Additions of urea progressively increased the heat stability of milk outside of its coagulation time (CT)–pH minimum. In the region of the CT–pH minimum larger amounts of urea were required before an increase in heat stability occurred. The effect of urea was observed over the temperature range 125–140 °Cfornaturalmilk, milk which had been dialysed against synthetic sera, and milk to which a sulphydrylblocking agent had been added. Urea additions did not affect the activation energy of the heat coagulation reaction or the stability of milk to rennet or ethanol.

Bovine milk-fat globule membrane was solubilized with a zwitterionic surfactant and subjected to chromatography on agarose, with the surfactant in the eluant. Fractions were tested for their effects on the oxidation of buffered linoleate. The maximum oxidative capability was greatly enhanced by the addition of Cu, and became associated with the phospholipids.

Further chromatography of the retarded protein peak from agarose on Sephadex G-200, again in the presence of surfactant, gave 2 protein peaks. Oxidative effectiveness resided almost entirely in the first peak, which was devoid of phospholipid, but high in xanthine oxidase activity. This fraction was subjected to isoelectric focusing, and the xanthine oxidase from this was highly pro-oxidative. Furthermore, its oxidative capability was almost doubled on heat treatment.

The enzymes chymosin, bovine pepsin A and bovine pepsin B in 40 commercial rennets were assayed by rocket immunoelectrophoresis, using monospecific antibody preparations against the 3 enzymes. Concentrations found were chymosin, 0–1208 mg/l; bovine pepsin A, 359–3818 mg/l; bovine pepsin B, 0–320 mg/l. Bovine pepsin B was found to be a minor enzymic component in bovine rennets, representing maximally 2·9 % of the milk-clotting activity or 9·7 % of the weight of the enzymes. The sensitivity for the quantification of the 3 enzymes was 0·5–1·0 mg/l. The coefficient of variation was 3·7 %. The sum of the immunochemical data converted into milk-clotting activities was in good agreement with the directly observed milk-clotting activity, indicating that there was no immunochemical overestimation due to the presence of non-enzymic active antigens. A high correlation was found between the enzymic composition determined immunochemically and by activity ratio assay.

Casein samples from untreated milk and stored ultra-heat-treated (UHT) milk were hydrolysed with pronase (protease ex Streptomyces griseus K-1) and subsequently with a mixture of peptidases prepared from the microsomal fraction of hog kidneys. Incubation of casein from unheated milk with pronase alone hydrolysed 70–80% of peptide bonds involving Ile, Leu, Tyr, Phe and His residues; other amino acids were released less well and proline hardly at all. The pronase/peptidase treatment resulted in 90–100% hydrolysis of peptide bonds involving all amino acids, including proline.

Caseins from stored UHT milks were more resistant to proteolysis than casein from unheated milk. Reduced release of all amino acids was observed from samples taken after storage at 37 °C for 12 months or longer and for Lys, Arg and Asn residues from samples taken after storage at 30 °C for 14 months. Resistance to proteolysis was attributed to the Maillard reaction between milk proteins and lactose during storage of UHT milk.

The coagulation of caseins is considered in the light of a polyfunctional condensation model for protein. This model was originally developed to explain aspects of the polymerization behaviour of small molecules, but it is shown here that it can be used to explain certain aspects of the polymerization of caseins, particularly the Ca2+-induced association of αs-casein and the heat-induced coagulation of casein in whole milk.

Lactuloselysine (ε-N-deoxylactulosyl-L-lysine) and fructoselysine (ε-N-deoxyfructosyl-L-lysine), formed by Maillard reaction, were identified in enzymic hydrolysates of casein from stored ultra-heat-treated (UHT) milk. Furosine (ε-N-(2-furoylmethyl-L-lysine) and pyridosine (ε-(3-hydroxy-4-oxo-6-methyl-1-pyridinyl)-L-norleucine) were identified in the corresponding acid hydrolysates. Enzymic hydrolysis of the caseins, gel filtration and preparative scale amino acid analysis were used to isolate the compounds which were then identified by reference to synthesized authentic compounds. Lactuloselysine formation was extensive and involved 10–30% of lysine residues in UHT milks stored at 30–37 °C for 6 months to 3 years. Fructoselysine concentration was generally about 10% of the lactuloselysine concentration.

Differential scanning calorimetry (DSC) was used to study thermal transitions of the following whey proteins and enzymes in milk ultrafiltrate solution: β-lactoglobulin, α-lactalbumin, serum albumin, γ-globulin, apo- and Fe-lactoferrin, lysozyme, ribonuclease, α-chymotrypsin and xanthine oxidase. Denaturation enthalpies (ΔHD), denaturation temperatures (TD) and the half width of the denaturation peaks in DSC thermograms (ΔT½D) were determined and the degree of renaturation was estimated by rescanning previously denatured samples. A fair correlation between the results obtained by DSC and other more classical methods was found in general. However, for some proteins (α-lactalbumin, lysozyme, ribonuclease and xanthine oxidase), which have so far been considered relatively thermostable, calorimetry reveals conformational changes starting at temperatures as low as about 45 °C. In these cases thermostability observed after heat treatment of milk should be interpreted in terms of renaturation and not of high temperatures of denaturation.

Three cationic materials markedly reduced the rennet clotting time of casein micelle suspensions, the efficacy of each being primarily dependent on the charge and the amount absorbed by the micelles. The reduction in coagulation time was unaffected by components of the milk serum other than salts. No enzymic action by lysozyme on casein micelles was detected. All materials acted by the same mechanism, increasing the affinity of rennet for the micelles and accelerating the aggregation phase. Coagulation did not occur until a minimum amount of κ-casein had been hydrolysed to para-κ-casein. All additives increased the proportion of added rennet retained by the casein in the coagulum. The results indicated that coagulation occurs by specific interactions between micelles modified by rennet.

The theory of branching processes is used to present a unified kinetic model for the heat induced coagulation of milk. Nitrogen depletion curves for the overall reaction are fitted to a theoretical curve to demonstrate that the reaction is a single process and that the idea of a lag phase up to the coagulation time, followed by the aggregation reaction proper, is an unnecessary complication. Coagulation times as a function of temperature are then used to determine the activation energy of the reaction; the average value of 126 kJ/mol agrees with that found from analysis of the post-coagulation reaction.

Four chemical methods were used in a comparison of the measurement of the extent of modification of lysine residues in casein during storage of ultra-heat-treated (UHT) milk. Determination of homoarginine after reaction with o-methylisourea appeared to be the best measure of total lysine modification. Reduction with borohydride before acid hydrolysis prevented recovery of lysine from lactuloselysine (ε-N-deoxylactulosyl-L-lysine) and fructoselysine (ε-N-deoxyfructosyl-L-lysine). This procedure gave values for these derivatives which were in good agreement with those obtained from the determination of furosine (ε-N-(2-furoylmethyl)-L-lysine) in acid hydrolysates. Determination of lysine after acid hydrolysis showed that lysine loss was accounted for by lactuloselysine and fructoselysine formation in most of the samples.

Browning was most apparent in UHT milk stored at 37 °C for 3 years and in this sample only part of the lysine residues was accounted for as lactuloselysine and fructoselysine, thus indicating that lysine had become involved further in sugar degradation products.

Water vapour adsorption and desorption properties of spray- and roller-dried milk baby foods were determined by a dynamic method at 25, 35 and 45°C. The equilibrium moisture content increased with increase in relative humidity and decrease in temperature; values for the spray-dried milk baby food were higher than those for the roller-dried product. The desorption and adsorption isotherms exhibited hysteresis. The desorption isotherms were of sigmoid shape: the adsorption isotherms were of similar shape except for a region of discontinuity. The relative humidity at which this discontinuity occurred depended on the temperature at which sorption was determined. Monomolecular moisture contents and latent heat ratios are reported.

Changes in the appearance of casein micelles and fat globules in ultra-heat-treated (UHT) milk were comparatively slight within a 14-month storage period although closer inspection often revealed the beginnings of some of the changes observed after even longer storage. Within this period the milks remained liquid and no evidence for any proteolytic breakdown was seen on electrophoretio examination. After 34 months the milks had all gelled and there was evidence of a small amount of proteolytic breakdown. Electron microscopic changes were striking and depended on storage temperature. At 4 °C the casein micelles in the gel phase had become very ‘spiky’ and long tendrils often bridged micelles, forming extensive networks. At room temperature and at 30 °C this tendril formation was less marked and the average micelle size was considerably greater than in freshly processed milk. There was also clear evidence for the coalescence of micelles. After storage at 37 °C the micelles were much larger than in the other samples, often exceeding 0·4 mμ in diam.

The specificity of chymosin on αs1-casein was shown to be dependent on the reaction pH and on the state of aggregation of the substrate. In aqueous solution αs1-casein was optimally hydrolysed to αs1-I at pH 5·8; if the casein was solubilized in the isoelectric region by the use of 5 M-urea, optimum proteolysis occurred at pH 2·8. Hydrolysis of αs1-I to yield αs1-II, αs1-III and αs1-IV occurred at pH values > 5·8 in the presence or absence of urea. In the isoelectric region αs1-II, αs1-III and αs1-IV were not formed in the absence of urea where the substrate was aggregated: instead a peptide αs1-V was produced; at the same pH and using urea as a solubilizing agent αs1-II, αs1-III and αs1-IV were formed together with a further peptide αs1-VI.

A method for lactose hydrolysis in sterile milk or whey with Saccharomyces lactis lactase is described. The properties and stability of this enzyme are very favourable for use in milk and neutral whey. The long shelf life of sterilized milk and whey products is utilized for a prolonged incubation time – one to several weeks – at room temperature. Thus, minimal amounts of enzyme are needed which makes enzyme costs very low.

Changes in the dependent variables flow velocity, inlet pressure and temperature affected the rate of concentration of separated milk by reverse osmosis; their values were measured for minimum fouling at the membrane, resulting in maximum rates of concentration. Most of the deposit formed at an early stage of concentration and its extent depended on the operational variables.

Electron micrographs showed that the major component of the deposit was casein micelles linked by bridges to form a lattice. Chemical analysis of the deposit confirmed a high casein content and showed that Ca phosphate was precipitated in the deposit throughout concentration. When fat globules and fat globule membrane were present in the feed, they appeared to be caught up in the deposit, but did not affect its initial formation.

Thermal conductivities of creams of about 20, 30 and 40 % fat were measured at several temperatures between 5 and 75 °C. Because of the meagre data on cream, special attention was given to the prediction of conductivity using different mathematical models. An equation was derived for the thermal conductivity of cream as a function of temperature and volume fraction of the fat phase.

Of 8 strains of Streptococcus cremoris tested, 5 grew almost as well in defined media in which various essential amino acids were supplied in dipeptides as they did in media containing the equivalent free amino acids. The remainder grew poorly or not at all in the peptide-containing media. Growth of peptide-utilizing strains was inhibited by also including structurally-related dipeptides in the medium, presumably due to competition for uptake by transport system carriers. Both types of starters produced cell-free dipeptidases recoverable from the medium of exponential phase cultures. Addition of the partly-purified extracellular dipeptidases to dipeptidecontaining test media initiated growth in strains unable to use peptides. Str. lactis grew in defined peptide media, but the further addition of structurally-related dipeptides did not inhibit growth, either bcause each dipeptide was transported by a specific carrier or because peptides were hydrolysed extracellularly. The presence of cell-bound extracellular dipeptidase was indicated by the hydrolysis of dipeptides with washed whole cells in buffer. This was not observed with Str. cremoris strains.

The nutritional protein quality of lactose-hydrolysed milk after some industrial processes was studied with N balance experiments on growing rats.

Ultra-high-temperature sterilization and evaporation at 25 °C did not influence the nutritional value significantly, whereas spray-drying under conditions usually used for ordinary milk gave a considerable reduction in protein quality. This was caused mainly or entirely by loss of biologically availabe lysine.

The stability of homogenized ultra-heat-treated (UHT) aseptically-packed cream of 18 % fat content when added to hot coffee was indirectly monitored in 10 batches of cream, stored for 10 weeks at different temperatures. A tendency to coagulate under standard conditions (instability) is known as feathering. Stability was determined by a procedure in which acetate buffer instead of coffee solution was used as the test solution. The distribution of components between the fat and aqueous phases of the stored cream was also monitored. Feathering score (stability) decreased during storage but was not accompanied by any increase in non-protein nitrogen levels. The amount of casein and Ca in the fat phase of the cream increased with time and was inversely related to feathering score. Some batches of cream contained 0·1 % Na2CO3 and 0·1 %Na citrate as stabilizers and these additions were found to improve feathering scores early in the storage period. Initially, the fat phase of these creams contained less casein and Ca than those without the additives, but after 10 weeks the differences in feathering score and in casein and Ca distribution, between the 2 types of cream diminished. Examination by electron microscopy showed that casein micelles were associated with fat globules, apparently linking several globules together, which was less in the creams containing additives, but after 10 weeks the appearance of all creams was similar. Electrophoretic separation of the fat globule membrane (FGM) proteins by SDS-polyacrylamide-gel electrophoresis showed that casein was the principal protein component, but that β-lactoglobulin and native FGM proteins were aslo present.

A small process analyser employing the pulsed nuclear magnetic resonance technique has been used to determine both water and fat contents of milk powder and cottage cheese samples. The procedures used for establishing and eliminating cross-interferences in these analyses are described. For milk powders, water and fat contents in the ranges 1–5 and 0·5–25 % respectively were determined with standard deviations of 0·20 % water and 0·64 % fat. Cottage cheese samples with water and fat contents in the ranges 77–81 and 2–7 % respectively were determined with standard deviations of 0·30 % water and 0·16 % fat.