Constituents of a flavour concentrate from fishy butterfat were separated by gas chromatography on silicone oil and ‘Carbowax 400’ columns. Two fractions with distinct flavours were obtained, the one of oily flavour and containing n-hexanal, n-heptanal, hex-2-enal and heptan-2-one, the other of metallic flavour, containing a single carbonyl compound present in relatively small amount. It is considered that these compounds with the exception of heptan-2-one are mainly responsible for the characteristic fish-oil flavour.

Other compounds isolated include propanal, n-pentanal, n-octanal, n-nonanal, n-decanal, acetone, pentan-2-one, nonan-2-one, undecan-2-one, acraldehyde, pent-2-enal, hept-2-enal, oct-2-enal, non-2-enal, hepta-2, 4-dienal, a compound resembling octa-2, 4-dienal, a compound with a mushroom flavour, and a compound that formed the 2, 4-dinitrophenylosazone of methylglyoxal.

It has been shown that iodine present as iodide in milk can be quantitatively removed by passing the milk through a column of an anion-exchange resin in the chloride form. Any residual iodine is present as non-ionic species which are not removed by the resin.

This technique was used to determine the non-ionic iodine in milk at various times after the oral administration of a single dose of 131I to dairy cows and also during a multiple-dose experiment. Non-ionic131I was present in milk 6 h after its administration and the fraction in this form varied considerably in milk from different cows, ranging from 1·2 to 14·5% of the total 131I content. The chemical form of the nonionic iodine was not investigated.

The sterilization of homogenized milk at 115·5°C for 15 min in bottles having a partial vacuum in the headspace produced conditions inhibitory to the growth from very small numbers of spores of Bacillus subtilis, B. licheniformis, B. cereus and B. brevis when compared with growth in the same milk sterilized in open bottles. B. circulans differed from the other strains tested in showing greater inhibition in milk sterilized in open bottles than in milk sterilized under partial vacuum.

The extent of the inhibition became less as the size of the inoculum was increased. It became less also as the temperature of incubation approached the optimum growth temperature of the bacillus, and was influenced by the strain of the bacillus and the source of the milk but not by the degree of heat treatment within the range 107–117·5°C for 15 min. Inhibition was manifested by a prolongation of the lag phase, and in addition with some strains inhibition of spore germination could be demonstrated.

Spore formation following vegetative growth occurred more readily in milk sterilized in open than in evacuated bottles.

Milks sterilized under partial vacuum frequently failed to show any growth from small inocula in 30 days at 22°C although growth occurred readily in milk sterilized in open bottles.

A detailed study has been made of spectroturbimetric methods for testing the degree of homogenization of milk, and it has been shown that, provided care is exercised in the design of the turbimeter, the turbidity index at a specified wavelength provides a useful practical measure of the degree of homogenization. This measurement can easily be made with a simple filter turbimeter.

The effect of ultra-high-temperature (U.H.T.) sterilization and of subsequent in-bottle sterilization on the availability of milk calcium has been studied in experiments with rats. The heated milks and the raw milk from which they were prepared were freeze-dried and added to a diet low in Ca so that the resulting diets contained some 0·11% Ca and 0·42% P.

Ca retention by groups of six litter-mate male weanling rats was 98·4, 98·4 and 98·5% respectively from diets containing raw milk, U.H.T.-treated milk and milk treated by U.H.T. followed by in-bottle sterilization. It is clear that these processes did not affect the availability of milk Ca.

Removal of colloidal phosphate leads to changes in the propertics of milk of which the increased viscosity and greatly increased sensitivity to calcium salts, and the apparently diminished degree of mutual integration of the various casein fractions are the most striking. Re-introduction of colloidal phosphate restores incompletely and only in certain instances the original properties of the milk.

These observations are interpreted as favouring the conception of chemical links between colloidal phosphate and casein in milk.

The effect of the temperature of laboratory sterilization of homogenized milk on the titratable acidity, the colour of the milk, the yellowing of the filtrate from the Aschaffenburg turbidity test, the acid-ferricyanide-reducing substances and on pH has been determined. The greater changes in these properties in milk sterilized under reduced oxygen tension compared with milk sterilized in open bottles was shown to be due, in part at least, to the slower rate of cooling in the evacuated bottles. Samples of commercial sterilized milk have also been examined and their properties have been compared with those of the sterilized milk produced in the laboratory.

Two experiments are described in which spring-calved Ayrshire cows were fed supplementary food when grazing good, leafy pasture.

In the first experiment 6 lb of a mixture of three parts bruised oats and one part flaked maize was fed daily to three groups of five cows. One group received the supplement from 1 July until 30 October, and the other two groups from 1 August and 1 September respectively. A control group received no supplement.

In no month was there any statistical difference in daily milk yield between supplemented and unsupplemented groups of cows. The group fed from 1 July gave the highest total milk yield in the period June to October, but with a return of only 1 gal of milk from 34 lb of meal.

The length of the lactation was not significantly affected by the treatment.

In the second experiment, three groups of four cows on a 9-week change-over trial (4 May-5 July) grazed herbage containing an average of only 17·6% dry matter and 3·8% crude fibre. On the three treatments the cows ate pasture only, pasture plus 1·6 lb hay/day and pasture plus 5·7 lb of meal/day. The meal consisted of four parts by weight of undecorticated cotton cake, four of bruised oats and three of locust beans.

Milk yield was significantly increased from 41·8 lb on the all-grass ration to 43·2 lb/day on the meal treatment but the hay did not affect milk yield.

The fat content of the milk was significantly increased from 3·77 to 3·91% on the hay treatment, whereas the solids-not-fat content of the milk was not significantly affected by any of the treatments.

At the present-day prices of the feeds and with cows on good leafy pasture it was clearly uneconomic to feed the meal supplements.

Batches of cream inoculated with 1% of a culture of Streptococcus diacetilactis or of one or other of two mixed starters were held at different temperatures over the range 37–57°F before churning. Diacetyl and acetoin were estimated in the creams at churning and in the butters fresh and after storage.

The diacetyl production of all three starters in the cream increased with increase in temperature, although at the lower temperatures there was no indication of any rise in titratable acidity. At 52 and 57°F there was some rise in acidity during the holding period.

The butters were of good keeping quality. There was a small but definite reduction in the diacetyl content of the butters during storage, but the extent of the change was not related to the holding temperature of the creams.

It was concluded that if sweet cream starter butter is to be made with uniform intensity of flavour the proportion of starter added must be adjusted to counterbalance changes in the cream-holding temperature.

For milk sterilized under partial vacuum in closed bottles the oxidation-reduction potential measured under nitrogen was markedly lower (—280 to —300 mV) than that of milk heated in bottles open to the air, measurements being made in air (+10 to —30 mV).

The Eh (under nitrogen) of commercial sterilized milk and of ultra-high-temperature short-time heated milk was in the range —200 to —300 mV and of the same order as that obtained in the laboratory for deaerated milk sterilized in closed bottles.

A positive 80% alcohol test was not obtained in sterilized milks containing fewer than 105 bacteria/ml with Bacillus subtilis, B. licheniformis and B. cereus, and usually did not occur until the stationary phase of growth had been maintained for some days. A positive alcohol test with less than 105 bacteria/ml was associated with the lower maximum population density of B. brevis and B. circulans 152.

The alcohol test after 24 h at 37°C was found unreliable as an indicator of the keeping quality of sterilized milk at 22°C. The possibility of using the alcohol test after 3 days at 37°C or a bacterial count after the same incubation period is discussed briefly.

Production of diacetyl and acetoin by starters in cold skim-milk and cream was shown to increase with increase in the proportion of starter culture added, with some limitations at the higher rates of starter addition.

With Streptococcus diacetilactis starter in skim-milk at 50°F the relation between proportion of starter added and production of diacetyl was linear up to addition at the 4% level, whereas at 43°F it was approximately linear up to the 10% level. At both 50 and 43°F the relation between the proportion of starter added and the production of acetoin was linear up to the 10% level.

With Camb starter in skim-milk at both 50 and 43°F there were regular increases in production of diacetyl up to the 4% level of addition, but only minor changes thereafter with increase in the proportion of starter added up to 10%. At both temperatures the maximum production of acetoin was reached with the 7% rate of addition.

Production of diacetyl and acetoin in skim-milk was greater at 50°F than at 43°F with both starters for all proportions up to 10%, and it was greater for Str. diacetilactis than for the mixed cultures.

Except at the higher rates of addition of starter and at the higher temperature there were no concomitant increases in the acidity of the milk or lowering of the pH values. It appears that at low temperatures production of diacetyl by starters in sweet milk and cream proceeds independently of production of lactic acid.

Similar results were obtained in a series of experimental buttermaking trials and some small commercial-scale trials, in which varying proportions of starter were added to creams after pasteurizing and before holding overnight for churning. With the cream-holding temperatures used, mainly 40–50°F, the pH values of the butters were not appreciably lowered by the starter additions to the cream. At all the rates of addition there were with Str. diacetilactis starter higher contents of diacetyl in the butter than with Camb starter. There was no indication of any relationship between the proportion of starter added and the keeping quality of the butter.

Three groups of eight Friesian cows, four of high and four of low milk yield, were given during the last month of the winter-feeding period rations that provided respectively low, normal and high planes of energy nutrition, at approximately 75,100 and 130% of the Woodman standards for starch equivalent (S.E.). They were then all changed to the grazing of good, spring pasture, without supplementary food. The changes with grazing in solids-not-fat (S.N.F.) content and yield of milk were markedly influenced by the previous level of winter feeding. After 15–18 days of grazing the mean changes in S.N.F. percentage were +0·33, +0·11 and –0·12 for the low, normal and high energy groups, and the corresponding changes in milk yield were +7·70, +2·13 and –0·68 lb per cow per day. The changes in S.N.F. reflected mainly changes in protein content, and to these casein, β-lacto-globulin and α-lactalbumin all contributed. Small, differential changes were found in the fat and lactose contents of the milk. The effects of the level of initial milk yield of the cow on the various responses to grazing are recorded.

These observations support the view that when a rise in S.N.F. with spring grazing occurs it probably results from an improvement in the plane of energy nutrition. There appears to be no need to postulate a special effect of any specific herbage constituent.

Results are reported of an experiment with fattening pigs, made under commercial conditions, on the effect of adding formalin to liquid skim-milk, to delay souring, on the nutritive value of the milk given either in restricted or in unrestricted quantities.

There were four treatments: (i) Meal, fed to a scale based on live weight, plus sour skim-milk restricted to a maximum of 5¼ pints per pig per day. (ii) As (i) except that milk kept sweet by the addition of 0·15% formalin shortly after separation, was fed. (iii) Daily allowance of 2 lb meal/pig fed throughout the fattening period plus an unrestricted supply of sour skim-milk. (iv) As (iii) except that milk kept sweet by the addition of 0·15% formalin shortly after separation, was fed.

There were five pens of nine group-fed pigs on each treatment, involving a total of 180 pigs. The pigs were on experiment from about 9 weeks of age to bacon weight. Comprehensive carcass measurements were made on all the pigs, and bacon-tasting tests were made on samples of both green and smoked bacon from a number of the carcasses.

There were no significant differences between any of the four treatments in rate of growth or overall efficiency of food utilization.

Pigs given formalin-treated milk had a significantly thinner layer of back fat over the loin than those given sour milk. Although the difference was relatively small, it was apparent whether the milk was fed in restricted amounts or ad lib. and was reflected in the commercial grading results. There were no significant differences between treatments in dressing percentage, carcass length, shoulder back fat thickness, belly thickness or size of eye muscle.

No taint or off-flavour was detected in samples of green or smoked bacon from pigs fed milk to which 0·15% formalin had been added.

The feeding of liquid skim-milk in unrestricted quantities is discussed in relation to the comparative prices of meal and skim-milk.

The importance of adding the formalin to the skim-milk as soon as possible after separation, particularly during warm weather, is stressed in order that souring might be delayed for at least 1 week.

The lactobacillus flora of the mixed morning and evening milk of a single herd and of sixty Cheddar cheese made from this milk, unheated, was examined over a period of 12 months. Few lactobacilli (usually not more than 1/ml) were found in the milk samples, and were identified as Lactobacillus plantarum, L. casei, L. brevis, L. fermenti and unclassified strains. In addition, Pediococcus cerevisiae and leuconostoc species were isolated on the selective medium used, the leuconostocs occurring more frequently than either the lactobacilli or the pediococci. The lactic-acid bacteria in the cheese varied considerably, both quantitatively and qualitatively. Although in general the same species of organisms were isolated from the cheese as from the milks, the relative frequencies of occurrence were not the same, leuconostocs being found much more often in milk than in cheese, and L. plantarum and L. casei more frequently in cheese. No association between numbers or types of lactic-acid bacteria and flavour of the cheese was noted, nor did the season of the year or the ration fed to the milking herd affect the lactobacillus flora of the milk.

The numbers of lactobacilli, including pediococci and some leuconostocs, in raw and heat treated milks at twenty-six cheese factories in different parts of the country were determined over a period of 12 months, by plating on acetate agar. Counts in the raw milks varied from < 1 to > 100000/ml, but were usually within the range 100 to 10000/ml. Widely different times and temperatures of heat treatment were in use at the different creameries, some of which destroyed most of the lactobacilli present, whilst others had less effect. Post heat treatment re-infection of the milk in the vat with lactobacilli usually occurred.

Data available for fifty cows mainly infected with Staphylococcus pyogenes have been analysed to determine the effects of udder infections and mastitis on milk, fat and solids-not-fat yield and on fat and solids-not-fat percentages over full lactations.

It was found that lactation yields of milk, solids-not-fat and fat were depressed by 10, 11 and 12% respectively in the presence of infection. These depressions were significant (P < 0·05). The composition of the milk averaged over whole lactations was not significantly affected (P > 0·05), although the mean solids-not-fat percentage of the milk was significantly depressed by 0·196±0·092 during the period of infection.

It is suggested that the depressions in lactation yield and changes in milk composition would have been greater in the absence of antibiotic therapy for mastitis.