1. It was planned to study the variations of the protein content of milk during the lactation period.

2. The total protein nitrogen and the casein nitrogen were determined by approved methods in about 380 weekly samples of milk taken from twentyseven individual cows of different breeds and at different stages of lactation, from February to July 1937. The albumin+globulin nitrogen was calculated in each case by difference.

3. The statistical analysis of the figures was made by the methods of “factorial arrangement” and by “pairing”, and also by the estimation of the 95% probability range of the mean of the figures available for each week of lactation.

4. It was found that the percentage of total protein nitrogen decreases very significantly from the beginning until the 4th week of lactation, and then rises slowly until the end of lactation, the rise being more pronounced to wards the end. The same was true for casein nitrogen and for albumin+globulin nitrogen, but with the latter only the decrease at the beginning and the rise at the end were sharp. The rise for the rest of the lactation curve, although it did exist, was very slow.

5. The lactation curves for the yield of total protein nitrogen, casein nitrogen and albumin + globulin nitrogen were practically parallel with the lactation curve for the yield of milk.

6. The correlations between the yield of milk and the percentages of total protein nitrogen, casein nitrogen and albumin + globulin nitrogen are studied and the correlation tables with all details are given for each. It is noteworthy that the regression lines are not linear and that with a poor milker an increase of 1 lb. in yield of milk corresponds to a relatively high decrease in the percentage of protein nitrogen, whereas with a high milker this decrease is relatively much smaller.

7. The correlations between the yield of milk and the yield of total protein nitrogen, casein nitrogen and albumin + globulin nitrogen are examined and the correlation tables are given. It is shown that there is a close correlation between the yield of milk and its protein content, from which the conclusion is drawn that a very low protein food does not decrease the percentage of protein in milk as one might expect, but decreases the yield of milk and increases the percentage of protein in milk. On the other hand, an increase of adequate protein in the ration may raise the yield of milk if it is not at its maximum level.

8. It is emphasized that the above conclusions are drawn from a study carried out on a group of cows, and may not be strictly applicable to individual cows.

Variations in protein yield of a cow are due to changes both in her yield of milk and in the percentage of protein in her milk. If it were possible to have, for each day of lactation, figures for the yield of milk and its percentage of protein, the estimation of lactation yield of protein would be easy, but in practice this is not possible since the determination of protein in milk is a long and laborious process.

An account has been given of observations made for over two years on the composition of the milk of the breeds of goat distributed throughout Great Britain. The average percentages of fat and solids-not-fat in the milk of the breeds, together with their standard errors, are recorded. The frequency with which evening and morning samples fell within specified limits of these constituents is indicated for each breed, and diagrammatically for all goats. Detailed analyses of the milk of each breed are given, with the maximum and minimum values encountered and the standard errors of the average figures. Comparison is made of the composition of the milk of the goat with that of the cow. The effect of stage of lactation on yield of milk and on percentages of fat and solids-not-fat are shown by means of curves. the effect of age on the yield and composition of milk is indicated. The average yields per lactation for each breed, based upon the records of animals officially recorded during the last ten years are noted.

1. In a field investigation covering twenty-nine herds, an average of 19·5% of the animals in milk during the first three months of 1937 were yielding milk low in solids-not-fat, estimated by the routine method, at two consecutive milkings.

2. Of the low solids-not-fat cows which were not stale, evidence of mastitis infection was found in an average of 61·5% of cases. Re-examination of some negative animals suggests that the total infection amounted to about 70%.

3. There were apparently at least twice as many “subclinical” cases of mastitis as clinical cases.

The accuracy of the casern number of milk as the basis for a chemical method for the detection of mastitis has been investigated. The number was determined for 247 samples of milk from the individual quarters of sixty-two cows, and the samples were examined bacteriologically for mastitis streptococci. Of the 247 quarters 49% were found to be infected—mostly subclinically—involving 74 % of the cows.

The total nitrogen content of the samples varied from 362· to 687·4 mg. per 100 ml., and the casein nitrogen from 153·0 to 554·0. The casein number varied from 42·3 to 83·4. The casein number averaged 79·2, 80·4, 79·5 and 80·3 in the samples from the uninfected quarters of the Shorthorn, Friesian, Ayrshire and Guernsey cows respectively; and 71·7, 74·1 and 70·3 in the samples from the infected quarters of the Shorthorn, Friesian and Ayrshire breeds respectively.

For the diagnosis of mastitis from the casein number, a figure of 78·0 and less was taken as indicating an infected quarter, and on comparing this chemical method with the bacteriological method, out of a total of 243 quarters, the chemical result differed from the bacteriological for 21 quarters—representing 8·6 % of the total.

Reasons are given for considering that this estimate of the error is excessive and that in practice the casein number would provide a valuable and rehable method for the diagnosis of mastitis.

The solids-not-fat content (expressed as a percentage of fat-free milk) has been determined for 247 samples of milk from the individual quarters of sixty-two cows. The samples were also examined bacteriologically for the presence of mastitis streptococci and 121 were found to be infected, 114 uninfected, and twelve “doubtful”.

The solids-not-fat content of the uninfected samples varied from 8·24 to 10·28 and of the infected samples from 4·26 to 9·92. For the uninfected samples of the Shorthorn, Friesian, Ayrshire and Guernsey breeds it averaged 9·36, 9·11, 9·82 and 9·83 respectively, but in the infected samples only 8·44, 8·38, 8·28 and 9·49%.

Of the 247 samples, 36 % were below 8·80 in solids-not-fat content. Of 121 infected samples, 63% were below while of the 114 uninfected samples only 9% were below this standard. Hence 88% of the samples below 8·80 were infected samples, thus showing that subclinical mastitis accounts for a very high percentage of samples low in solids-not-fat content. In these samples the average value for solids-not-fat was 7·93%. The few uninfected samples that were low in solids-not-fat averaged 8·56%.

1. Fifteen samples of raw and fifteen samples of commercially sterilized milk from the same bulk were analysed for vitamin C by the chemical method (titration with dichlorophenol-indophenol).

2. The raw milk contained on an average 1·83 mg./100 ml. of total (reduced and reversibly oxidized) ascorbic acid. The corresponding figure for sterilized milk was 1·03 mg./100 ml., a loss of 43% of the original value.

3. Storage of the sterilized milk for a period of 4·6 weeks resulted in a further reduction equivalent to about 30%.

1. The effect of commercial sterilization on the total nutritive value of milk was studied in five separate experiments.

2. In the first experiment raw and sterilized milks supplemented with minerals were given in equal amounts to pairs of litter-mate male rats. No differences were found in the growth performance of the rats, but the rats on sterilized milk consumed it more readily than their mates on raw milk. This difference was statistically significant.

3. Rats getting limited but equal amounts of milk in addition to a basal diet which supplied only protein, energy and minerals grew better on raw than on sterilized milk.

4. When the basal diet contained in addition 5% brewer's yeast and the milk intakes were the same as in the previous experiment no difference was found between the two groups.

5. Rats receiving raw milk to which 15 g. of cane sugar was added per 100 ml. grew better than rats receiving equal quantities of similarly treated sterilized milk.

6. Rats given raw milk containing 30 g. per 100 ml. of sugar grew as well as the rats on sterilized milk in the preceding experiment, but rats receiving similarly treated sterilized milk took much less milk than the raw milk rats, grew very poorly and some of them developed symptoms of beri-beri.

7. The difference between raw and sterilized milk is discussed and it is concluded that in these experiments on the total nutritive value of milk vitamin B1 was the first limiting factor of sterilized milk.

The work carried out by us on the nutritive value of commercially sterilized milk shows definitely that certain factors are injured in the course of the heat treatment. The biological value of the proteins is slightly but unmistakably decreased, most probably because of the partial destruction of one or more essential amino-acids. Of the vitamins, vitamin C is the most markedly affected. It is decreased by half and, in view of the severity of the heat treatment, it is remarkable that the loss is not greater. The anaerobic conditions which exist during the application of the highest sterilizing temperatures may account for the survival of a part of this labile factor. Vitamin B1 also suffers serious loss, the destruction amounting to 30 % of the original value. Neither vitamin A and carotene nor vitamin B2 (flavin) are affected by the heat treatment. The fate of other components of milk was not studied by us, but experiments on rats on the effect of sterilization on the total nutritive value of milk indicates that vitamin B1 was the first limiting factor of sterilized milk.

1. The problem of the physical nature of spreading capacity of butter is discussed, and the advantages of methods measuring individual physical properties in absolute units are given.

2. A simple rheometer is described with which the rate of compression of cylinders of butter can be measured. A simple equation for calculating absolute (mean) viscosities from the data obtained with the apparatus is given.

3. Preliminary experiments, in which temperature control was not as exact as might be desired, nevertheless gave an effective measure of hardness. These are quoted to show the applicability of the method, but they also show the extent to which the phenomenon of “false body” occurs when butter is reworked.

4. A preliminary experiment on the work-hardening of butter is described.

1. Direct steam distillation of cheese does not give a true figure for the volatile acids present in cheese.

2. The fat portion of a cheese has a retarding effect on the distillation of higher volatile fatty acids, although it is possible that a complete recovery of the acid may be effected from the fat if the distillation were to be continued for a sufficiently long period.

3. Cheese protein is capable of a permanent retention of a part of some of the acids present, it being impossible to extract these portions by continued distillation.

4. If direct steam distillation of a cheese is to be carried out the results seem to indicate that by collecting only “twice the original volume” a more accurate picture is gained of the distribution of the acids present in the cheese than would be obtained by distilling to “5 times original volume”, since the retention factor of the protein becomes more significant after “twice the volume” (Fig. 2).

An extraction method for estimating volatile acids in cheese has been described. The values obtained were considerably higher than those given by direct steam distillation.

The loss of calcium and phosphorus at various stages in the manufacture and ripening of British hard pressed cheese (Cheddar, Cheshire, Leicester and Lancashire) has been followed. As all the cheese were made from portions of the same bulk milk the results are comparable.

It has been shown that there is remarkably little difference in the values of the hard pressed varieties, but that Stilton contains much less calcium and considerably less phosphorus than any of the other varieties examined.