The radioimmunoassay for brucellosis previously reported from this laboratory was a sensitive and useful method for detecting antibody against Brucella abortus in bovine serum. Changes in the procedure have improved sensitivity, have apparently increased interassay precision, and have made the assay easier to perform.

Serum samples were obtained from 281 heifers vaccinated with Brucella abortus strain 19, and from 50 heifers that had received two injections of killed B. abortus strain 45/20 adjuvant (K45/20A) vaccine. The serological response measured by the brucellosis radioimmunoassay (RIA) was compared with responses measured by other tests.

The serological responses of cattle during the first weeks after strain 19 vaccination were found to give little guide to the frequency of persistent reactions.

In the case of strain 19, persistent reactions were considered to be those occurring 12 or more months after vaccination. In heifers vaccinated at the recommended age, small numbers of persistent reactions were given by the RIA (four in 374 sera), the complement fixation test using warm fixation (CFTW) (six in 383) and cold fixation (one in 185), the serum agglutination test (two in 222) and the indirect haemolysis test (IHLT) (two in 369). The Rose Bengal plate test gave 74 persistent reactions in 374 sera.

Five of the 50 heifers gave particularly prolonged responses to K45/20A vaccine. In these animals the RIA and IHLT remained positive for longer than the CFTW.

The serology of 27 heifers found to be positive to culture after inoculation with Brucella abortus strain 544, was studied. Eighteen heifers had previously been vaccinated with strain 19 or strain 45/20 and nine were unvaccinated. Post-infection serum samples were tested for Brucella antibodies by radioimmunoassay (RIA), complement fixation test (CFT), indirect haemolysis test (IHLT) and Rose Bengal plate test (RBPT).

All of the unvaccinated heifers showed strong humoral responses to experimental infection in the RIA, CFT, IHLT and RBPT. The CFT and RBPT became positive sooner after infection than the other tests in the unvaccinated heifers. However, in vaccinated heifers the RIA was the most sensitive test early in infection and the results of the RBPT were variable. Three of the vaccinated heifers showed weak and inconsistent humoral responses and, in these animals, the RIA gave fewer false negative reactions than the other tests.

Thirty-one laboratories examined a total of 6169 meat samples, 1236 from abattoirs and 4933 from retail and other outlets. Campylobacter jejuni was isolated from 98 (1·6%). A higher isolation rate of 49/1236 (4·0%) was found among abattoir than among retail and other samples (49/4933–1·0%). Twenty-two of the laboratories looked for salmonella; although 94/4002 (2·3%) were positive, in only one sample of minced beef were campylobacter and salmonella found together. Isolation rates for salmonellae were 75/3576 (2·1%) from retail and 19/426 (4·5%) from abattoir samples.

Analysis of the results revealed that (1) the contamination rate of raw red meat by C. jejuni is, in general, very low; (2) when contaminated, numbers of organisms are generally also very low; (3) enrichment procedures were of some value; 41/98 (42%) isolates were detected by enrichment only, but, on the other hand 8 (8%) were direct plate positive/enrichment negative; (4) practice at looking for the organism and increased seasonal temperatures over the survey period did not result in a noticeable increase in isolations; (5) there was no apparent correlation between campylobacter and salmonella isolations.

Twelve strains of Bordetella pertussis, of various serotypes, were tested by agglutination with serial dilutions of six antisera which were thought by their donor to be monospecific for six distinct agglutinogens.

The first three sera were shown, by agglutination of the twelve donated strains and of standard pertussis serotypes, to be monospecific for the three well-recognized major agglutinogens (factor 1, factor 2, factor 3).

The ‘factor 4’ serum gave weak agglutination, but only of strains that possessed factor 2, though not all of such strains. This suggested that it may be a distinct minor component of factor 2.

The ‘factor 5’ serum was very weak, and gave agglutination only with a few of the factor 2 strains.

The titres of the ‘factor 6’ serum were very closely parallel to those of the factor 3 serum, and it may be inferred that these ‘factors’ probably constitute a single antigenic entity.

Seven of the twelve donated strains were shown to be mixtures of two or more serotypes, offering an explanation for discrepancies in serological tests involving the use of these strains on different occasions.

It seems likely that ‘factors 5 and 6’ do not actually exist; and, since ‘factor 4’ is only a minor component of factor 2, any protective role that it may have would probably be covered by the presence of factor 2 in vaccine.

One hundred and five shigella isolates from Addis Ababa were studied to determine serogroup frequency and in vitro antibacterial drug sensitivity.

About 70 % of the isolates were Shigella Flexneri followed by Sh. dysenteriae (15%), Sh. boydii (10%) and Sh. sonnei (5%). All or most of the strains were susceptible to cephalothin, gentamicin, kanamycin, polymyxin B and trimetho-primsulphamethoxazole. Frequencies of susceptibility to ampicillin, carbenicillin and chloramphenicol were, respectively, 79, 80 and 75%. Only 37, 23 and 58% were susceptible to streptomycin, sulphadiazine and tetracycline, respectively.

Resistance to one or more drugs was detected in 85 % while 72 % were multiply resistant. There were 24 different resistance patterns, varying from resistance to one drug to resistance to seven drugs.

The findings have been compared with reports from other countries. This study and several others cited support the view that trimethoprim-sulphamethoxazole is the best alternative drug for treatment of shigellosis particularly in regions with multiple drug-resistant strains.

Drinking water is a recognized mode of transmission of various parasitic and bacterial infections and high prevalence rates of water-borne infection have been found during past studies in rural areas of the Nile Delta (Farag et al. 1979; Khairy. Barakat & Omar, 1978).

Whilst the major danger associated with drinking water is that it may be contaminated with sewage or human excreta, the danger of pollution from animals must not be overlooked. Outbreaks of infection from drinking water are frequently reported (Melnick & Gerba, 1979) and high prevalence rates of diarrhoeal diseases have been found in the Nile Delta (Sallam et al. unpublished data). The supply of potable water is a critically urgent national problem and is of especial importance to communities which must rely on poorly designed supply systems and which lack even the minimum quality control services.

The development of the bacterial flora of neonates with congenital abnormalities of the gastro-intestinal tract was studied in 31 infants during the first 10 days of life. Specimens were collected from the umbilicus, mouth and gastro-intestinal tract on the pre-operative day, at operation and on post-operative days 1, 2, 3, 5, 7 and 10. Bacteria were isolated semi-quantitatively on a variety of plain and selective media and identified by conventional methods.

Staphylococcus albus was the predominant species isolated from the umbilicus; it was recovered from 24 of the 31 babies. The viridans group of streptococci and Streptococcus salivarius were the commonest species isolated from the mouth; there were no differences between the babies with different abnormalities and treatment with antibiotics had no effect on the bacterial flora. Ten babies were colonized by each species on the pre-operative day, and 25 and 19 respectively by the tenth post-operative day. Anaerobic gram-positive cocci were the predominant oral anaerobes. Bacteria were not isolated from the rectal swabs of babies with tracheo-oesophageal fistula (TOF) or small bowel atresia on the pre-operative days. Post-operatively the predominant faecal isolates from babies with TOF were Str. faecalis, Escherichia coli and Clostridium perfringens. About 80% of the babies with small bowel atresia were colonized by Str. faecalis and Bacteroides vulgatus, 60% each by E. coli, Klebsiella aerogenes and Str. faecium. The five babies with necrotizing enterocolitis were colonized by Str. faecalis, E. coli, Cl. perfringens and Cl. difficile; Bacteroides spp. were not recovered from any of them. The commonest facultative species recovered from babies with large bowel obstruction were Str. faecalis and E. coli. B. vulgatus, Cl. perfringens and Bifidobacterium spp. were the commonest anaerobes and anaerobes outnumbered aerobes. No significant isolates were recovered from the wound swabs and none of the babies developed post-surgical sepsis.

Surveillance of influenza in England and Wales is made by monitoring weekly data. Principal indices are deaths, sickness-benefit claims (SBC), laboratory reports and observations from general practitioners (GPs). The 12 winters 1968/9 to 1979/80 have been studied to see which indices best described size and timing of influenza epidemics. A method of plotting the data (called cusums) is suggested which makes it easier to see the effect of small epidemics.

Cusums for GP statistics and respiratory deaths were found to be the most helpful indices for describing both size and timing of the epidemics, followed by total deaths and SBC, which were less specific to influenza, and influenza deaths, which lagged behind other indices. Deaths certified as pneumonia have been increasing over these years, whereas bronchitis deaths have been decreasing and these indices should not be used separately for monitoring.

The laboratory reporting system is important. It confirms the presence of influenza virus in the community and indicates prevalent strains. Because it is a voluntary system with no defined population base the reports are not reliable numerically for estimating relative size of epidemics or for developing cusums.

Cusum plots were unanimous in describing the winter of 1980/1 as one of little influenza activity.

Lumpy skin disease virus strains isolated in Kenya over a period of some 20 years have proved to be serologically identical. They were indistinguishable by indirect fluorescent antibody and serum neutralization tests from the South African Neethling and West African serotypes. These two serological methods proved of value in studying the antibody responses to infection.

While epizootic spread of LSD has occurred in Kenya, most cases are of a sporadic nature and are thought to be the result of accidental contacts with a maintenance cycle. There is evidence of antibody to LSD in the African buffalo (Syncerus coffer) in those areas where LSD is considered to be enzootic in Kenya, and also in small numbers of domestic cattle. No buffalo or bovine sera contained antibody to cowpox virus. An area enzootic for LSD is proposed and it is suggested that the maintenance cycle involves the buffalo. No antibody was found in the other wild ruminant species examined.

The production of enterotoxin A and B by strains of Staphylococcus aureus isolated from nasal swabs of healthy carriers, from lesions of hospital patients and from foods unconnected with outbreaks of food-poisoning was investigated. Sixty-six strains of S. aureus were obtained from human beings, two produced enterotoxin A, 45 produced enterotoxin B. seven produced enterotoxins A + B. Thirty-six strains were isolated from 111 samples of food, one produced enterotoxin A, 16 produced enterotoxin B. The relative incidence of A, B and A + B enterotoxigenicity was assessed.

Groups of pigs were brought to an abattoir by truck and approximately 25 were killed on each of the next 3 days.

While the pigs were in lairage they were given water but were not fed. After slaughter the caecal contents of all pigs were cultured to detect Salmonella spp. The organism was isolated from 70% of 145 pigs killed after 1 day in lairage, 49% of 143 pigs that had been in lairage for 2 days and 41% of 135 pigs that had been held for 3 days.

The influence of multiple plating of fluid cultures on salmonella isolation from animal feeding stuffs was examined. Four platings were made from broth culture after 24 h at 37 °C and four platings from selenite enrichment from 24 h at 43 °C. Selenite enrichment followed broth culture which was used as a pre-enrichment stage. Brilliant green MacConkey agar plates were employed for broth subculture and brilliant green MacConkey and desoxycholate citrate agars for selenite subculture. The eight brilliant green plates subcultured from broth and selenite were examined for salmonellas after incubation for 24 h at 37 °C. The four desoxycholate citrate agars after 24 h at 37 °C were used for motility enrichment. The food sample size was a single 100 g instead of 4 x 25 g cultured in an earlier study. This pooling of samples aimed at technical economy. Quadruple plating played an important part in salmonella isolation from 100 g specimens. The combination of multiple plating with motility enrichment was the most successful technique used.

In this study 197 strains of Escherichia coli isolated from cowpats and the faeces of healthy humans were examined for the presence of enterotoxins. Six strains representing five different serotypes were found to be enterotoxigenic. Four of these strains produced the heat labile enterotoxin and two, the Vero cell cytotoxin.

One hundred and eighty-two raw, 112 pre-cooked and 750 cooked hamburgers composed mainly of beef or beef and pork were subjected to microbiological examination.

Raw hamburgers gave total bacterial counts from 105 to 108 per g, counts of Enterobacteriaceae from 104 to 108 per g, of Escherichia coli from 103 to 105, of group D streptococci from 102 to 104, of Staphylococcus aureus from 3 to 102 and of Clostridium perfringens less than 10 bacteria per g. Of the samples, 32 % contained salmonellas; the highest most probable number was 102 per g but most estimates were below 1 per g. Corresponding figures for the pre-cooked samples were 2–3 log cycles lower, and only one sample contained salmonella. Yersinia enterocolitica was not isolated from any raw or pre-cooked sample.

Three hundred and ninety-five of the cooked hamburgers were prepared by grilling raw hamburgers for between 2 and 5·5 min. These gave total bacterial counts from 105 to 107 per g, and counts of Enterobacteriaceae from 102 to 105 per g. Of the samples, 9·4 % contained salmonellas, always in numbers below 1 per g. The remaining 355 cooked hamburgers were prepared from samples pre-cooked for 10 min at 80 °C. Some were grilled and some fat fried. The total bacterial counts were from 103 to 105 per g, and counts of Enterobacteriaceae below 102 perg. Salmonellae, again in small numbers only, were recovered from 3·5 % of samples.

When hamburgers were artificially contaminated with Salmonella typhimurium it took 5·5 min on a commercial grill, 2·25 min frying in a frying pan and 1·75 min on a household grill to reliably reduce the salmonella count one hundredfold. This means that at many vending places hamburgers are often cooked for too short a time.

D-values were determined for S. typhimurium in hamburger meat at 50, 55, 60, 65 and 70 °C, these values were 7·1, 5·1, 1·2, 0·9 and 0·6 min respectively. It can be concluded that the heating action in the centre of the hamburgers will take place more slowly than in the hamburger as a whole, and that the time between cooking and consumption is very important in reducing the microbial load to acceptable levels.

Pre-cooking (10 min at 80 °C in a water bath) gives a reduction in the numbers of salmonella of about 4 × 103, after which cooking gives a further reduction as mentioned above.

The contamination of linen and air in a bed centre, supply station and ward were compared, as well as the contamination of gowns used by the staff working in the ‘clean’ and the ‘dirty’ rooms of the bed-centre. The contamination of linen and air was low and there was no significant difference between the tested areas. The contamination on gowns used by the staff working in the ‘dirty’ room was significantly higher than that on gowns used by the staff working in the ‘clean’ room. This stresses the importance of dividing a bed centre into ‘clean’ and ‘dirty’ rooms. So organized, a bed centre does not seem to be a hygienic hazard.