1. In a winter-feeding experiment with eight Ayrshire cows the effects of feeding a high- and a low-protein concentrate at two rates (4 and 2 lb./gal.), in addition to dried grass and silage, were studied.

2. With these cows and the feeds used increasing the rate of concentrate feeding from 2 to 4 lb./gal. increased the total dry-matter consumption by 2 lb./day and the average milk yield by 1·2 lb./day.

3. For every 1 lb. increase in concentrate fed there was a decrease of 0·68 lb. grass dry matter eaten.

4. The level of protein in the concentrates had no effect on total dry-matter intake, and only a slight, though positive, effect on milk yield.

5. It is concluded that in this work where the diet was made up with high-quality dried grass and silage, feeding either a high- or a low-protein concentrate mixture at the rate of 2 lb./gal. milk gave the greatest profit margin per cow.

A description has been given of a dairy herd, its breed composition, history and management over a period of 8 years. A close study of infection and mastitis was made in the herd, and the bacteriological methods used have been outlined. The data collected have been analysed and the methods of analysis have been described. There were 530 complete lactations for which full records were available and, because of the frequency of milk sampling and the care with which the bacteriological and other tests were made, these data are considered to be suitable for a precise study of the effect on the incidence of infection and mastitis of stage of lactation, season of the year, age and previous infections. These factors will be the subject of a later report.

1. The detailed udder health records of 530 lactations have been analysed on a monthly and lactation basis, so that separate studies of the first infection rate, re-infection rate, total new infection rate and total infection and mastitis incidence for both cows and quarters could be made. These data have been used to study the relationship of stage of lactation, age and season of the year with udder infection and mastitis.

2. With advancing lactation there was a decline in the rate of first infection, reinfection, total new infection and in mastitis in infected cows and quarters.

3. In early lactation there was a marked tendency for infected cows and quarters to develop clinical symptoms.

4. During the last 4 months of lactation there was no significant rise either in infection or in mastitis.

5. There was an increase in infection and mastitis with advancing age, but infected older cows were not more liable to develop clinical symptoms than younger cows.

6. There was an indication that the longer a quarter was free from infection the greater were its chances of remaining so.

7. No major seasonal trends were found in the incidence of udder infection and mastitis, although the incidence was slightly higher in the summer months.

1. An experiment was carried out with seventy-five dry cows to study the effect of taking samples of udder secretion for bacteriological examination on the incidence of new infection in the early dry period. The incidence of new infection in two quarters of each cow, sampled at weekly intervals after drying off, was compared with that in the remaining two quarters sampled in the fourth and fifth week of the dry period.

2. No difference in the new infection rates was found between the quarters that were sampled early in the dry period and those that were not.

3. Significantly fewer (P <0·05) staphylococcus infections occurred in quarters sampled in the early dry period compared with those sampled at a later stage.

4. About half of all the new infections persisted until the next lactation.

5. It was concluded that if natural sealing of the teat orifice or teat sinus takes place it is of minor importance in preventing infection in the early dry period.

1. An experimental study was made of the effect of drying-off by ‘stop’ or intermittent milking on the incidence of new infection in the early dry period of dairy cows. Ninety-one and ninety-seven dry periods were examined following the treatments, and the results were analysed on a cow and quarter basis.

2. The incidence of new infection was similar for groups of cows dried off by ‘stop’ or intermittent milking.

3. The quarters of cows that were not infected on drying-off showed a significantly higher incidence (P <0·05) of new infection following ‘stop’ than following intermittent milking.

4. Of the cows that were not infected at the time of drying off, those which had a previous history of infection showed a significantly higher (P <0·01) incidence of new dry-period infection than cows that had not been infected before.

5. There was a much higher incidence of new infection in cows that were already infected on drying-off compared with cows free from infection at that time.

6. About half of all the new dry-period infections persisted until the next lactation, and about one-third of these resulted in mastitis in the affected quarter.

7. The method of drying-off had no effect on infections that were already established in quarters going dry.

8. The types of pathogen causing infection were similar for both methods of drying-off.

9. The method of ending lactation did not affect milk yield in the lactation following the dry period studied.

1. Records were collected from a herd where many cows were producing relatively high daily yields of milk at the time of drying-off. Animals giving 7 lb. of milk or less daily were dried off by intermittent milking, and those giving over 7 lb. daily were dried off by ‘stop’ milking. The data from 113 dry periods were analysed so that a comparison could be made of the new infection rate in the early dry period according to level of production at drying-off.

2. The new dry-period infection rate was found to increase with yield at drying-off. This was not due to age differences between the groups of cows.

3. The rise in total infection rate with yield was due mainly to an increase in Str. agalactiae infections.

4. A high proportion of clinical quarters were detected in the dry period by the cowmen.

5. The yield in the lactation following the dry period studied was not related to yield at drying-off.

6. New dry-period infections were significantly (P <0·01) more numerous in quarters of cows that had been hand-stripped during lactation compared with quarters of machinestripped cows.

1. The effect of disinfecting the skin of the teats at drying-off on the incidence of new infection in the early dry period has been studied. After the last milking of lactation the udders of alternate cows to go dry were washed with sodium hypochlorite solution (800 p.p.m.), dried, and each teat immersed in a 5% iodine tincture for 20 sec., and then again immersed 24 hr. later.

2. There was significantly (P <0·05) less new infection in cows that had teats disinfected than in control cows. This difference in infection was highly significant (P <0·01) for cows that were not infected on drying-off.

3. The treatment effect was due to a highly significant (P <0·01) reduction in the incidence of infection by Staph. aureus. New infection caused by Str. uberis was not affected.

Further studies of immersion cleaning have shown that no deposition of calcium soap and inorganic calcium salts occurs if the immersion solution contains ethylene diamine tetraacetic acid. Compared with results for the previous year there was a marked improvement in the bacteriological state of liners as a result of the absence of calcium deposits. It has also been shown that the addition of tetramethylthiuram disulphide to the rubber mix from which liners are made gives useful bactericidal properties. The intensity of the bactericidal action was little reduced after liners had been in farm use for 4 months.

Antibiotic treatment eliminated organisms from 61% of staphylococcal and from 80% of streptococcal udder infections. In 84% of quarters from which organisms were eliminated the cell counts of the milk were reduced, about half of them to a normal and about one-third to a late-lactation type of cell count in which the majority of the cells are epithelial cells. The remainder became reinfected before there was time for the cell counts to return to normal. It was found that it may take as long as 5 weeks for the cell counts to return to normal after elimination of mastitis organisms, but in quarters in which the cell count eventually fell there was a decrease in the percentage of polymorphs in the milk to less than 30% one week after treatment.

One of the most important stages governing the accuracy of lactose determinations in milk is the preparation of the protein-free serum. Factors influencing the lactose content of the serum are: (1) the complete precipitation of protein, (2) the formation of lactose-protein complexes, and (3) the degree of hydration of the protein coagulum.

Experiments are described which show how these factors are influenced by different clearing agents operating at different pH values. On the basis of the results obtained a reagent containing zinc acetate, phosphotungstic acid and acetic acid is suggested; it reduces to a minimum any errors introduced by the above factors.

Iodine values of milk fats of grazing cows show wide variations which are chiefly determined by differences in oleic acid content. Among the factors promoting the variations are: individuality of the cow, stage of lactation, breed differences, diurnal and incidental (irregular) day-to-day effects, nature of the food and plane of nutrition.

Results of experiments with feeding an unsaturated vegetable oil indicate that considerable proportions of the unsaturated long chain acids of milk fat were derived from the ingested oil. It is deduced that herbage lipids may contribute materially to the production of milk fat of grazing cows.

From comparisons of the extents to which dietary long chain unsaturated acids are incorporated into milk fats of grazing and semi-fasting cows, it would appear that the rise in iodine value following partial inanition does not result from increased uptake of fatty acids withdrawn from the more highly unsaturated glycerides of the body reserves.

1. Reactivation of alkaline milk phosphatase is affected by the presence of ions, Mg2+, Zn2+ and Mn2+ being activating and Cu2+, Ni2+ and Co2+ being inhibitory.

2. Cu2+ will inhibit reactivation induced by Mg2+, but has no effect upon Zn2+-induced reactivation.

3. Using washed cream, either alone or in admixture with boiled whey treated with cationic resin, reactivation occurs in the presence of added Mg2+, but not when Zn2+ is added.

4. Experiments with EDTA. confirm that metallic ions play a part both in alkaline phosphatase activity and in reactivation.

5. It is suggested that either Mg2+ or Zn2+ or both play a part in reactivation.

1. The nutritive value of milk sterilized according to the Stork procedure has been studied in laboratory experiments by the method of Mitchell, the rat-repletion method and growth trials with rats and baby pigs.

2. Determinations of biological value of the milk proteins with three groups of six rats by the Mitchell method yielded mean values of 87·0, 88·0 and 85·1 for raw, pasteurized and sterilized milk respectively. The corresponding digestibilities amounted to 96·8, 96·3 and 94·4. The differences were not statistically significant.

3. The relative nutritive value of the three types of milk for the repletion of protein-depleted rats was studied with three groups of eight rats. The average weight gain on feeding daily 40 ml. of either raw, pasteurized or sterilized milk during 5 days was 27·7, 26·3 and 22·9 g. respectively. Again these differences were not statistically significant.

4. No differences in growth rate were obtained if three groups of twelve male and twelve female young rats were fed on raw, pasteurized or sterilized milk supplemented with trace minerals for a period of 6 weeks. The growth rate of rats on exclusive milk diets was, however, far below normal.

5. In growth experiments with baby pigs, fed equal amounts of pasteurized or sterilized milk supplemented with trace minerals, small but statistically non-significant differences in weight gain were found, in favour of the pasteurized milk. The average weight gain on sterilized milk in four separate growth trials amounts to 97, 95, 98 and 98% of that on pasteurized milk. The average 3% less gain on sterilized milk was ascribed partly to more difficult adaptation and less intake of milk solids.

6. These experiments failed to demonstrate a significant decrease in nutritive value of milk sterilized according to the Stork procedure.

1. A method for preparing rennet reference standards by freeze-drying is described.

2. A substrate has also been developed using ordinary spray dried separated milk powder which will give reproducible results under specified conditions. These include:

(i) A constant temperature (37° C.) for the duration of the test.

(ii) The exclusion of added calcium ions until actual testing begins.

(iii) The equilibration of the milk at 37° C. for 2 hr. before use.

3. The construction and use of an automatic apparatus for recording rennet coagulation time is also described.

4. The instrument error of a single observation is about 1·0%.

5. With liquid rennet extracts, the total experimental error, based on one sample of three replicates, is 2% or less: with two samples, it is 1% or less.

6. With freeze-dried rennet preparations, the experimental error with two ampoules, sampled thrice with three replicates per sample, was ±2·.0%. With three ampoules it was ±1% or less.

An extraction method is described for determining the volatile fatty acids in cheese. The method, dependent upon Soxhlet extraction and micro-distillation, is speedy with regard to actual manual operation and gives results in accordance with those obtained by lengthier established methods.