A detailed calculation of the growth of molecular weight during the renneting of milk is given, based on a first-order breakdown of κ-casein followed by development of instability caused either by a decrease in the intermicellar repulsive potential or by the formation of holes in the stabilizing surface layer of the micelles. Unlike most of the models which have been described, this model allows a complete analytical solution. The solution is, however, complex and difficult to use simply, although it is shown that the calculations are in accord with experimental observations of the dependence of the coagulation process upon the enzyme concentration and the concentration of the milk. The calculations are also compared with those from other models of the reaction.

Significant effects of parity and stage of lactation were found on ATP content, somatic cell count and antitrypsin content when using morning milk samples from 39 cows over a period of 18 months. A close similarity was evident between variation in ATP and somatic cell count, except during the first 10 d after parturition when the variation in ATP was more pronounced. The same result was obtained in the whole material as well as in a healthy subset. This study shows clearly that parity and stage of lactation must be taken into account when establishing normal values for ATP as an indicator of mastitis.

A reference milk system (Clarke system) using primary reference milks made from preserved natural bovine milks (PNM) and standard reconstituted milk (SRM) has been developed to enable accurate calibration of milk analysing instruments. The PNM have values of 2, 3, 4, 5 and 6% fat and 3·5, 2·7, 3·4, 4·2 and 3·3% protein respectively and can be stored for 4 months without detectable change in fat or protein content. The above reference milk system utilizes SRM as a secondary reference milk to enable regular checking of the stability of the instrument calibration during routine testing (e.g. after every 100 samples). Over several months seven laboratories using 13 fat-testing instruments and seven protein-testing instruments achieved high levels of accuracy in weekly calibrations (coefficient of variation 1·1%) when they used the reference milk system and adhered to the prescribed calibration criteria.

At pH 7·0, the emulsifying capacity (Ec) of ion-exchange fractionated whey protein powders in corn oil/water systems increased with increase in protein concentration up to 0·l% (w/v), where it was maximal. At higher concentrations Ec decreased because of packing effects of the oil droplets. At low protein concentrations Ec could be determined from linear equations involving the relative concentration of the individual proteins in the powder. The surface coverage of β-lactoglobulin on the oil was ∼ 1 mg m–2. Oil droplet size and Ec were maximal at pH 4·0–6·0, corresponding to the isoionic points of the proteins.

Milk fat was fractionated into liquid (m.p. ⋍ 12 °C), intermediate (m.p. ⋍ 21 °C) and solid (m.p. ⋍ 39 °C) fractions by three different processes—melt crystallization, short-path distillation and supercritical CO2 extraction—and the cholesterol content of these fractions determined. Cholesterol was enriched in the liquid fractions from all three processes, in particular about 80% of the cholesterol being found in the liquid fraction obtained by short-path distillation. The basis of migration of cholesterol into various milk fat fractions was explained by its affinity to various triglycerides (melt crystallization) and by vapour pressure and molecular weight (short-path distillation). It was more complex in the supercritical CO2 extraction process; the interplay of cholesterol affinity toward CO2 and its molar volume, and its vapour pressure enhancement under applied pressure play a role.

Milk samples of four species of wild equids (onager, Equus hemionus onager; mountain zebra, E. zebra hartmannae; plains zebra, E. burchelli; Przewalski horse, E. caballus przewalskii) and two domesticated equids (ass, E. asinus; pony, E. caballus) were analysed. At mid to late lactation the milks of all species were very similar, containing on average 10–12% total solids, 1–2% fat, 1·6–1·8% true protein, 6–7% ‘lactose’, 0·3–0·5% ash, 0·08–0·12% calcium, 0·04–0·07% phosphorus and a calculated energy content of 2·0–2·4 kJ/g. Milk samples collected in the first 2 weeks after birth showed elevated levels of total solids and protein, and some had reduced ‘lactose’ levels, but there were no observable trends in milk composition during mid to late lactation (1–12 months post partum). It was concluded that these closely related species produce milks that are nearly identical in gross composition and that the domestic horse is a representative model for the study of equid lactation.

The rennet coagulation times of infant milk formulae or fresh skim milk (milk) samples heated at temperatures in the range 70–140 °C for 1–10 min decreased on acidification, usually to pH < 6·0. Heated milk samples acidified to pH 5·5 and reneutralized to pH 6·6 retained good rennet coagulability. Acidification of such milk samples before heating also reduced the adverse effect of severe heat treatment (95 °C for 1 min) on rennet coagulation. Addition of low concentrations of CaCl2 to heated milks offset the adverse effects of heating. Acidification of heated milks increased the [Ca2+], and reneutralization of acidified milk only partly restored the [Ca2+], i.e. acidified/reneutralized milk had a higher [Ca2+] than normal milk, suggesting this as the mechanism via which acidification/neutralization improves the rennet coagulability of heated milk. Approximately 50% of the whey protein can be incorporated into rennet gels in heated milks while retaining good coagulability and curd tension; this may be a useful technique for increasing cheese yield.

The dissociation of caseins of different types from casein micelles in milk, acidified to different pH values in the range 4·9–6·7, and at temperatures of 4, 20 and 30 °C, has been studied. In contrast to a number of previous findings, it was shown that caseins of all types were dissociated from the micelles, although in all cases β-casein was in highest concentration. The amounts and proportions of all of the caseins were found to be pH- and temperature-dependent, especially the former. Studies of the proportions of the different caseins liberated suggested that, at a defined temperature, the proportions of κ;- and αs2-caseins were independent of pH, while the proportions of β- and αsl-caseins were variable, changes in one being compensated by changes in the other. The manner in which the proportions of the αsl-casein and β-caseins changes with pH was found to be temperature-dependent.

Mixtures of κ-casein and β-lactoglobulin (β-lg) were heated at 70 °C in 20 mM-imidazole buffer, pH 6·8 containing 20 mM-EGTA. Aggregation of the κ-casein/β-lg mixture occurred within 90 s and susceptibility to hydrolysis by chymosin decreased significantly within 180 s even though covalent interaction was not detected until after 4000 s. UV-absorbance indicated initial structural destabilization of the heated κ-casein/β-lg mixture followed by a return ( > 350 s) to a spectrum comparable to the native state, indicating molecular rearrangement. Apparent hydrophobicity of κ-casein decreased 80% within 250 s compared to a 38% decrease for β-lg. Under similar conditions, the κ-casein/β-lg mixture (1:1) showed a faster (2·5 times) decrease in apparent hydrophobicity than κ-casein alone with concomitant exposure of acidic (hydrophilic) groups. The results suggested that the driving force for the rearrangement was mainly hydrophobic, i.e. entropic in origin. The tendency of heated and subsequently cooled κ-casein/β-lg to aggregate reached a maximum after heating at 70 °C for 720 s.

The effect of the covalent binding of reducing sugars such as galactose, glucose, fructose, lactose and maltose on the flow properties of casein solutions and on the voluminosity of casein molecules was investigated, voluminosity being calculated from viscosity measurements. Rheological parameters appeared to be efficient indices of structural changes occurring in proteins as a result of chemical modifications. Results showed an increase in voluminosity of casein after the binding of sugars, possibly explained by an increase in the net negative charge or by an increase in the steric hindrance of the molecule. At high concentrations (above 0·03 g/ml), viscosity of the solutions depended on the nature of the attached sugar and on the level of the modification. Thus galactose- and glucose-modified caseins were more viscous than the control. With disaccharides, the level of modification appeared to be more important than the amount of sugar bound.

Cultures of Pseudomonas spp. strains P10, P12 and P15 grown in whole milk which contained ∼ 1 × 108 viable bacteria ml−1 demonstrated near linear increases in the concentration of short-chain free fatty acids and trichloroacetic acid soluble free amino groups at 20 °C, following either ultra high temperature (UHT) treatment (140 °C for 5 s) or dual heat treatments (140 °C followed by either 57, 60 or 65 °C). The dual heat treatments reduced the rates of lipolysis and proteolysis compared to the UHT treatment by up to 25-fold. The dual heat treatment utilizing 60 °C for 5 min also effectively limited both lipase and proteinase activities in raw milk culture samples which had contained either 6 × 106, 5 × 107 or 1 × 108 viable bacteria ml−1. In this system enzyme activities were reduced by up to 10-fold following dual heat treatment compared to UHT treatment alone.

A heat stable extracellular proteinase from the psychrotroph Pseudomonas fluorescens AH-70 was purified to electrophoretic homogeneity by affinity chromatography on a gramicidin S–Sepharose-4B column. Bacitracin linked to Sepharose-4B was unable to retain any proteolytic activity, whereas the same antibiotic bound to AH-Sepharose-4B retained ~ 25% of the total activity. The purification procedure on the gramicidin S–Sepharose-4B column was easy to perform, fast and reproducible; it resulted in a 207-fold increase in the specific activity and a yield of 41% of the original activity. The purified enzyme was a monomer with a mol. wt of 33000. The enzyme hydrolysed whole casein and its fractions whereas no activity was observed against bovine serum albumin. The enzyme was a metalloproteinase. It was heat stable, having D-values at 121, 135 and 150 °C of 3·8, 1·9 and 0·6 min respectively.

A peptic hydrolysate of proteinaceous material of whey from Mozzarella cheese manufacture was incubated with α-chymotrypsin under conditions leading to the formation of plastein products. The influence of incubation time, concentration and molecular size of substrate was studied. The most favourable conditions for the plastein synthesis were 12 h incubation time and 35% (w/v) concentration of peptides having mol. wt in the range 451–1450.

Soluble proteins and Ca and P salts in the fouling deposit on a mineral microfiltration membrane obtained in static conditions were analysed and compared with those found using dynamic filter conditions in a laboratory pilot installation for milk processing. The serum milk proteins were retained by the porous micellar deposit under both static and dynamic conditions, resulting in progressive fouling of the dynamic micellar membrane, enabling milk to be processed with a mineral microfiltration membrane. Ca and P salts increased fouling probably by allowing better adsorption of casein micelles on the alumina and by acting as intermicellar bonds in the deposit. Fat globules modified porosity, permeability and resistance to matter transfer of the deposit. The dynamic conditions gave a deposit of the same type and structure as with the static conditions although made thinner by the velocity effect.

An industrial refractometer connected to a microcomputer was used to study the rennet coagulation of milk. The amplitude of the refractometer's signal changes during coagulation was about 104 times more important than the resolution of the computerized refractometer. The signal v. time derivatives were calculated and the inflection point time (ti) was correlated with pH, Ca concentration and temperature. The inflection point time was inversely proportional to the rennet and Ca concentrations and the logarithm of the proportionality factor (k) was linearly correlated with pH. The relation between ti and temperature was curvilinear. It is concluded that the refractometer tested could be used for research on the coagulation process.

The effect of several parameters on the decarboxylation of α-acetolactic acid (ALA) to diacetyl and acetoin was studied in model experiments. At temperatures > 50 °C, relatively more diacety1 than acetoin was produced. Regardless of temperature, HC1 increased the production of acetoin (58 mol% conversion at max.) and prevented the production of diacetyl. Steam distillation of ALA caused ~ 55 mol%/h conversion to acetoin and 7 mol%/h to diacetyl. Production of acetoin was proportional to the ALA concentration. Based on these results two methods for the quantitative measurement of ALA were developed. Only one of 17 strains of Streptococcus diacetylactis (Lactococcus lactis subsp. lactis biovar. diacetilactis, new nomenclature) and three of three strains of Leuconostoc spp. produced ALA. The implications of these findings are discussed.

Synergic effects of proteolytic enzymes from two different microbial sources on the ripening of Swedish hard cheese were studied. When extracellular proteolytic enzymes from Bacillus subtilis (Neutrase) and/or heat treated cells of Lactobacillus helveticus (now L. delbrueckii subsp. helveticus) were added to the cheese milk, cheese ripening was accelerated; Neutrase effectively hydrolysed casein to give a softer body. Addition of heat treated lactobacilli did not accelerate hydrolysis of casein, but accelerated the breakdown of peptides which increased the amount of amino acid N in the cheese and also enhanced the intensity of cheese flavour. A bitter taste which developed in cheeses with added Neutrase could be eliminated by the simultaneous addition of heat treated lactobacilli.

Samples of caseins having different Ca contents as used in cheese processing were analysed by techniques using differential scanning calorimetry and thermally stimulated currents (TSC) before and after treatment with Na poly-phosphate, a food additive used in the manufacture of processed cheese. These techniques revealed structural changes induced by the salt, and the different types of water molecules associated with the protein are evident. This characterization is in agreement with results obtained by other techniques, particularly X-ray diffraction of proteins. Transmission electron microscopy of the same samples confirmed that the changes observed by TSC were associated with an unravelling of the protein.

Cows were infused with Escherichia coli endotoxin to provide a pool of milks with high somatic cell counts (SCC). This was mixed with bulk tank milk obtained from either of two Institute herds, producing a range of milks containing 0, 10, 20, 30 and 100% high SCC (HCC) milk. Creams containing 38% fat were produced from these milks on nine separate occasions; five times from winter milks and four times from summer milks. The SCC of the milks, the free fatty acid concentration of the unwhipped creams and the whipping time, stiffness and overrun of the whipped cream were measured. Whipping time and stiffness increased and overrun (volume increase) decreased as the proportion of HCC milk in the mixture increased. However, these changes were statistically significant only when comparing the 100% HCC milk with the other treatments containing lower levels of HCC milk. Creams produced in the summer took less time to whip and were less stiff than winter creams but there was no change in overrun. The source of bulk milk supply affected the stiffness and overrun of the whipped creams. Free fatty acid concentration appeared to have no direct effect on whipping characteristics.

Deposit formation was measured in a model laboratory plant in which whole milk was in contact with a heated surface at 100 °C for 1 h. The effect of the interfacial properties of various poly mer-coated surfaces on the amount and the adhesion strength of deposit was determined. The nature of the surface influenced the formation of deposit only slightly, but had a large effect on its adhesion strength. From correlation analysis, the polar contribution to surface energy was identified as the main factor influencing the deposit adhesion strength. These results suggest that the type of interactions at the surface govern the ease of removal of deposit.