Campylobacters isolated from human, animal and environmental sources during the investigation of a milk-borne outbreak, an incident involving goats' milk and an outbreak associated with puppies were serotyped using the Penner and Lior schemes and biotyped using the Preston scheme. Application of these three methods to the incident strains demonstrated that heterogenicity amongst similar strains exists. This study has confirmed the need to use at least two typing methods when investigating epidemiologically-related strains and we suggest that a combination of a serotyping scheme and an extended biotyping scheme is the most useful.

The intradermal inoculation of four badgers with small numbers of Mycobacterium bovis resulted in localized lesions with ulceration which slowly healed by 5 months after inoculation. Lesions of generalized tuberculosis were seen in three badgers, one of which died at 17 months post-inoculation and in the remaining two killed 22 months post-inoculation. In the fourth badger lesions were confined to the draining lymph node of the inoculation site but M. bovis was isolated from the liver. Monthly clinical sampling of faeces, urine, tracheal aspirate and inoculation site exudates detected only the excretion of M. bovis from the inoculation site of one badger. There were marked seasonal variations in body weight but significant weight loss was observed during the second year in all four badgers, particularly prior to death. Four badgers inoculated intratracheally with a similar inoculum of M. bovis and another two control badgers showed no evidence of infection with M. bovis.

Optimal conditions were determined for performing antibody measurements (ELISA), lymphocyte transformation tests and, to some extent, skin tests in badgers. These parameters, together with the bacteriological and pathological studies reported previously (Pritchard et al. 1987), were used to follow the course of intradermal and intratracheal challenge of badgers with bovine tubercle bacilli. Two challenge doses were used for each route of infection and two animals received each dose. None of the four animals challenged by the intratracheal method showed any evidence of infection, suggesting that adult badgers may have some resistance to challenge by this method. All four animals challenged intradermally developed lesion of tuberculosis.

Immunologically the disease passed through three phases. There was an early phase in which lymphocyte transformation to whole BCG steadily and significantly increased, and skin tests to tuberculin became positive but there was little change in antibody levels. This was followed by an intermediate phase of variable skin responses, fluctuating lymphocyte transformation and significant increase in antibody levels. The final phase, which was only seen in two animals with extensive disease, was associated with changing skin reactions and falling lymphocyte responses, together with a sudden increase in antibody levels.

This paper presents the first formal evidence of cell-mediated immunity to tuberculosis in the badger, which may delay onset and prolong the survival of challenged animals.

During the period 1968–85 shigella organisms were isolated from stool specimens of 1562 patients attending Tygerberg Hospital, situated in the south-western province of the Cape of Good Hope of the Republic of South Africa. Shigella flexneri (72% of patients) was the commonest subgroup identified. Sh. sonnei was the second-commonest isolate (20%), with smaller number of Sh. boydii (5%) and Sh. dysenteriae (3%). Sh. dysenteriae has not been isolated since 1979. In 1985 30% of isolates were resistant to ampicillin and 52% to trimethoprim-sulphamethoxazole. During this period 12 cases of shigellaemia were seen, 11 in young infants less than 13 months of age who were malnourished in 6 cases. The single adult had had a previous gastrectomy and splenectomy.

We assembled an international collection of strains from sporadic and epidemic human infection with Streptococcus zooepidemicus (Lancefield group C) for laboratory study. Cultural and physiological characteristics of the isolates were determined, including biotyping with the API 20 STREP test kit and susceptibility testing with penicillin, erythromycin and tetracycline. The strains were examined for bacteriocin production and sensitivity and typed with a specially developed group-C streptococcal bacteriophage system incorporating a panel of 14 phages. Results of these tests gave useful discrimination between many of the strains: differences were shown between each of the major outbreak strains, including those complicated by post-streptococcal glomerulonephritis.

Serious group C streptococcal infection may be caused by S. zooepidemicus and isolates should be identified to species level; the application of a typing scheme such as this may help to distinguish epidemiological patterns of infection.

Three unrelated severe infections with Streptococcus zooepidemicus occurred in England in 1985. The first patient developed septic arthritis, which has not been recorded before with this organism. The second died with septicaemia, pneumonia and post-streptococcal glomerulonephritis, the only record so far of nephritis following sporadic S. zooepidemicus infection and of nephritis and systemic sepsis in the same patient. The third patient experienced septicaemia during pregnancy but recovered without complications. A likely animal source of infection was found in only one case.

Grivet monkeys experimentally infected with two different strains of Leptospira interrogans serovar hardjo showed no signs of severe clinical disease. There were no significant macroscopic lesions in any of the tissues examined, but the organisms were demonstrated in various tissues by immunofluorescent technique and were isolated from the blood and urine of two monkeys and the kidney of one. Abraded skin was shown to be a viable route of infection in non-human primates.

An enzyme-linked immunosorbent assay (ELISA) incorporating the sensitive biotin-streptavidin system was developed to detect IgG antibodies to tetanus toxoid in human serum. Serum samples obtained from 557 normal persons aged 1–65 years from different areas in New Zealand were tested. The proportion of those immune ranged from 60–93% in males, and from 46–86% in females. In the 1–9 years age group 85% were immune. The indirect ELISA is suitable for serological surveys as it is simple to perform, economical and reproducible.

The serotype distribution of 874 strains of Streptococcus pneumoniae was determined in relation to patients' age and to frequency of isolation from systemic disease. Types 14 and 18, in pre-school children, and types 1, 4, 7, 8 and 12 in patients over 5 years of age were significantly associated with systemic disease whereas type 23 in pre-school children, and type 6 in older patients was associated with upper respiratory tract carriage. No significant difference was found in the incidence of other types in systemic disease compared to upper respiratory tract carriage.

Fifteen diagnostic pneumococcal antisera (to types 1, 3, 4, 6, 7, 8, 9, 11, 12, 14, 17, 18, 19, 22 and 23) sufficed for typing 87% of strains.

A sero-epidemiological survey, using an indirect immunofluorescence test for IgG against Coxiella burnetii (phase II), was carried out in the Netherlands. Serum samples taken in 1968, 1975, 1979 and 1983 were tested. Occupational groups with a supposedly high risk of infection (veterinarians, residents of dairy farms and taxidermists) showed a significantly higher percentage of seropositives than defined controls. The percentage of seropositive amateur wool spinners was significantly higher than that of the controls from the same region. Since 1968 there has been no increase in the percentage of infected persons, indicating that, contrary to earlier assumptions, Q fever has been endemic in The Netherlands for a long time already. The increase in numbers of notified cases of overt Q fever is considered to be the result of the recent introduction of a sensitive indirect immunofluorescence test for IgM antibodies against C. burnetii. Antibody percentages in all age classes between 1 and 64 years were much alike, suggesting that most infections occur in early childhood. This is in accordance with the finding that 35% of our patients are younger than 3 years. The possibility of infection related to childbirth and lactation is discussed.

In a recent paper in this Journal by Marshall et al. (1985), which described the bacterial endonuclease DNA analysis (BRENDA) of several Escherichia coli 0126 strains, the following statement was found in the summary: ‘The isolates from the outbreak produced indistinguishable DNA electrophoretic patterns in spite of their assignment to seven different H serotypes… These results support the epidemiological evidence that a single-strain outbreak had occurred, and they cast doubt on the value of H typing for this particular investigation.’ The same 0126 strains that were enterotoxigenic (ST) were treated in two earlier papers (Bettelheim & Reeve, 1982; Bettelheim, 1984) and also on these two earlier occasions the authors raised doubt about the stability of E. coliH typing results.