THE factors affecting strains of Escherichia coli carried in the bowel of man are poorly understood. Some workers have found the coliform population of the bowel to be relatively stable (Sears, Brownlee and Uchiyama, 1950), but others have shown that the faecal coliform flora may change (Cooke, Ewins and Shooter, 1969; Wiedmann and Knothe, 1969) and this, in one instance, has been related to the ingestion of large numbers of E. coli in food . It has also been postulated that these strains of E. coli may be of animal origin, and may reach the prepared food by cross-contamination in the kitchen .It is not known whether all strains of ingested E. coli will establish themselves in the bowel, what numbers if ingested will be detectable in the faeces, and whether strains from different sources or of different serotype or colicine type behave differently. The experiments reported here were performed in an attempt to answer these questions.
MATERIALS AND METHODSVolunteers. These were one man and two women who were members of the laboratory staff. They were aged 33-37 yr, and had no known abnormality of the gastro-intestinal tract. They were not receiving antibiotics and were eating a normal mixed diet. The food eaten by these volunteers was not examined for the presence of coliform bacteria.E. coli strains for ingestion. Twelve strains of E. coli, some of human and some of animal origin, were ingested. The sources, serotype, antibiotic sensitivity, colicinogenicity, and colicine sensitivity of the strains are shown in table I.The human strains were isolates from normal faeces of patients in a medical ward. The four animal strains were faecal isolates from healthy animals.Except on two occasions, the strains were rendered resistant to streptomycin or nalidixic acid in the laboratory before they were ingested. From a 5-hr broth culture of the test strain 0.5 ml was added to 200 ml of cold milk which was drunk in the late afternoon, 3-4 hr after the last meal. The numbers of organisms ingested, determined by surface-viable counts on blood agar, ranged from 105 to 108 except that on one occasion 1011 freeze-dried organisms were ingested. Examination of faeces. Specimens of faeces were examined on at least three occasions before ingestion of E. coli, and on three occasions after its disappearance. In each experiment, every specimen of faeces passed was examined.Specimens were examined as soon as they were passed or were stored for up to 24 hr at