the period late September 1935 to November 1936.

2. While no outstanding differences in composition are evident between morning and evening samples, the time of rennet coagulation was appreciably greater with the evening's samples. Evidence is presented to show that the longer time of cold storage of these samples must be regarded as a contributory cause of this effect.

3. In the discussion of the mean values, special treatment has been given to the partition of calcium and phosphorus between the four phases, soluble and insoluble, organic and inorganic. The caseinogen values are examined in the light of current theories.

1. The effect of heating on the creaming power of milk and the role of truly hydrophilic colloids in increasing the creaming power of raw and of heated milk are discussed. The suggestion is made that a proper understanding of changes underlying the reduced creaming power of heated milk has been handicapped by a lack of reliable data on the extent to which the hydrophilic colloid content of milk is reduced by denaturation of albumin and globulin.

2. Information is obtained concerning the relationship between the denaturation of albumin and globulin and the reduction in creaming power from a series of determinations of the amounts of protein denatured, and of the changes in creaming power produced, by heating samples of the same milk for 30 min. at temperatures of 57·5, 60·0, 61·5, 63·0, 65·0, 67·5, 70·0, 74·8 and 80·0° C.

1. Experiments were carried out during the years 1931–5 in which mice on a basal diet of raw milk and white-flour biscuit were compared with mice on a similar diet except that the raw milk was replaced by commercially holderpasteurized (145–150° F. for 30 min.) milk belonging to the same batch. Each group of mice comprised in aggregate about 1100 animals, which were usually 4–6 weeks old at the commencement of observation. For reasons given on p. 206 no attempt was made to standardize the volume of milk consumed by the individual mice. Except in the preliminary experiments, when the milk was given in excess, the amount supplied was regulated more or less in accordance with the requirements of the average animal.

1. In a series of virulence experiments a comparison was made between the resistance to infection of 551 mice fed on raw and 495 mice fed on pasteurized milk, each dietary being supplemented by a mineral mixture and a whiteflour biscuit. In one set of experiments the mice were inoculated intraperitoneally with about 100 or 1000 living Bact. typhi-murium and observed for a fortnight; in another set they were fed by the mouth with about 100–200 million living organisms of the same strain and observed for 4 weeks. All animals dying during, and all survivors killed at the end of the experiments were bacteriologically examined for evidence of specific infection.

2. In the intraperitoneal inoculation experiments the results obtained with the raw and pasteurized groups respectively were as follows: Proportion of animals dying of specific infection 51·83 and 51·20 per cent; proportion of all animals found to be specifically infected at death or when killed 95·29 and 91·57 per cent; average survival time of animals dying of specific infection 8·06 and 6·93 days. The proportions of animals dying of specific infection when standardized for weight distribution were 55·1 and 50·6 per cent respectively.

Various strains of lactobacilli isolated from Cheddar cheese were added to the milk employed in cheese manufacture. Different strains of these organisms were found to impart specifically characteristic flavours to the cheese. The range of flavours developed included both desirable and objectionable types, many of which are commonly encountered in commercial cheese, e.g. “pleasant aromatic”, “diacetyl-like”, “metallic”, “bitter”, “unclean” and “fermented”.

The inoculation of cheese-milk with natural mixtures of lactobacilli from mature cheeses led almost always to the development of unclean “fermented” flavours. The causative strains of lactobacilli were isolated and shown to be of frequent occurrence in normal mature cheeses but under ordinary circumstances the actual numbers of these bacilli present in most cheeses were too small to spoil the flavour.

It has been shown by experiments in milk that the enzymes of commercial rennet in conjunction with the lactic acid bacteria occurring in Cheddar cheese can bring about protein breakdown similar in extent to that found in the ripe cheese as far as the non-protein nitrogen is concerned. The amino nitrogen produced is, however, much less than in cheese. This may be ascribed to the higher pH of cheese as compared with that of the milk cultures, since acidity adversely affects the peptidases present. Attention is drawn to the differences between the conditions.

Rheology is denned as the science of the deformation of matter. It includes the study of all deformations, recoverable (elastic) and non-recoverable (plastic). In this paper cheese is primarily considered although the fundamental principles apply to all materials.

The elastic and plastic properties of cheese exercise a great influence in the ripening processes and the incidence of faults. Although much progress has been made in the bacteriological and chemical investigations of cheese ripening, little fundamental work on the nature of and the factors controlling the physical properties of curd and cheese has been carried out.