1. In a winter feeding trial, using eight Ayrshire cows, two concentrates with digestible crude protein contents of 15% and starch equivalents of 63 and 79 were each fed at 2·6 and 4·6 lb./gal. of milk.

2. The basal ration was hay, silage, dried grass and fodder-beet.

3. Milk yield was significantly increased from 30·9 to 34·7 lb./day when the amount of concentrate fed increased from 2·6 to 4·6 lb./gal., but the average milk yields when the concentrates low in starch equivalent were fed did not differ significantly from those obtained when the concentrates high in starch equivalent were fed at the same levels.

4. The fat content of the milk was similar on all four rations with an average value of 3·84%, whereas the solids-not-fat varied from 8·53% on the low starch concentrate fed at the lower rate to 8·77% on the high starch concentrate fed at the higher rate.

5. The cows which were fed at the rate of 4·6 lb./gal. were significantly heavier by an average of 33 lb./cow than the ones fed at the rate of 2·6 lb./gal. The two different qualities of concentrate did not cause differences in live weight.

6. The weight of dry matter in the basal ration consumed per day was similar on all four treatments.

7. A comparison of feeding costs and the value of the milk showed that the largest margin was obtained with the low starch equivalent concentrate fed at the rate of 2·6 lb./gal.

8. It was concluded that because of the loss in live weight on the lowest level of feeding, and because the quality of the concentrate gave no improvement in milk yield or liveweight and little improvement in milk composition, the optimal level of feeding in this experiment was 4·6 lb./gal. of the low starch equivalent concentrate.

This series of four papers has outlined a method for predicting the bactericidal efficiency of an ultra-high-temperature sterilizing plant. The method has been illustrated with reference to the performance of an A.P.V. 200 gal./hr. plate-type sterilizing plant, treating at different temperatures milk heavily inoculated with suspensions of B. subtilis and B. stearothermophilus spores.

The method involves the determination of the temperature and flow time distributions in the plant, and of the thermal death characteristics of the organism to be considered. The investigations of plant performance are considered in Part I. The determination of the sporicidal effect of the plant from this information is described in Part II. The effect is expressed in terms of the slope of the thermal death line for the organism considered, at the operation temperature of the plant. By expressing the result in this form it becomes general and applicable to any type of organism, at any operating temperature close to that for which the plant is designed.

The thermal death time of any organism at the operating temperature may be obtained by laboratory experiment, by extrapolation if necessary. In Parts III and IV, laboratory data for B. subtilis spores and B. Stearothermophilus spores are used to calculate the performance of the plant, and the calculated results are compared with the results of direct plant experiment. The agreement is satisfactory.

The interpretation of the results with spores of B. stearothermophilus is complicated by their very marked inhibition by u.h.t.-treated milk. Only about one spore per million germinates and grows in such milk. A false impression of the number of truly surviving organisms after the heat treatment may therefore be obtained. It is not known to what extent this effect occurs with spores of other strains of B. stearothermophilus or with other organisms.

It is not suggested that the theoretical method of estimation of performance will give reliable information on the spoilage level to be expected from a plant under practical dairy conditions. The uncertainties as to the number and types of incident organisms are too great. Comparisons can be made between different plants, however, by comparing the results of the theoretical analyses for the plants. No bacteriological data are then required. This may be a less difficult procedure for manufacturers than a direct bacteriological experiment using heavily inoculated milk. The analysis also enables the contribution to the overall lethal effect of the different part of a sterilizing plant to be assessed.

Improved growth of Streptococcus cremoris 924 was obtained in the presence of CO2.

When fast and slow acid-producing variants of Str. cremoris 924 were mixed and the mixture subcultured daily the resulting cultures were all active acid producers, irrespective of the initial ratio of fast to slow cells. A balance is established between the two types of cell; probably this will vary with the cultural conditions and vary between strains of Str. cremoris and Str. lactis for any particular conditions of maintenance.

Providing the strains are correctly selected, single-strain starter cultures can be maintained by daily subculture in litmus milk, apparently indefinitely. Activity varies with the present method within wide limits and this variation seems to be a function of the culture.

Freeze drying is a satisfactory method of maintaining active single-strain starters.

The author wishes to thank Miss Shelagh Gibbs for technical assistance, Miss H. R. Chapman for making the cheese, and Messrs. Cow and Gate, Ltd., for supplying the dried milk powder for these experiments.

A cover for a 40 gal. cheese vat has been specially designed to allow cheese to be made under conditions which prevent the entry of micro-organisms into the vat from the dairy environment. The technique employed is described. Using milk free from lactobacilli it has been possible to make Cheddar cheese in which no lactobacilli grew during the ripening period. An assessment of the role of these and other micro-organisms in cheese ripening should now be possible.

The technique should also be of help in other cheese investigations where control of the microflora is desired.

The use of an autogenous vaccine/toxoid in the prophylaxis of clinical and subclinical staphylococcal udder infections in a dairy herd is described. Mainly first calving cows were included in the trial, and the results are given on 25 vaccinated and 18 uninoculated control animals. Four injections of vaccine at weekly intervals were given and every 6 months a further inoculation was given with a view to reinforcing immunity.

Observations were made over a period of 2½ years and there was a significant difference in the number of heavy staphylococcal infections and clinical cases in treated and control groups during the first 18 months, only one clinical case occurring in the treated group compared with 9 in the controls. Major infections totalled 8·1% during this period compared with 1·9% in the treated cattle, total staphylococcal infections being 16·2% compared with 7% in the respective groups.

In the last 12 months of the experiment eight clinical cases occurred in the treated group compared with five in the controls. It is clear from these results compared with those for the first 18 months that the reinforcing doses of vaccine/toxoid every 6 months were insufficient to maintain immunity. Subclinical infections, however, remained at a lower level in the treated group than in the control animals.

The significance of subclinical staphylococcal infections in relation to Whiteside tests is discussed, also the possible relationship of toxoid and bacterial vaccination to actual immunity produced. Some evidence is given that the presence of blood serum antitoxin is not necessarily associated with immunity.

The melting dilation of butterfat between 10° and 30° C. has been determined from results obtained for the thermal expansion of solid and liquid butterfat. Measurement of the solid fat percentages obtained with different cooling treatments showed that at higher rates of cooling there was an increase in the solid fat content, which is in accordance with the theory of mixed crystal formation. A cooling rate existed beyond which no further increase in solid fat content took place, since butterfat cooled by immersion in a 5° C. water-bath had practically the same solids content as it had when cooled in a 0° C. water-bath. Differences in cooling rate changed the solid fat content mainly in the region of lower melting-point glycerides. Seasonal variations in solid fat content indicated relatively large differences in the content of higher melting glycerides. There is an indication that not only the quantity of solid fat, but also the composition of the crystals is important as a factor influencing the hardness of butter. It was possible to recrystallize mixed crystals in butterfat at 22·5° C. and thereby lower the solid fat content of the butterfat.

1. In commercial practice dairy liquids of fat contents ranging from 3 to 90% are subjected to deodorization by steam distillation treatment.

2. Increase in fat test of cream causes a rise in the vapour/liquid equilibrium coefficient for tainting substances of low butterfat/skim-milk distribution ratio, and a fall for substances of high-distribution ratio, in cream.

3. Increase in fat content of cream reduces the amount of tainting substance in the cream per unit weight of butterfat in the cream, for substances of all butterfat/skim-milk distribution ratios.

4. Separation of milk to cream of a high fat content greatly facilitates the elimination of substances of low butterfat/skim-milk distribution ratio from the cream and butter.

5. Variation in temperature of separation, by its effect on the distribution ratio for tainting substances, could affect the content of these substances in the final cream, and could possibly cause a change in the character of the cream.

6. In the partition of tainting substance between butter and buttermilk in the churning process, fat content of cream affects the proportion of residual taint passing from the treated cream into the butter.

7. It is pointed out that, because of the effect of salt in increasing the vapour/liquid equilibrium coefficient of some steam volatile substances, the threshold concentration for a tainting substance in butter can be expected to vary with variation in the rate of salting of the butter.

1. Results are reported of an experiment with fattening pigs, made under commercial conditions, to obtain further information on the comparative value of dried skim-milk and white fish meal as protein supplements in a basal meal fed in conjunction with unrestricted amounts of whey.

2. All the pigs received, together with the unrestricted supply of whey, 3 lb. of a basal meal, reduced to 2 lb./pig/day by 13 weeks of age, fed dry once daily. Four different basal meals, comprising the four experimental treatments, were used, containing respectively 10, 7 and 5% white fish meal and 10% dried skim-milk as the protein supplement.

3. There were five pens of nine group-fed pigs on each treatment, involving a total of 180 pigs. The pigs were on experiment from approximately 8 weeks of age to bacon weight. Comprehensive carcass measurements were made on all the pigs.

4. None of the treatment differences for any of the variables studied was significant at the 5% level. There was, however, a trend for rate of growth to be reduced, efficiency of food conversion to decline and dressing percentage to improve with reduction of protein level in the diet. There was also an indication that the rate of growth and efficiency of food utilization of the pigs given the basal meal containing 10% white fish meal were rather better than for the pigs receiving the basal meal with 10% dried skim-milk as the protein supplement, although the mean differences were relatively small.

5. The results are discussed in relation to those obtained in earlier experiments completed under similar conditions at the Cow and Gate farm.

6. On the basis of the experimental evidence obtained so far, it was concluded that either 10% white fish meal or 15% dried skim-milk should be considered the minimum amounts of these protein supplements that should in practice be included in the basal meal fed with unrestricted whey in the manner described.

7. Possible effects of season on whey consumption are discussed in relation to some preliminary findings on the effect of whey acidity on the performance of fattening pigs.

8. Results of another preliminary experiment are mentioned in which a reduction in the daily allowance of the basal meal to 1 lb. or less had a very adverse effect on the performance of the pigs. It was concluded that a daily allowance of 2 lb./pig of meal given in conjunction with an unrestricted supply of whey, should be considered the absolute minimum in order to obtain a reasonable rate of growth.

We thank the Director of Messrs C. & G. Prideaux, Ltd., Bacon Curers, Motcombe, Shaftesbury, Dorset, for permission to take carcass measurements of the experimental pigs, and Messrs K. Lawrence and P. Prideaux who most conscientiously took these measurements. The help of Mr J. Rowell of the Agricultural Research Council Statistics Group, Cambridge, in making the statistical analyses is also very gratefully acknowledged. We should also like to thank Mrs J. Sedgwick for making the whey acidity determinations.

1. Measurements, by means of a continuous vaporization equilibrium still, of vapour/liquid equilibrium relationships in the steam distillation of dilute aqueous solutions showed that for mesityl oxide the relationship was linear, equilibrium coefficient (CV/CL)= 74.

2. For acetylpropionyl it was linear over the concentration range l–10 p.p.m. in the liquid, CV/CL=51, and above a concentration of 20 p.p.m. it was again approximately linear. At 20 and 50 p.p.m. in the liquid the values found for CV/CL were 34 and 21·5 respectively.

3. For acetyl-iso-butyryl it was not linear even at very low concentrations, of the order of 1–3 p.p.m. in the liquid, but there was an approach to linearity at concentrations 0–1 p.p.m., CV/CL=240. At concentrations above 1 p.p.m. in the liquid there was a steady fall in the value for the equilibrium coefficient.

4. For acetylvaleryl it was curvilinear even over the concentration range 0·1–4·0 p.p.m. in the liquid. At 0·5–1·0 p.p.m. in the liquid the equilibrium coefficient was of the order of 240. Above a concentration of 4 p.p.m. in the liquid the relationship was linear.

5. For acetylcaproyl it was approximately linear only over the concentration 0·1–0·4 p.p.m. in the liquid, CV/CL=180. At 4 and 8 p.p.m. in the liquid the coefficients were 90 and 70 respectively.

6. For acetylbenzoyl it was curvilinear at low concentrations, i.e. 0·1–0·5 p.p.m. in the liquid, CV/CL=30 decreasing to 16. Above this concentration there was a gradual fall in the equilibrium coefficient to 11·5 at 70 p.p.m. in the liquid.

7. Of the substances investigated only diacetyl and acetylpropionyl and, within limits, acetylbenzoyl were considered to be of suitable properties for use as reference substances in detailed investigations on cream deodorization equipment.

1. Continuous growth of lactic acid streptococci in autoclaved milk allows strains, originally insusceptible, to become susceptible to inhibition by raw milk.

2. Variants of the original organisms, the growth of which is inhibited in raw milk but not in autoclaved milk, appear to predominate in strains maintained in autoclaved milk.

3. The time required to allow these changes to occur in autoclaved milk varies with the strains used.

The viscosity of concentrated skim-milk changed markedly with time upon the addition of urea, calcium-complexing agents or alkali, the viscosity-time curve passing through a distinct maximum. Also, the characteristic opacity of the milk disappeared in the presence of these reagents.

These phenomena are explained in terms of a postulated expansion of the casein micelles followed by their disintegration into ‘molecular’ units.

Electrostatic repulsions within the casein micelles are considered to be the cause of expansion.

Calcium, either in the ionic atmosphere of the casein micelle or as calcium bridges within the micelle, and hydrogen bonds are regarded as two of the major factors responsible for maintaining the stable structure of the casein micelle.

The authors wish to thank Mr E. F. Woods, Division of Protein Chemistry, C.S.I.R.O., Melbourne, and various staff members of the Dairy Research Section, C.S.I.R.O., Melbourne, for helpful discussion during the preparation of the manuscript. The technical assistance of Miss Audrey K. Morris is also gratefully acknowledged.

1. The effect of 4·5% lactose in solution on the vapour/liquid equilibrium relationship for acetylpropionyl, acetyl-iso-butyryl, acetylvaleryl, acetylcaproyl and acetylbenzoyl has been investigated.

2. There was no appreciable effect on the relationship for acetylpropionyl and acetylbenzoyl.

3. For acetyl-iso-butyryl there was a lowering of the coefficient at low concentration of diketone in the liquid, but at concentrations above 4 p.p.m. there was no appreciable effect.

4. For acetylvaleryl at low concentrations in the liquid the coefficient was lowered, and at concentrations above 3·0 p.p.m. in the liquid it was increased.

5. For acetylcaproyl the coefficient was lowered by presence of 4·5% lactose in the solution.

6. Change in concentration of lactose over the range 0·5—10% had little effect on the vapour/liquid equilibrium coefficient for acetylpropionyl and acetylbenzoyl, but it caused a progressively increased reduction in the coefficient for acetyl-iso-butyryl, acetylvaleryl and acetylcaproyl.

Flavour concentrates from mature New Zealand Cheddar cheese were prepared by steam distillation methods. Using 2:4-dinitrophenylhydrazine reagent the presence of carbonyl compounds in the distillate was established. By means of paper chromatography, and other methods, the following compounds were identified in the distillate: acetoin, diacetyl, acetaldehyde, acetone, 2-butanone, 2-pentanone, 2-heptanone, 2-nonanone and 2-undecanone, and their approximate concentrations in New Zealand Cheddar cheese were determined. The possible role of these compounds in the formation of Cheddar cheese flavour is discussed.

The authors gratefully acknowledge the helpful advice and criticism of Dr H. R. Whitehead and Dr W. A. McGillivray, and the technical assistance of Mr P. L. McLaughlin.

The citric acid content of 440 samples of milk has been determined. The average citric acid content of bulk milk of an estimated 10,000 cows was 0·172%. The average citric acid content of the mixed milk of a Holstein herd of 35 cows was 0·168%, and the average citric acid content of the milk of individual cows was 0·171–0·189%. The coefficients of variation of the citric acid content were: in bulk milk 3·49, in mixed herd milk 7·74 and in the milk of individual cows 10·6–13·5. The citric acid content of the milk of individual cows may differ significantly. The citric acid content of milk in the grazing season was slightly lower and more variable than in the housing season. The stage of lactation had no significant effect on the citric acid content of milk. The time of day of the milking had no influence on the citric acid content of milk. Marked differences may occur in the citric acid content of milk from two successive milkings of an individual cow, and in this study differences up to 0·034% or 23·6% of the lowest value have been found.

1. Data are given on the changes which occurred in the composition of sow's milk during the course of 24 lactations. The mean percentages of the major constituents (192 samples) were: total solids, 20·0%: crude protein, 5·7%: lactose, 4·7%; fat, 8·6% and ash, 0·89%.

2. There were differences in milk composition between sows and between successive lactations, but these were only appreciable for milk fat percentage between sows.

3. A highly significant negative correlation (P < 0·001; r = –0·62) was found to exist between the percentages of protein and lactose in the milk, quite apart from the general trend for protein content to fall with advance in lactation. Possible reasons for this are discussed.

1. Coefficients of distribution for acetoin, diacetyl and acetylpropionyl between butterfat and skim-milk were measured over a range of temperatures up to 190° F. by passing preheated milk containing the reference substance through an hermetic cream separator and analysing the cream and skim-milk obtained.

2. The coefficients increased with rise in temperature, and the relationship between change in coefficient and change in temperature was linear for all three substances.

3. Coefficients of distribution for acetyl-iso-butyryl, acetylvaleryl, acetylcaproyl and acetylbenzoyl between butterfat and 4·5% lactose solution were measured at 40° and 60° C. by direct separation after equilibration, followed by estimation of the diketone in the lactose solution.

4. Coefficients of distribution at 100° C., obtained by linear extrapolation from these results, were: acetoin, 0·255; diacetyl, 1·27; acetylpropionyl, 5·9; acetyl-iso-butyryl, 20; acetylvaleryl, 23; acetylcaproyl, 152; acetylbenzoyl, 48.

5. The coefficients of distribution for diacetyl and acetylpropionyl between butterfat and 4·5% lactose solution were not affected by change in concentration of the diketone in the lactose solution.

An essentially exponential relationship between maximum viscosity and concentration of casein was observed for concentrated skim-milk undergoing treatment with calcium hydroxide. Positive temperature coefficients of viscosity were found for samples treated with alkali, while negative coefficients were found with sodium hexametaphosphate. The time of appearance of the maximum viscosity was independent of temperature in concentrated skim-milk treated with sodium or calcium hydroxide; however, when sodium hexametaphosphate was used the maximum appeared earlier at higher temperatures. The viscosity of all systems studied decreased with increasing rate of shear, the deviation from Newtonian being greatest when the viscosity was at its maximum. The effects are explained in terms of a swelling and disintegration of the casein micelles as was postulated in the previous paper.

The authors wish to thank Miss Audrey K. Morris, Dairy Research Section, C.S.I.R.O., Melbourne, for valuable technical assistance, and to express their appreciation for the helpful advice of other members of this Section during the preparation of the manuscript.

1. The relationship between vapour/liquid equilibrium coefficient for a tainting substance in cream (mc) as compared with the coefficient for the substance in skim-milk (ms) at 100° C. is given by the formula

where F is the percentage fat content of the cream, and k is the butterfat/skim-milk distribution coefficient at 100° C.

2. The values for the vapour/liquid equilibrium coefficients for acetoin, diacetyl, acetylpropionyl, acetyl-iso-butyryl, acetylvaleryl, acetylcaproyl and acetylbenzoyl in cream, computed by the formula from the equilibrium coefficients for the substances in 4·5% lactose solutions and the distribution ratios at 100° C, were in good agreement with the experimental results for these substances in cream.

3. It was shown that high solubility of a substance in butterfat relative to solubility in skim-milk greatly reduces the steam volatility of the substance from cream.

4. It was concluded that steam volatility of a substance from cream may be affected by temperature of deodorization because of the effect of temperature on the butterfat'skim-milk distribution coefficient.

The volatile sulphur compounds present in New Zealand Cheddar cheese have been studied by means of gas entrainment and steam-distillation techniques. Hydrogen sulphide was the only sulphur compound which could be detected, and after removal of this compound from the gas stream or steam distillate only a mixture of carbonyl compounds remained. Subsequent removal of these compounds completely destroyed the original cheesy aroma of the gas stream or distillate.

The parts played by hydrogen sulphide and carbonyl compounds in Cheddar cheese flavour are discussed.

The author wishes to record his gratitude to Dr H. R. Whitehead and Dr W. A. McGillivray for their helpful advice and criticism and to Mr P. L. McLaughlin for technical assistance.