During November and December 1981 more than 50 residents in a village in Derbyshire had an acute gastrointestinal illness. One month later a second outbreak occurred affecting another 22 people. Campylobacter jejuni was isolated from 12 patients; no other gastrointestinal pathogens were identified. A casecontrol study showed an association with the consumption of unpasteurized milk from one particular farm. No new cases were identified for 6 months following the application of a Pasteurization Order from 26 January to 23 February 1982.

Two enrichment broths (Preston and Roman & Doyle's) and four solid media (Preston, Skirrow's, Butzler's and Blaser's) were compared to determine their relative efficiencies in recovering campylobacters from 389 freshly voided seagull faeces, 276 of which were found to contain campylobacters by one or more of the procedures used.

A combination of enrichment in Preston medium followed by plating on to Preston agar gave the highest number of isolates (263). Enrichment in fluid media was shown to be an important part of the technique, as only 85 (30·8 %) of the 276 isolations were made as a result of direct plating.

Very little difference was seen between the two forms of enrichment (P > 0·5) but of the four selective media, Butzler's was significantly less efficient than any of the other three (P < 0·01), because it failed to grow more than a few strains of Campylobacter coli and the NARTC group, which together made up nearly two-thirds of the total number of Campylobacter spp. isolated.

The production of enterotoxins A, B, C and F by strains of Staphylococcus aureus isolated from various clinical sources and from isolates implicated in food poisoning was investigated.

One hundred and ninety one of the 374 clinical strains (51·1%) were found to be enterotoxigenic; of these, 81 (27·7 %) strains produced enterotoxin A, 57 (15·3 %) strains produced enterotoxin B, 23 (6·2 %) strains produced enterotoxin C, and 64 (17·1 %) strains produced enterotoxin F. These enterotoxigenic strains were most frequently lysed by phages of group III (21·5%) or were not typable (22%).

Eighteen of the 29 strains implicated in food poisoning were enterotoxigenic.

The correlation of antigens and bacteriophage patterns with enterotoxigenicity was determined: enterotoxin A being related to a4 antigen, enterotoxin B to phages of 94/96 complex with c1, o antigens, and enterotoxin F to phages of group I with 2632, k1k2, m antigens.

At a pig-fattening farm in the south-western Veluwe which was infected with salmonellas it was sought to achieve salmonella-free fattening in a specially adapted piggery. The test piggery was thoroughly cleaned and disinfected and measures were taken to exclude birds, insects and rodents. An attempt was also made to obtain salmonella-free piglets. Clean clothing, special footwear and disinfectants were used when entering the piggery.

During the experiment an infection was detected in the test piggery caused by the same salmonella serotypes as had only been found immediately before the test at the breeding farm. Other salmonella serotypes occurring at the fattening farm did not find their way into the test piggery, and therefore it can be concluded that after the pigs had been brought in all the hygienic barriers functioned adequately. The test showed that the hygienic measures taken had a beneficial effect on growth performance, even though salmonellas were not entirely excluded.

After fattening the pigs were slaughtered in two groups. The first group was slaughtered in the usual way, but with the second group extra care was taken with the individual singeing of the carcasses and the careful removal of the intestines. Tests on the carcasses showed that 46% of the pigs in the first group were contaminated with salmonellas as against only 7% in the second. From this it can be concluded that slaughter need not lead to further contamination by salmonellas present in the intestines; indeed, carefully carrying out the slaughter process can even reduce the contamination of the surface of pig carcasses by salmonellas.

An outbreak of salmonellosis due to Salmonella saint-paul in two dairy herds was first detected during routine examination of calves soon after calving; infection reached 100% amongst calves and up to 60% amongst milking cows. Excretion by cows continued for over 12 months and by calves for up to 18 months.

The most important factor in controlling the spread of infection was reduction of environmental contamination by removal of carriers, prompt antibiotic treatment of sick calves and reduction in stocking densities.

The development of Rappaport's enrichment medium and elevated temperature incubation as methods of salmonella isolation is traced. The recent recording of a merger of the two techniques by means of Rappaport–Vassiliadis medium is noted (RV medium).

In Cardiff, we have found an earlier modification of Rappaport's enrichment (R25) by Vassiliadis to be efficient in salmonella recovery from environment samples. The current study compares the two media using sewage polluted natural water as test material. Under the conditions of experiment, R25 was more successful in salmonella isolation than RV, although the latter medium inhibited competitive organisms better.

R25 is a convenient enrichment broth for routine use. In combination with pre-enrichment it allows the use of a loop for subculture rather than a pipette. This increases safe manipulation. It also produces a high proportion of positive isolations at the 24 h subculture time in contrast to other enrichment broths. For these reasons it forms an integral part of salmonella isolation methodology in our laboratory.

Studies of group A, M type 49 streptococci from England, Trinidad and Alaska indicate that isolates of this serotype often differ with respect to phage subtype from one geographical areato another, but are generally homogeneous in one place at one time. The findings support the conclusion that acute glomerulonephritis can be associated with a variety of phage subtypes of M type 49 streptococci.

In outbreaks of skin sepsis without nephritis in England, the phage subtypes of M type 49 streptococci isolated from skin lesions of meat handlers were the same as those recovered from skin lesions of non-meat handlers in the same community.

The findings on the Trinidad isolates suggest that M type 49 streptococci of one phage subtype may persist in a population for 9 years and may result in a second outbreak of acute glomerulonephritis.

In an Alaska Eskimo population in whom acute glomerulonephritis was occurring, most of the M type 49 isolates available for testing were of a single phage subtype. Equally prevalent in this population were group A streptococci that exhibited the same T antigen as the type 49 isolates but differed in their serum opacity reaction and phage subtype. This apparently related strain was not typable with available M antisera but showed functional evidence of M protein and is probably a new M type.

A natural outbreak of Salmonella saint-paulinfection in two Institute herds was monitored clinically, bactcriologically and immunologically. This paper describes the findings in calves. Morbidity and mortality became apparent 30 days after S. saint-paul was first isolated on routine sampling of neonatal calf faeces. All heifer calves were treated with a tetracycline or ampicillin preparation when they showed clinical signs of disease, while the effects of intradermal vaccination with heat-killed S. saint-paul were assessed in a proportion of the bull calves.

Antibiotic treatment reduced mortality and the number of persistent excretors; vaccination did not affect mortality but, if given during the first week of life, reduced the duration of faecal excretion. Calves which were untreated and unvaccinated, or vaccinated when older than 16 days, excreted S. saint-paul for periods of up to 18 months. Of six ‘recovered’ calves, which had been negative on faecal culture for 5–8 weeks, four yielded S. saint-paul at necropsy. Agglutinating antibody titres were highest in those unvaccinated calves which were persistent excretors.

In two surveys of three commercial abattoirs a minimal apparatus method for making bacterial counts, the ‘loop-tile’ method, detected the same trends in bacterial numbers on beef carcasses as the ISO reference method applied to the same samples. Both methods showed the carcasses from one abattoir, that with an export licence, to carry consistently higher numbers of bacteria, and one of the four sites sampled on each carcass to be consistently dirtier than the other three.

A total of 495 diarrhoea and non-diarrhoea patients whose ages ranged between 5 and 39 years were examined for the presence of Campylobacter jejuni, Salmonella and Shigella species. About 12% of the specimens from diarrhoea patients were positive for Campylobacter jejuni compared with 6% and 10% for Salmonella and Shigella species. In contrast 2%, 0% and 1% of the samples from non-diarrhoea patients were positive for Campylobacter jejuni, Salmonella and Shigella species respectively.

Most (62%) of the Campylobacter jejunifrom diarrhoea patients were isolated from children under the age of 10 years. This compared with 26 % and 37 % for Salmonella and Shigella species in this age group. The frequency of isolation of Campylobacter jejuni in diarrhoea patients was highest during the dry months of the year. This study demonstrates the importance of Carnpylobacter jejuni as a major bacterial cause of diarrhoea in this part of the world.

Stools, known to have contained salmonellas were cultured in three enrichment media on arrival at the laboratory. The three media were laboratory prepared selenite F, laboratory prepared Muller-Kauffmann tetrathionate and commercially obtained Oxoid Muller-Kauffmann tetrathionate (CM343). Where faecal samples were sufficiently large, they were stored at room temperature and re-examined using the same three enrichment broths. Whether fresh or stored stools were investigated, the laboratory prepared media were significantly more efficient than the commercial medium. In a straight comparison of the two laboratory prepared media, no significant difference in efficiency was evident with fresh stool samples. With stored specimens, however, Muller-Kauffmann tetrathionate was significantly better than selenite F. This finding might be relevant in examining samples delayed in the post.

Experiments have been performed to simulate accidents which may occur during large-scale production of micro-organisms. Four types of accident, which were considered to be the most likely to result in the greatest hazard to health, were simulated using a bacterial model. The accidents were all concerned with faults occurring in the operation of the microbial fermenter. Gross contamination of surfaces occurred in all experiments, but only three types of accident produced a measurable aerosol.

The growth of an enterotoxin A producing strain of Staphylococcus aureus in corned beef was investigated. In the inoculated 6 lb. canned product the bacteria spread throughout the meat and attained high numbers. The rate of spread of the organisms was related to the temperature and length of storage of the cans and the numbers of bacteria inoculated. Cans which had been stored for more than four months showed high counts of the bacteria throughout the meat. It was noted that with the long term contaminated product counts of S. aureus on some selective media may give falsely low results.

Numbers of S. aureus on meat inoculated by handling after removal from the can were initially extremely variable. More uniform distribution and higher counts were attained only if the meat was exposed for some hours at ambient temperature or above. The significance of the results to the investigation of outbreaks of food poisoning suspected of being associated with canned corned beef is discussed.

The incidence of Staphylococcus aureus on turkeys sampled at various stages of processing and further-processing was determined on four occasions at each of three different processing plants. For freshly-slaughtered birds, counts from neck skin varied from plant to plant over the range < 102 to > 105/g but in all cases the corresponding counts obtained from carcasses sampled after chilling rarely exceeded 103/g and the same was true for samples of mechanically recovered meat (MRM), the final raw product examined.

Despite the limited susceptibility of isolates from the different factories to typing by means of either standard human or poultry bacteriophages (55–94% untypable), evidence was obtained with the aid of biotyping for the presence of both human and animal-derived strains. However, some biotypes isolated from MRM were not detected at earlier stages of processing.

At one processing plant, an ‘indigenous’ type of S. aureus was clearly demonstrated. It occurred in high numbers in the defeathering machines (up to 105/swab), was found on carcasses at all subsequent stages of processing over the survey period and was shown to survive routine cleaning and disinfection procedures. Isolates of this type produced unusually large amounts of extracellular ‘slime’ in artificial culture.

Two of the three processing plants yielded isolates which were enterotoxigenic. Of 55 strains from Plant 1, 60% produced enterotoxin C and all were of the ‘indigenous’ type. In the case of Plant 2, only two type D-and one type F-producing strain were found.

The resistance of spores of B. subtilis var. niger produced in liquid synthetic medium and exposed to ethylene oxide on a nylon surface, has been shown to be almost identical to that for spores produced on a traditional solidified complex medium with exposure to the sterilant on aluminium foil. The use of short lengths of nylon tube as carriers allowed easy production and handling, with self-protection of the spore-bearing surface. Addition of a dye provided visual evidence of inoculation without affecting resistance to ethylene oxide. Such a monitor is suitable for use as a standardized biological challenge in routine ethylene oxide sterilization cycles.

The reliability of colicine typing of Shigella sonnei: as an epidemiological marker has been investigated. Strains from 13 of 110 excretors showed variation in type as determined by typing serial isolates and in 14 of 106 epidemics there was a lack of uniformity in typing results.

The simultaneous presence of two types was found in 11·4 % of patients. In vitro variation was seen after 10 days in 6·8 % of isolates from 44 stools; variation had risen to 9% when isolates were re-examined after one month.

These results suggest that restrictions should be placed on the use of colicine typing as an epidemiologic tool.

Resistance of faecal Escherichia coli to ampicillin, tetracycline, sulphamethoxazole and gentamicin was studied in patients admitted to seven different departments in two hospitals. The resistance to ampicillin, tetracycline and sulphamethoxazole in the seven patient groups was 27–57%, 26–56% and 35–63%, respectively. Resistance to gentamicin was found in only one department. An E. coli flora predominantly resistant to ampicillin, tetracycline or sulphamethoxazole (> 50% of the E. coli strains in a faecal sample resistant) was found in 10–38%, 4–30% and 21–35% of the samples. A cross-sectional study focusing on the influence of the use of antimicrobial agents on the occurrence of resistant strains revealed a positive correlation between the annual turnover of broad-spectrum penicillins in various departments and the occurrence of predominantly ampicillin-resistant E. coli strains in these departments.

The microbial and viral flora in the ventilating ducts of an apartment building was evaluated. Several types of sampler (slit sampler, Andersen sampler, large volume air sampler) were used to evaluate the hourly, weekly and seasonal variation of this flora. The mean bacterial concentration was 17·2 c.f.u./m3 with a maximum level at 07.30 h (41·3 c.f.u./m3) and a minimal concentration in the early afternoon (8 c.f.u./m3). The bacterial concentration observed correlated with the relative humidity in the air-ducts although there were no seasonal differences. The bacteria were mainly gram-positive cocci (73·5%) represented by a large number of Micrococcaceae (47·1 %); gram-positive bacilli accounted for 14·2% of the isolates, gram-negative bacilli 120% and gram-negative cocci 0–3%. The majority of the bacteria-carrying particles were in the respirable range with 80·4 % of them being less than 5 γm. The methods used did not result in the isolation of viruses during the winter sampling period.

In April 1982, Katsina, in Northern Nigeria, was affected by an outbreak of gastroenteritis associated with Vibrio cholerae serotype ‘Ogawa’ and 662 patients were admitted to the Katsina General Hospital during a 16-week period. The outbreak affected all ages and both sexes and all parts of the town and its immediate surroundings except the Government Residential Area (GRA). The overall case fatality rate was 7·7%. Male specific case fatality and female specific case fatality rates were 9·7 and 6·2% respectively. ‘Adults’ and those in the 11–20 and 21–30 age groups accounted for most of the cases. The epidemic curve was of a propagated and protracted nature. About 51·7% of all the patients spent between 2 and 5 days in the hospital. A similar pattern was observed for all age groups regardless of sex. Cholera vibrio ‘Ogawa’, Shigella spp., Salmonella spp., Proteus spp. and Escherichia coli were isolated from 16 patients. All well-water samples obtained from the compounds of the cases were contaminated with MPN/100 ml index ranging from 540 to greater than 2400. All samples were positive for faecal coliforms. Salmonella spp., Shigella spp., Proteus spp. and non-0, group 1 (non-0 1) V. cholerae were isolated. Water sellers probably facilitated the spread of the outbreak.

Faecal carriage of bacterial enteropathogens (enteropathogenic Escherichia coli (EPEC), shigellae and salmonellae) was studied in 265 individuals: 65 infants 3–6 months of age (50 bottle-fed and 15 breast-fed), 100 school-age children 8–10 years of age and 100 adults 21–50 years of age. All were apparently healthy, did not have gastrointestinal symptoms, had not received antibiotics in the proceding fortnight and were not malnourished. Enteropathogens were isolated from the faeces of 24 individuals (9–1%). Cultures were positive for enteropathogens in 20% of the infants (both breast- and bottle-fed), 8% of school-age children and 3% of the adults. EPEC was the most frequent isolate. Twelve different serotypes were detected. The highest recoveries were E. coli026: K60 and 044·K74. Shigella was detected only in school-age children (2%) and salmonella only in adults (1%). Campylobacter jejuni and Yersinia enterocolitica were studied only in the school-age children: there was one isolate of each of them. Most enteropathogens isolated were susceptible to the majority of the antibiotics tested. Only four E. coli strains, isolated from bottle-fed infants, could be considered multi-resistant. Two of the strains were E. coli 044: K74 and 020a020c: K61. The remainder were E. coli 0111: K58 and were capable of transferring some of their antibiotic resistance traits to a recipient strain.