Rappaport's magnesium chloride/malachite green medium was employed in the cultural examination of 2476 faecal specimens in parallel with selenite F broth enrichment and direct plate culture on deoxycholate citrate agar. The use of the medium increased the salmonella isolation rate by 26% over the number of isolations from the other two methods, but proved of no additional value in the isolation of shigella organisms. The addition of Rappaport medium is recommended in routine faecal examination.

a large proportion of samples examined had high bacterial counts (more than 100,000 bacteria/ml.); many contained coliform bacteria and some contained socalled faecal coliform strains. Since most of the samples had been pasteurized or manufactured from pasteurized milk, it has been assumed that the bacterial content of these creams is largely due to contamination after pasteurization. It has also been shown in these investigations that the methylene blue reduction test served as a reasonably reliable guide to the hygienic quality of fresh cream despite occasional anomalous results. The purpose of this investigation was first to examine a number of creams, identify as many as possible of the bacterial strains present and arrive at some conclusion about their source of origin. Next it was hoped to compare the results of the methylene blue test with the bacteria in the creams and study any anomalous results that might occur. MATERIALS AND METHODSOne hundred and twenty-nine samples of fresh cream from sources in Worcestershire were brought to the laboratory by Public Health Inspectors or the County Council Milk Sampling Officer. The samples had either been heat-treated as cream or, if not, had been manufactured from heat-treated milk and were usually submitted within 2 hr. of purchase. They were examined as soon after arrival in the laboratory as possible. Viable counts were carried out by the pour-plate method using decimal dilutions of cream in i-strength Ringer's solution. Nutrient agar and McConkey agar incubated aerobically at 370C. were used for the total bacterial count and coliform count. McConkey agar plates were incubated in a water-tight brass canister (Burman 1967;Barrow & Miller, 1967) immersed in a water bath at 44°C. for the so-called faecal coliform count. Plates were examined after 24 and 48 hr. incubation. Where coliform bacteria and especially Escherichia coli were suspected confirmatory biochemical tests (indole production test, methyl-red test and citrate utilization) were carried out. Also a loopful of undiluted cream was spread on a blood agar plate and the plate examined after 24 and 48 hr. incubation. Colonies were picked on to other blood agar plates to obtain pure cultures. Gramstaining, motility, oxidase production, catalase production and glucose fermenta-

(1) An aerobic coccus has been obtained from cultures of the motile butyric acid bacillus under conditions which exclude the possibility of contamination.(2) Descriptions of the coccus and the bacillus are given.(3) The coccus does not fix nitrogen in soil extract containing dextrose.

Pasteurisation at 62.8° C. (145° F.) for 30 minutes leads to a marked reduction inthe total bacterial contentof milk, varying for different specimens from 94 to 99·8 per cent., when the process is carried out under the most exact conditions. At lower temperatures,e.g.60° C., the percentage reduction may vary widely—from 51·6 to 97·4 per cent. The percentage reduction has been found to be slightly less when the same specimens of milk are pasteurised in a large scale apparatus than when treated in small quantities under laboratory conditions, the average reduction being 91·2 per cent, in the former case and 98·4 per cent, in the latter. The reasons for this are discussed.An effectively pasteurised milk should not contain lactose-fermenting bacilli in 1 c.c, and theB. colicontent is a valuable index of the efficiency of a pasteurising process.Milk containingB. tuberculosis, when pasteurised at 62·8° C. for 30 minutes (1) in tubes in a water bath in the laboratory, and (2) on a large scale by a commercial plant, is rendered non-infective to guinea-pigs. By laboratory pasteurisation, even at 60° C. for 30 minutes, milk containing tubercle bacilli is rendered non-infective.