In cross-over experiments it has been established that milk yield increased and the fat solids-not-fat (SNF) contents of the milk decreased when cows were given approximately 15 lb dry matter/day of early-cut silage compared with the values obtained by giving the same amount of silage made from herbage cut at a later stage of maturity.

Increasing the proportion of concentrates to silage in the cow's diet so that the milk yield remained unchanged increased the SNF content of the milk but had no effect on the fat content.

Supplementing early-cut silage with 5 lb hay/day increased the fat content of the milk, but had no effect on either milk yield or the SNF content of the milk.

Compared with a silage to which molasses had been added, untreated silage decreased milk yield, and increased the fat content without affecting the SNF content of the milk.

Solutions of labelled benzyl mercaptan-(35S) in water and cream were subjected to steam distillation in a continuous vaporization equilibrium still, and the concentrations of mercaptan in distillates and residues were obtained by measurement of the radioactivity in toluene extracts.

In the steam distillation from aqueous solution the vapour/liquid equilibrium relationship was not linear. The vapour/liquid equilibrium coefficient increased from 73 to 108 over the range of concentrations 0·01–0·05 ppm. in the liquid fraction at the time of partition.

In the steam distillation from cream containing 30% fat the relationship was linear, with an equilibrium coefficient of 1·35 over the range 0·1–0·6 ppm. in the cream at the time of partition.

The values of the vapour /liquid equilibrium coefficients are of the same order as those reported previously, 105 and 1·30, respectively, for which estimations were made by an absorption method.

The butterfat/water partition coefficient for benzyl mercaptan was found to be 160, which is consistent with the large difference found between the steam volatilities of the substance from water and from cream.

The effect of incomplete milking or of an extended milking interval on the yield and composition of cow's milk has been investigated. At the 1st milking after either treatment, the yield of milk was augmented by a carry-over of milk but at the 2nd milking the yield was invariably depressed and then increased until the original yield, or slightly less, was recovered within 5 days. The concentration of lactose and potassium decreased and of sodium, chloride, whey proteins and casein increased during the period of the treatment, and on the resumption of normal milking the original composition was rapidly recovered. The changes in the composition of fat were less clearly defined because of residual milk effects but over the period of the treatment and the succeeding 6 milkings the effect was invariably to increase fat percentage. The differing effects on composition resulted in a more marked effect on the yields of lactose, potassium and casein than on those of sodium, chloride or the whey proteins throughout the treatment and recovery periods. When milk was removed in portions after an extended milking interval there were decreases in the concentrations of lactose and potassium and increases in those of sodium and chloride throughout normal milking. When the residual milk was obtained after an injection of oxytocin the composition of the portions of milk resembled that of the first portions of milk removed from the udder. The possible effects of the treatments on the secretory cells and the permeability of the udder tissue are discussed.

The changes in yield and composition of milk when 4 cows were milked throughout the whole of pregnancy have been studied. The yield decreased to a minimum 1–20 days before parturition and then gradually increased. The changes in composition before parturition were similar to those observed in late lactation but more pronounced. The content of proteose-peptone plus globulin increased in late pregnancy with a sharp peak, of about 5 times the value observed in mid-lactation, within the last 10 days before parturition, and then decreased rapidly. After parturition milk yield increased rapidly and the changes in composition were less marked than in animals with the usual ‘dry period’. The results are discussed in relation to current knowledge of colostrum formation.

After an injection of oxytocin and the removal of residual milk, the milk obtained at the next milking had an increased content of sodium, chloride and whey proteins and a decreased content of lactose. The effect persisted for several milkings. It is suggested that this was due to a direct effect of oxytocin on the permeability of the mammary epithelium and that oxytocin should be used with caution in certain types of milk secretion experiment.

Samples of butter were reworked at various times after removal from the churn, either in a laboratory mixer or by pressing between a microscope slide and cover glass. Moisture distribution was studied by microscopic examination and by indicator papers.

Reworking within 30 min of completion of the original working caused no change in moisture distribution but delayed reworking caused aggregation of water droplets. The further the setting of the butter had proceeded before reworking, the greater was the extent of aggregation.

Cream-cooling treatments which resulted in a softer butter delayed or decreased the aggregation of droplets caused by reworking.

A theory is suggested relating changes on reworking with the formation of a crystal structure on setting.

Cream-cooling rates, cream-pre-cooling treatments (Alnarp) and seasonal changes in the chemical composition of butterfat were found to influence the setting rate of butter. Vacreator pasteurization, however, especially at high steam intensity, had the greatest effect. Comparative hardness values of butter made from platepasteurized and Vacreator-pasteurized cream were 1·5 and 2·9 kg and 4·4 and 8·3 kg at the end of ½ and 5 h, respectively, after the completion of working. The highest setting rate, which resulted in a hardness of 6 kg at the end of 30 min, was recorded for butter made from Vacreator-pasteurized cream which was rapidly cooled to 5 deg F below normal churning temperature and churned after holding for 30 min. Abrupt breaks noted in the setting curves of this rapidly setting butter lead to speculation that changes in the butterfat crystal form may be involved

In general, the final hardness values of butter stored at 45 °F for 30–50 days were not greatly influenced by the setting rates observed at 55 °F during the 5-h period immediately after the completion of working. Exceptions to this observation were obtained when the cream was rapidly cooled to or slightly below churning temperature and also when the cooling of plate-pasteurized cream was delayed at 65 °F for 3–4 h before final cooling to the churning temperature.

The causes of the differences observed in the setting and final hardness values of butters made from plate- and Vacreator-pasteurized cream were not determined and warrant further investigation.

A simple, quick, batch-type method for the laboratory-scale HTST pasteurization of milk is described and details of construction, operation, cleaning and sterilization of the apparatus are given. The heating and cooling curves approximate closely to those of commercial HTST milk-pasteurizing plants and the end-product, as measured by residual phosphatase, is similar. The method has a relatively high sample throughput and is suitable for the routine pasteurization of large numbers of milk samples of up to 300 ml in volume. Other possible applications of the method are discussed.