The magnitude of the forces causing closure of the bovine teat canal was estimated by measuring the pressure required to cause retrograde flow of milk into the canal from a manometer attached to the external teat orifice. Before milking, the hydrostatic pressure required to cause retrograde flow was 12 kPa or greater. Retrograde flow occurred at a pressure of 4–6 kPa once the teat canal surface had been wetted. This 2- or 3-fold fall in the closing forces presumably resulted from the combined effects of fracturing the teat canal epithelium and loss of initial resting tension in the smooth muscle fibres surrounding the teat canal. After milking, retrograde flow into any teat canal occurred at a pressure ∼ 1 kPa lower than the wetted, pre-milking value. This small drop, and subsequent recovery ∼ 30 min after milking, reflects the change in tonicity of the smooth muscle fibres which are stretched during milking. At hydrostatic pressures below 7–9 kPa, retrograde flow could be arrested momentarily by the occurrence of a teat contraction. At 4–7 kPa, the wave of contraction was also capable of expelling 4–14 μl milk from the teat canal indicating the outwardly directed, peristaltic nature of the phasic muscular activity.

The magnitude of the cyclic load applied to the teat apex by the collapsed liner during milking was estimated by measuring the pressure required to cause retrograde flow of milk into the teat canal from a manometer attached to the external teat orifice. The study was extended by observation of the milk flow cycle within a transparent teatcup assembly and by pressure recordings within the teat canal using a catheter-tip transducer. The source of the force compressing the teat is the small airspace, within the collapsed liner, directly below the teat apex. The total force generated is determined mainly by the size of this airspace and the pressure difference (PD) acting across the opposing liner walls. When the PD reached 80–90% of the liner vacuum, the load was just sufficient to occlude the teat canal. Further increase in PD provided the compressive load capable of offsetting vascular congestion of the teat apex. Increasing liner tension increased the load applied. The narrow range of compressive loads applied by conventional liners (5–12 kPa above atmospheric pressure) may be sufficient to offset congestion and oedema in the teat. Loads greater than the mean arterial pressure within the teat apex might serve only to expose the tissues to non-productive compression.

During the course of an acute, fatal mastitis in mice caused by intra-mammary injection of ∼ 105 cfu of Staphylococcus aureus, small numbers of staphylococci were detected in liver, spleen, kidney and heart. Transport from the mammary gland was by the venous blood system with no evidence of lymph node involvement. Intravenous injection of 100 times more staphylococci (∼107 cfu) than were fatal by the intramammary route failed to kill the mice and small numbers of staphylococci were found in the visceral organs. However, intravenous injection of ∼ 108 cfu of Staph. aureus was fatal and large numbers of staphylococci were found in the visceral organs. Treatment with cloxacillin (500 μg) by the intramammary or subcutaneous routes, or equally divided between these two routes, cleared the visceral organs of staphylococci; subcutaneous therapy had a limited and variable effect on numbers of staphylococci in the mammary gland while intramammary and divided treatments significantly reduced the numbers of staphylococci. Thus, staphylococci are disseminated to visceral organs during acute staphylococcal mastitis, but this makes a negligible contribution to the clinical disease.

Addition of reducing agents such as 2-mercaptoethanol (2-ME), dithio-threitol and Na sulphite to milk markedly reduced its heat stability at pH values below 7·1. 2-ME reversibly destabilized milk or serum protein-free casein micelle dispersions and promoted the release of κ-casein-rich protein from the micelles. Reduction of either casein micelles or β-lactoglobulin (β-lg) with 2-ME and subsequent blocking of the newly formed –SH groups with N-ethylmaleimide irreversibly reduced the maximum to minimum ratio in the heat stability profile. 2-ME disrupted κ-casein/β-lg complexes and stripped κ-casein from the micelles on heating. The milk or caseinate systems were thus destabilized. Addition of KBrO4 or iodosobenzoate to milk at 5 HIM eliminated the minimum but destabilized milk in the region of the maximum. However, KIO3 at 5 mm had a strong stabilizing effect throughout the pH range 6·5–7·3.

Goats’ milk caseins were separated into four components in a single run using reversed-phase gradient high-performance liquid chromatography. The purity of the isolated components was checked by sodium dodecyl sulphate polyacrylamide gel electrophoresis, amino acid analysis and determination of the N-terminal residue. By a comparison with previously published results for goats’ milk caseins the four peaks were identified as κ-, αs1-, αs2- and β-casein. In order to confirm the existence of αsl-casein in goats’ milk, this component was sequenced for 44 steps, revealing a sequence homologous to bovine αsl-casein and almost identical to the N-terminal sequence previously published by Boulanger et al. (1984).

Alkylation of whole casein samples by reaction with cysteamine and cystamine in a bis-tris-propane–urea buffer (pH 7·0) followed by fast protein liquid chromatography (FPLC) at 20°C on a Mono Q HR5/5 column in the same buffer and using a NaCl gradient led to good resolution of the whole casein into fractions representing (i) γ2- plus γ3-caseins, (ii) κ-caseins, (iii) β-casein, (iv) αs2-caseins and (v) αsl-caseins, together with small amounts of unidentified materials. Quantitatively the FPLC values agreed well with those for αs1-, β-, αs2- and γ2- plus γ3-caseins obtained by ion-exchange chromatography on DEAE cellulose, Whatman DE52 and with those for º-caseins obtained by gel-permeation chromatography on Sephadex G–150.

Heating milk at 140 °C caused an initial increase in the percentage of total casein sedimented at 10000 g for 1 h and in the relative viscosity, but as heating continued both parameters decreased before increasing again just before the onset of visible coagulation. This suggests that heating milk at high temperatures caused an initial aggregation of casein micelles with a concomitant increase in particle size and weight, followed by dissociation of these aggregates with reaggregation just before coagulation. Chromatography of heated milk on controlled pore glass confirmed the above suggestion. Calcium appeared to play a major role in the initial association of casein micelles which was also influenced by the initial pH and total solids concentration but whey protein had no effect.

The ethanol (EtOH) stability of skim milk and the stability towards aggregation of casein micelles diluted into ethanolic buffer solutions were compared using data obtained from previously published experiments. Differences in absolute stability and in relative response were observed when Ca2+ level and pH were adjusted, the buffer system results lying below those from skim milk in both cases. Increasing the ionic strength of skim milk adjusted to pH 7·0 lowered its EtOH stability whereas increasing the ionic strength of the diluting buffer increased the stability of the casein micelles. The hypothesis is put forward that the differences are due to the simultaneous precipitation of Ca phosphate when EtOH is added to skim milk. This draws calcium from the caseinate sites of the micelle, counteracting the destabilizing effects of the EtOH towards the micelle. Such removal and the consequent restructuring are kinetically controlled and micellar precipitation in skim milk finally occurs when the micellar coagulation time falls within the time scale of the restructuring reactions.

Lactobacillus bulgaricus strains LT1 and LT4 previously isolated from a yogurt factory were shown to be lysogenic and inducible by mitomycin C or u.v. irradiation, induction being optimal at the beginning of exponential growth. Five indicator strains (four of L. lactis, one of L. bulgaricus) were found in which the two phage lysates propagated well in a liquid medium, but formed plaques on bacterial lawns less readily. Ca2+ was required for the phage infection of the indicator strains to reach termination. Phage particles from each strain exhibited a comparable ultrastructural morphology under the electron microscope, having an isometric head, a triple collar and a noncontractile fibre-containing tail.

Milk clotting activity of Mucor miehei proteinase (commercial name Fromase) was determined in combined rennets as the difference between the total activity of combined rennet and the residual activity after inhibition of Fromase by the purified immunoglobulin fraction of a Fromase antiserum. The proposed method is simple, rapid, and makes possible the use of any arbitrary method for assaying milk clotting activity. The application of the immunoglobulin fraction, instead of the whole antiserum, is necessary owing to the presence of nonspecific proteinase inhibitors in blood serum.

A nutrient agar medium containing 0·1 % of a low melting point fraction of butterfat was shown to be suitable for detection, enumeration and isolation of lipolytic bacteria from milk. Bacterial growth was not inhibited by the butterfat and lipolytic reactions were clearly visible and easily interpreted. Lipolytic counts on the butterfat agar compared favourably with lipolytic counts obtained with other commonly used media.

Creams containing 25–45% fat were made from whole milks which had been concentrated 2- to 5-fold by ultrafiltration. Each series of creams (from any one concentrate) was standardized with its own skimmed milk, giving creams with higher than normal protein contents. These were then pasteurized and cooled for at least 24 h before being whipped. Increasing the fat content of creams gave increased stiffness, an overrun (volume increase) that passed through a maximum, decreased serum leakage and shorter whipping time. Increasing the protein content gave increased stiffness, less serum leakage, reduced overrun and a whipping time that passed through a maximum. Low fat creams, and also creams containing high amounts of fat and protein, did not produce a satisfactory whipped product. However, 35% fat creams from 2-fold concentrated milks compared well with ordinary 38% whipping creams. By lowering the fat content of creams an increase in yield was obtained.

Oggtt is a dried fermented milk product made and marketed primitively under uncontrolled conditions in the Arabian peninsula. By applying controlled conditions for production, the keeping quality of oggtt was greatly improved. The two types (cooked and salted) were produced from fresh pasteurized goats' or cows' milk or from skim milk powder. Oggtts flavoured with chocolate, date, mint or different fruits were also produced. Chemical composition and organoleptic properties of the final products were evaluated. Taste panel results indicated that the plain products, including those made from reconstituted milk, were acceptable, but flavoured oggtts, with the exception of date flavour, were not popular.

When Havarti cheeses were made from milk concentrated 5-fold using ultrafiltration (UF) and with a direct vat set type starter containing Streptococcus lactis, Str. cremoris, Leuconostoc cremoris and Str. lactis subsp. diacetilactis (i.e. a BD-type starter) their colour, texture, body, and flavour were equal to those of traditionally produced Havarti. The somewhat slower flavour development was associated with a retarded breakdown of protein. Although αs1-casein was almost completely broken down in Havarti cheese made with the ultrafiltration method (UF Havarti) and traditional cheeses after 120 d of storage, degradation of β-casein was retarded in UF Havarti. Rennet contents and numbers of starter organisms were similar in the two types of cheese and thus could not explain the difference in ripening rate. Str. lactis subsp. diacetilactis was dominant in both UF and traditional Havarti cheeses when the BD-type starter was used. By varying the composition of the starter, large variations in the bacterial composition and citrate fermentation in UF Havarti cheeses were obtained. The BD-type starter gave the best organoleptic quality.