Gas production in dairy products is a subject of considerable importance, since it is responsible for certain defects in milk, cream, cheese and occasionally condensed milk. Four groups of micro-organisms are concerned, namely, yeasts, lactic acid bacteria, coliform bacteria and spore-forming anaerobes, and of these the fixs two produce carbon dioxide only while the last two produce a mixture of carbon dioxide and hydrogen (with some anaerobes there may be methane as well). Where both gases are evolved the effectiveness of the two as regards ability to produce effervescence or to develop pressure in a closed container or in a hard-pressed substance like cheese differs by virtue of the fact that carbon dioxide is soluble in water (its solubility depending on the pressure and pH of the medium) while hydrogen is insoluble. Until the medium in which such micro-organisms are growing becomes saturated with carbon dioxide the effective gas production is therefore due to hydrogen. On the other hand, in some media the hydrogen may be involved in secondary reactions with some of the constituents while in the nascent condition and will therefore not form gas.

1. As regards the nutritive value of artificially dried spring grass, the protein appears to be unaffected by the temperature of drying even to the extent of scorching the grass. Early autumn dried grass has a biological value intermediate between spring and late autumn grass, while linseed cake has a very low value.

2. The supplementary effect of the maintenance on the production ration has been shown when hay was substituted for straw in the maintenance ration together with a production protein intake of poor biological value.

3. With an excess of protein in the ration most of the excess nitrogen was excreted in the urine, while both the milk yield and the biological value of the food protein fell considerably. It is possible that, together with an adequate amount of lysine in the protein ingested, a certain mixture of amino-acids is essential before maximum milk yield can be obtained.

4. A study of the nitrogen metabolism shows that high-temperature drying has no effect, whereas marked effects were noted with the early autumn grass and more especially with the linseed cake. The hippuric acid nitrogen output decreased with the high-temperature dried grass, possibly the effect of a loss on drying of benzoic acid. With the high-protein diet an increased excretion of creatinine was noted with no alteration in the creatinine intake. The purin metabolism varies with the total nitrogen excretion, the allantoin increasing markedly in the high-protein period. This is due to an increased intake of purin base.

5. The sulphur metabolism shows the results to be expected from previous publications except in the case of the high-temperature dried grass. It appears that the high-temperature treatment has in some way affected the nature of the sulphur compounds in the grass, with a consequent diminution in the retention.

1. The milk and blood iodine of a herd of thirty Guernsey and thirty Holstein cows has been determined, previous to and during prolonged increased iodine feeding to one-half of the herd. The increased amount of iodine was fed as potassium iodide which was mixed uniformly with a grain feed.

2. The blood iodine of the iodized cows was greatly and uniformly increased.

3. The milk iodine was increased. This increase was principally dependent on the amount of iodine which was fed. During late spring, however, the milk iodine from both the iodized and control cows was unusually low. During early autumn the milk iodine from the iodized cows was low.

4. The milk from the iodized cows contained 7–26 times as much iodine as that from the control cows. An average milk iodine of 80μg. % was obtained from the iodized cows throughout a mid-year period of 5 months.

5. This experiment reveals that a milk of increased iodine content can be readily produced by increased iodine feeding.

1. Sucrose and maltose behaved like dextrose, only more specifically in relation to type, in acclerating the rate of growth of lactic acid bacteria in milk. Sucrose was greatly superior to maltose but was never superior to dextrose.

2. Citrate was without significant effect.

3. The sterilization of yeast and dextrose in the medium produced a medium superior to milk + separately heated yeast and dextrose for Streptococcus III. For other types marked differences were not observed.

4. The plant extracts used were found to fall into the following order of ability to accelerate growth: yeast autolysed for 6 days, malt = beer wort, potato, alfalfa, carrot, tomato, clover, yeast autolysed for 2 days, and bean. Malt and beer wort were superior to yeast for some group III organisms.

5. Malt extract and beer wort contain growth factors for some lactic acid bacteria that are not present, or are present only to a small extent in yeast.

6. Alfalfa and clover appear to contain substances specifically stimulating gas production for group III types, especially the streptococci.

7. Normal variations in the content of vitamins B1 and B2 and co-carboxylase in milk do not appear to affect the rate of growth of starter organisms.

In a previous paper (Allen (1)) some methods have been described for measuring gas production by micro-organisms. The objects of the work described here were (1) to measure the gas production (if any) by pure strains of lactic acid streptococci and by starters growing in milk; (2) to measure the total gas evolved by Bact. coli in the early stages of growth in different media, particularly milk; (3) to find the effect on gas production by Bact. coli of growing Str. lactis in the same medium with it and to measure the proportion of Str. lactis relative to Bact. coli required to inhibit gas production completely—information which has a bearing on the defects due to coliforms growing in the early stages of cheese making; (4) to find the influence on gas production by Bact. coli, with and without the concomitant growth of Str. lactis, of subjecting it to anaerobic conditions from the outset; and (5) to measure the effect due to the hydrogen alone (by absorbing the carbon dioxide) on the gas produced by Bact. coli under different conditions.

The methods used for isolation and purification of bacteriophages active against lactic streptococci are described. An account is given of the relationships between nine apparently distinct phage races and eleven strains of lactic streptococci. The phages show a tendency towards strain specificity but there are some races which attack up to four different strains. Resistant strains usually arise on continued incubation of lysed cultures. Cross-resistance tests, in those cases where two or more phages act on one strain of streptococci, indicate that the relationships between the phages do not follow any simple rule.

Though the lactic acid bacteria are recognized primarily as saccharolytic, several workers have recorded observations on their slow proteolytic activity. Von Freudenreich(1) was the first to record the fact that cultures of these organisms in milk, to which chalk had been added to neutralize the acidity, formed appreciable amounts of soluble nitrogen, and these findings were confirmed by Orla-Jensen (2), Barthel(3), and Barthel & Sandberg(4). Anderegg & Hammer (5), in a study of a large number of strains of Str. Lactis, found an increase in soluble nitrogen in some cases and a decrease in others, while occasionally the same strain differed in different tests. In general, cultures which clotted rapidly were more inclined to proteolysis than those which were slower in forming acid. Str. citrovorus and Str. paracitrovorus did not cause protein breakdown. Addition of 0·3% peptone to the milk tended to retard proteolysis or to increase negative values while addition of chalk resulted in more extensive proteolysis. Barthel & Sadler (6) found that starters consisting of mixed cultures of streptococci produced more extensive proteolysis than single species, indicating a symbiotic effect. Sherwood & Whitehead (7) tested the proteolytic powers of several strains of Str. cremoris in chalk milk cultures and found some active and some comparatively inactive. Two strains appear to have formed surprisingly large amounts of non-protein nitrogen. In general they found that acid-producing power was linked with proteolytic power.

Generally speaking, serum and blood do not have a marked effect on the growth of lactic acid bacteria in milk, although certain types appear to be specifically stimulated or inhibited. Heated serum and, to a greater extent, heated blood, usually stimulate growth.

Unheated serum may inhibit certain types, e.g. Str. lactis, when a small inoculum of washed cells is used. Under conditions similar to those in practice, no effect was noticeable. It is unlikely, therefore, that “slow starter” is ever attributable to infiltration of serum or the constituents of serum.

Small amounts of dextrose are without effect on growth (apart from fermentation of the added sugar) and do not initiate the fermentation of lactose.

1. Seventy-five pure cultures of various types of coliform bacteria were grown in sterilized or partially sterilized milk for 24 hr. at 17° C. The cultures of B. aerogenes, B. oxytocus and B. cloacae types as a rule multiplied much more rapidly than those of the B. coli types. Similar results were obtained when twenty-nine cultures of various coliform types were grown in raw milk for 24 hr. at 17° C. There was generally little or no tendency to clumping of the organisms in the milk, even after 3 days' incubation at 17° C.

2. Forty mixed cultures, each consisting of an aerogenes-cloacae type and a B. coli, were grown in milk for 24 hr. at 17° C. In thirty-four instances the aerogenes-cloacae type multiplied more rapidly than the B. coli.

3. Twenty-one of these mixed cultures were also grown in milk for 24 hr. at 22, 30 and 37° C. At 22, as at 17° C, the aerogenes-cloacae type multiplied as a rule more rapidly than the B. coli, while at 30 and 37°C. the reverse was the case.

4. Twelve specimens of bovine faeces were inoculated into raw milk and the cultures kept at 17° C. for 36 hr. Aerogenes-cloacae types occurring in the faeces became enriched in the milk, the coliform flora of the milk at the end of the incubation period frequently consisting chiefly of these types.

5. The greater incidence of aerogenes-cloacae types in summer milk in Scotland as compared with winter milk, found in a previous investigation, may be explained by the fact that when milk is contaminated directly or indirectly with faeces, these types become enriched at the temperatures of holding commonly employed in summer. Such enrichment does not occur in winter, the holding temperature of the milk being as a rule so low that there is little or no proliferation of any coliform types. Consequently, the relative proportions of the various coliform types in winter milk tend to remain similar to those in the faeces.

Of the lactic acid bacteria isolated, 720 strains from thirty-six typical Cheddar cheeses have been classified. The dominant organism in New Zealand cheese has been shown to be Sbm. plantarum. Sbm. casei occurs much less frequently, while betabacteria and betacocci are found in still smaller numbers. Betabacteria tend to replace Sbm. plantarum in the final stage of cheeseripening. The flora of good quality cheese has been found to consist mainly of one or two varieties of Sbm. plantarum, often associated with Sbm. casei or a small proportion of betabacteria. In poor-quality cheese a larger number of types of lactic acid bacteria frequently occur. A simple routine method is described for distinguishing the different groups present in cheese.

The effects which different strains of lactobacilli and betacocci produce in cheese have been correlated with their biochemical characteristics. Sbm. casei in general appears to be beneficial to cheese quality. The strains of Sbm. plantarum isolated have been grouped in four varieties, mainly on the basis of sugar reactions; one variety has been shown to be beneficial to cheese quality, another variety has very little apparent effect, while the remaining two varieties produce serious defects—bad flavours, discoloration and, occasionally, open texture. The same defects are produced by most strains of betabacteria and betacocci when present in large numbers.

The biochemical characteristics of the lactobacilli studied have shown a more regular gradation from Sbm. casei through Sbm. plantarum to Bbm. breve (the only species of Betabacterium isolated) than is indicated from the generally accepted systems of classification.

1. Over 400 samples of spray-dried. milk powder from eight factories operating in England and Scotland have been examined. These samples were taken at different stages of the day's run and were obtained from several types of plant situated in widely different districts. Samples were taken in both the summer and winter months. The samples were, therefore, sufficiently representative to enable general conclusions to be drawn regarding the factors affecting bacteriological quality.

2. In reconstituting milk powders for bacteriological examination it is necessary to ensure (i) that the sample is sufficiently large to be representative, and (ii) that external contamination is avoided during weighing and handling. The use of a 20 g. sample is advocated. Complete reconstitution of the powder is best obtained by shaking the sample with glass beads in a 180 ml. water blank which is contained in a screw-capped bottle. Details are also given of the methods used in bacteriological examination of the samples and for the isolation of pure cultures from them.

1. In good quality raw milk reducing methylene blue at 37–38° C. members of the coliform group and Staph. aureus were among the commonest organisms present at the time of reduction. Other types of staphylococci, micrococci, and certain streptococci were not infrequently present in smaller numbers. In any given milk one type of organism tended to assume predominance.

2. Using the modified test described by Wilson et al. (1), various organisms isolated from milk at the time of reduction were examined for their rate of decolorization of methylene blue when growing in pure culture in raw “sterile” milk (milk freshly drawn from the healthy udder under aseptic conditions and having a plate count on milk agar at 37° C. of 200–300 per ml.). The most rapidly reducing organisms belonged to the coliform group. Following them, in decreasing order, came Str. lactis and some of the faecal streptococci, Staph. aureus, Staph. albus, Staph. citreus, some micrococci, group C haemolytic streptococci and some strains of Str. agalactiae, and aerobic spore-bearers

Attempts to improve the flavour of cheese by inoculation of good quality cheese-milk (pasteurized) with selected strains of streptobacteria gave the best results when relatively small inocula were employed. Large inocula imparted a sharpness to the flavour of the cheese during the early stages of ripening, and this sharpness often persisted when the cheese was mature.

Several possible methods suitable for the propagation of lactobacilli under factory conditions were investigated. The most convenient procedure consisted in the incorporation in the starter “mother” culture (enriched with a vegetable extract) of suitable strains of Sbm. plantarum; these strains were found to grow better in association with some starters than with others.

1. A survey of evaporated milk and canned cream samples from three condensed milk factories and from general sources has been undertaken to determine to what extent non-sterility occurred in commercial practice. In addition to the examination of sound samples many others have been tested which were obviously defective or which were associated with defective packs. 100% of the tins examined from general sources were sterile. The factory samples, especially those of canned cream, had a lower sterility percentage, although it appeared that this was associated with some special circumstance of operation and that a sterile pack should ordinarily be quite attainable.

2. An account is given of the methods used for sampling the can contents and for the recovery of bacteria from unsterile cans and their identification. The necessity for a large inoculum in such sampling is emphasized, since although the bacteria may survive for many months in the tins, they may be present in small numbers.

3. Data are presented regarding the types of defective samples encountered and the associated bacteriological flora. “Bloats” in evaporated milk and thin bitter and putrid cream were the two main causes of spoilage encountered.

4. By using the heat resistance of the organisms which were recovered from the cans as a basis of classification the. defects have been divided into two groups, namely (1) a mixed general flora which does not withstand high temperatures, and (2) a restricted heat resistant flora which is of the sporeforming type.

In recent years the attention of those interested in all aspects of the dairy industry has been focused upon the subject of mastitis, particularly the chronic streptococcal type. On account of the wide distribution of this disease and its adverse effect upon milk and milk products, it has become urgently necessary for the producer, distributor and manufacturer to be able to recognize mastitis in the cow and its effect upon the milk. It is, therefore, desirable that all those interested in dairying should be as fully informed upon the subject as possible. Many simple tests for the detection of mastitis have been described and are in common use, although so many papers have been published showing their limitations that it is surprising that workers should continue to use them(1, 2, 3, 9). Opportunity has been afforded to the writers to make an extensive examination and comparison of such, tests, with the main object of finding out if any simple test, or combination of simple tests, is really satisfactory for the diagnosis of chronic streptococcal mastitis. The enormous literature upon mastitis (much of it uncritical and, therefore, of little use) makes an adequate summary impossible, even if it were desirable in a paper of this nature, but reference may be made to a useful summary by Munch-Petersen (1).

The fermentation products of Betacoccus cremoris in neutral milk are acetic acid and carbonic acid. In acidified milk or in mixed cultures of this bacterium and lactic acid streptococci the products are acetic acid, C02, diacetyl, acetylmethyl carbinol and 2–3 butylene glycol. The latter three substances may be called “C4 compounds”. Diacetyl is only formed when an oxidation with atmospheric oxygen can take place. Carbinol is found in aerobic and in anaerobic cultures. It may be reduced to butylene glycol. This reduction is more complete at lower acidity, but also depends on the strain of Betacoccus used.

All these substances are formed from the citric acid of the milk. However, small amounts of acetic acid may be produced from sugar.

From 1 mol. of citric acid are formed: 2 mol. of carbon dioxide, 1–1·5 mol. of acetic acid, 0·5–0 mol. of C4 compounds.

A reciprocal relation exists between the quantities of acetic acid and C4 compounds. When the amount of acetic acid is high the C4 compounds content is low. The higher the acidity of the medium in which the betacocci are cultivated the higher is the amount of C4 compounds. This means that the fermentation in neutral media is the extreme case of the fermentation process. Then the production of C4 compounds is minimal (0) and that of acetic acid is maximal (1·5).

Many tests have been used for the detection of subclinical mastitis and these have been enumerated by Munch-Petersen(1) and Rosell(2). In view of the continued recommendation and use of these tests in many countries it was felt desirable to make a careful assessment of their value.

The pH ranges of growth of streptococci and lactobacilli in dextrose yeast casein digest broth have been studied. Nearly all the group I and II streptococci grew at pH 8·8, but only the enterococci grew at pH 4·8; all group III types (heterofermentative) grew at pH 4·4. Only a few lactobacilli grew at pH. 8·8 but most were able to grow at pH 4 if other conditions were favourable. The enterococci, Str. lactis and Str. cremoris could be differentiated by the alkaline limits of growth.

1. An investigation was made of the effect on the casein number of (a) advanced lactation, (b) storage of the milk sample at room temperature and in the refrigerator, and (c) storage of the sample at room temperature after the addition of the following preservatives: potassium dichromate, mercuric chloride, formaldehyde, crystal violet and sodium azide.

2. The casein number method of mastitis diagnosis is not applicable to cows in advanced lactation. It did not distinguish between infected and uninfected quarters, as all quarters in the drying-off stage produced milk with a number below normal. At the end of lactation the mean casein, soluble protein plus non-protein, and total nitrogen contents were 21, 86, and 36% greater respectively than when the cows were in full milk.

3. There was a decrease in the casein number of the milk samples on keeping at 16–18° C. or with preservatives owing to the hydrolysis of casein to proteose-peptone and non-protein nitrogenous substances. For the diagnosis of mastitis from the casein number, samples should not be kept at room temperature for more than a few hours before analysis. Keeping at 0–2° C, and at room temperature with the addition of 0·2, 0·05, and 0·01 g. potassium dichromate, formalin or mercuric chloride respectively per 100 ml. of milk, prevented significant reduction of the casein number for 24–48 hr.

By the term “udder count” in this series of papers is understood the colony count obtained when samples of early mid-milk (taken after about ½ pint has been removed from each quarter) are plated in milk agar and incubated for 2 days at 37°C. Dairy bacteriologists have long recognized the fact that cows with udder disease give high-count milk. In fact, some of those engaged in control work have made a practice of advising producers to search for mastitis when, in the absence of other causes, unusually high counts have been obtained in a series of samples. Therefore in low-count milk, at least, the milk agar plate gives some idea of infection in herds. The early observations, however, were not as a rule confirmed by satisfactory evidence of mastitis, i.e. by plating on blood agar or by cultural or serological typing of colonies picked off plates. This criticism also applies to many papers and statements in articles on mastitis as it affects other aspects of the dairy industry, e.g. off-taints, “slow starter” in cheese-making, and butter-making generally. Unless satisfactory evidence of mastitis is produced such work is valueless. Conclusions based on the results of the indirect tests are unreliable.