Of several strains of lactic acid bacteria which produced nisinase, Streptococcus thermophilus, strain TJ, was the most active. The optimum pH for extraction of the nisinase from an acetone preparation of bacterial cells and for incubation of the solution containing nisin and nisinase was in the neighbourhood of 7·0. The proportions of nisin destroyed by the nisinase under various conditions were determined. Also, it has been shown that nisinase has no action on a number of antibiotics, including subtilin.

In the years 1959, 1960 and 1961 sections of the College herd of pedigree Guernsey cows were fed indoors on cut grass. The grass, cut twice each day, was fed in excess of appetite.

The weight of grass dry matter consumed varied significantly in each year with the percentage of dry matter in the grass. In one year only did the dry-matter consumption vary significantly with the percentage of crude fibre in the grass.

The weight of drinking water consumed varied significantly with the percentage of dry matter in the grass. The total water intake, i.e. drinking water plus uncombined water in the grass, remained almost constant at approximately 14 gal per cow per day.

The use of a concentrate supplement in 1960 and 1961 produced small increases in yield which were not statistically significant. If, however, the differences recorded were real, then they would have been economic in one year and not in the other.

The intake of starch equivalent, calculated from the chemical analyses of the grass, could not be reconciled with the performance of the cows.

The viscosity and flow properties of fresh and age-thickened separated sweetened condensed milk have been investigated using a rotating cylinder viscometer. The rheological properties, and the changes therein brought about by stirring, were studied by means of flow curves obtained by plotting stress against the corresponding rate of shear. Condensed milk during age-thickening is shown to develop thixotropic properties, with a high initial viscosity or ‘gel strength’ which, however, is easily reduced by stirring. On prolonged stirring the viscosity reaches a minimum (the ‘permanent’ viscosity), which becomes progressively higher as the age-thickening process continues. On storage subsequent to stirring there is only an insignificant recovery of the lost viscosity at room temperature, but a substantial recovery at higher temperature.

Sweetened condensed whey shows no age-thickening, and the casein-free supernatant liquids obtained by centrifuging dilutions of both fresh and age-thickened condensed milk have the same nitrogen contents and the same viscosities. This is interpreted to mean that the whey proteins play no significant part, directly or in association with casein, in the age-thickening process. Also the sedimentation rate of the casein during centrifuging is greater with age-thickened than with fresh condensed milk, which indicates an increase in casein particle size during agethickening.

The following working hypothesis is suggested to account for this behaviour. During age-thickening a slow irreversible change occurs in the size or shape of the casein micelles, probably by aggregation, which produces the rise in ‘permanent’ viscosity. These micelles orient themselves in such a way as to form a loose network enclosing some of the dispersion medium, thereby producing the large increase in the initial viscosity. On stirring, this network is broken down, and although it can reform in time at high temperatures, such recovery is greatly retarded at room temperature by the high viscosity of the aqueous phase containing the soluble constituents of the condensed milk.

The rate of age-thickening of separated sweetened condensed milk increased with decreasing condensing temperature, over the range 55–22°C. Preheating the milk for 10 min at 85°C reduced the destabilizing effect of low condensing temperatures.

Changes take place after condensing which have the effect of decreasing the rate of age-thickening during subsequent high-temperature storage. These changes are complete in about 2 weeks at room temperature but take longer at lower temperature.

The hypothesis is put forward that the stabilizing effects of a moderately high condensing temperature, and of cool storage after condensing, are connected with the precipitation of insoluble salts, especially calcium phosphate. The effect of preheating is ascribed to the formation of nuclei of precipitated calcium phosphates which facilitates further precipitation during and after condensing.

Experiments on age-thickening at a series of different temperatures with four batches of commercial condensed milk gave mean values for the temperature coefficients of increase in ‘initial’ and ‘permanent’ viscosity of 4·9 and 7·5 per 10 degC, respectively. The difference between these values is thought to support our previous conclusion that two different processes are involved.

Age-thickening of sweetened condensed separated milk can be greatly reduced by removing part of the diffusible salts by dialysis of the milk before processing. This treatment is still effective after high-temperature preheating, which indicates that denaturation of the whey proteins does not influence age-thickening independently of the milk salts.

When added at the level of 20 m-equiv./kg of condensed milk to the milk before condensing, those polyvalent anions (phosphate, oxalate, citrate, ethylenediaminetetra- acetate) which have the property of forming undissociated or insoluble calcium salts greatly increase the rate of age-thickening. Fluoride, which forms an insoluble calcium salt but is monovalent, has the opposite effect.

It is suggested that both natural and added polyvalent anions destabilize the condensed milk by forming linkages between the calcium atoms of the calcium caseinate micelles, producing the larger, more asymmetric units which were postulated in Part I to account for the viscosity changes during age-thickening. Fluoride would have the opposite effect by combining with the calcium, and so preventing such linkage formation by the polyvalent anions.

When added to the milk at the same level as the anions, divalent cations, especially calcium and zinc, reduce the rate of age-thickening. It is suggested that this is due to removal of milk-serum anions, particularly phosphate and citrate, as insoluble or undissociated salts.

The effect of both anions and cations added after condensing is almost the opposite of that produced by the same ion before condensing, with the exception that phosphate increases age-thickening in both cases. There is as yet no clear explanation for this observation.

The freezing point and pH of milk ultrafiltrate were different from the values obtained for the original milk. It was estimated that from 3·2 to 4·6% of the soluble material in milk was retained by the ultrafiltration membrane. Comparison of the composition of ultrafiltrate and rennet whey showed that the soluble material retained consisted mainly of calcium and citric acid. A similar effect was observed with dialysis, although the quantity of soluble material retained was much less. Rennet whey was considered more representative of milk serum than either ultrafiltrate or diffusate.

Classical methods and methods based on the use of disodium ethylene-diaminetetra-acetate (EDTA) for the semi-micro determination of calcium and magnesium in milk and milk diffusate are described. With both types of method satisfactory accuracy and precision are attainable. The use of the EDTA methods leads to greater speed and convenience but practice in recognizing the correct titration end-points is necessary.

In earlier papers it was shown that the fall in viscosity of milk and caseinate solutions acted on by rennet, follows a first-order reaction equation. The reaction rate constant has been found to be independent of the source of the caseinate substrate used. This seems to be true not only for caseinate from normal cow's milk but also from colostrum, slow-clotting cow's milk and even the milk of buffaloes, sheep and probably goats.

With solutions of caseinate from sow's milk the reaction kinetics and the proportional viscosity loss were similar to those of cow's caseinate but the rate constants were much lower. Rennet showed no measurable reaction with human, whale, mare or rhinoceros caseinates but produced a constant rate of fall in the viscosity of a camel caseinate for more than an hour from rennetting.

Vegetable and microbiological rennet substitutes also produce a fall in the viscosity of sodium caseinate and the reactions follow first-order kinetics. They differ from cheese rennet in that the proteolysis involved is stronger and is not as limited as that of rennin since they attack at least twice as much of the casein ensemble. This might suggest a mixture of enzymes but since the proportionality between enzyme concentration and k1 was very close this seems unlikely.

Milk and butterfat yields, and the properties of butterfats from the milks, were compared in three successive years for early- and late-calving cows of monozygotic twin pairs grazed on the same pastures, over the period when both cows of the twin pairs were in lactation.

The seasonal fluctuations in the properties of the butterfat were closely similar for the two groups of cows in all three trials. This was taken to indicate that the known regular pattern of seasonal change in the properties of New Zealand butterfat over all North Island districts is not due to change in stage of lactation of the cows but is probably the result of a regular change in the feeding conditions.

The average milk and butterfat yields were lower for the early-calving cows during the comparison period, the iodine values were higher, and the Reichert and saponification values were lower. The early-calving cows were uniformly heavier than the late-calving cows. It is suggested that the consistent differences in the butterfat characteristics over the whole comparison period could be due to differences in the extent to which body fat reserves were drawn upon by the cows for maintenance of lactation.

The average carotene and vitamin A contents of the butterfats were not appreciably different for the two groups of cows, indicating that these values are not directly related to stage of lactation of the cows or to yield of milk or butterfat.

A rapid method for the estimation of carbon dioxide in Cheddar cheese is described which gives good reproducibility and is free from major sources of error.

For the past 10 years, a herd of reindeer (Rangifer tarandus) has been kept in the Scottish Highlands under the auspices of the Reindeer Council of the United Kingdom. It was established in 1952 when the Reindeer Co. Ltd. imported two small consignments of reindeer from Arctic Sweden. By 1962 the herd numbered about thirty, of which twenty-one have been bred in Scotland.