The last 30 years have been changes in emphasis in the study of streptococci and streptococcal diseases. Earlier work concentrated mainly on the sources and methods of cross-infection and descriptive epidemiology of Streptococcus pyogenes in its major manifestations of respiratory, cutaneous and invasive infection and in the complications of rheumatic fever (RF), scarlet fever (SF) and post-streptococcal glomerulonephritis (PSGN).

This report is based on a study of acute infections of the upper respiratory tract in 1965 and detailed records of such infections in 1963 and 1964. A change from illnesses mainly yielding viruses to illnesses mainly yielding group A streptococci was noted around the age of 5 years. A positive culture for group A streptococci in patients over 4 years of age was highly correlated with a complaint of sore throat and with serological evidence of streptococcal infection. A bimodal age distribution curve for pharyngitis associated with a positive culture for group A streptococci was consistently noted. The incidence was highest in children aged 5–9 but a second smaller peak occurred among adults in the 30–39 age group. The evidence suggests that being female increases the risk of acquiring group A streptococci and of experiencing sore throat.

In an uncompleted study in 1965 microscopic haematuria in the second or third week after acute pharyngitis was found four times more often in patients with either microbiological or clinical evidence of dual infection with both group A streptococci and a virus than in patients with evidence only of infection with group A streptococci.

Prospective studies of the role of viruses in the aetiology of transient haematuria and of acute post streptococcal glomerulonephritis are feasible in general practice and would be most productive if concentrated in children 5–9 years of age.

Over a period of 9 years in general practice temporary enlargement of the spleen was found in 29 episodes of pharyngitis or tonsillitis, in 2 episodes of acute upper respiratory tract infection other than pharyngitis and in 6 episodes of acute cervical lymphadenitis. In five patients more than one episode of illness associated with splenomegaly was recorded.

In 26 of the 37 episodes a possible aetiology was identified. Evidence only of infection with group A streptococci was found in 14 episodes, adenoviruses or coxsackie B viruses were isolated alone in 4 episodes and in 4 episodes the only finding was the presence in the blood of more than occasional atypical mononuclear cells; in 4 episodes there was evidence of both streptococcal and viral infection. Episodes with evidence of streptococcal infection only tended to be of shorter duration and to be more evenly distributed over the year than were episodes without such evidence. Temporary splenomegaly was noted also in two children with varicella (one of whom also had streptococcal infection) and in an adult with probable rubella.

Post-streptococcal complications are known to be common among Ethiopian children. Little is known, however, about the epidemiology of beta-haemolytic streptococci in Ethiopia. A total of 816 children were studied during a one-year period: 24 cases of acute rheumatic fever (ARF), 44 chronic rheumatic heart disease (CRHD), 44 acute post streptococcal glomerulonephritis (APSGN), 143 tonsillitis, 55 impetigo, and 506 were apparently healthy children. Both ARF and APSGN occurred throughout the year with two peaks during the rainy and cold seasons. The female: male ratio among ARF patients was 1·4:1 and 1:1·9 among APSGN. The monthly carrier rate of beta-haemolytic streptococci group A varied from 7·5–39%, average being 17%. T type 2 was the most frequent serotype. Marked seasonal fluctuations were noted in the distribution of serogroups among apparently healthy children. Beta-haemolytic streptococci group A dominated during the hot and humid months of February–May. Strains were susceptible to commonly used antibiotics, except for tetracycline.

The genetic relationships between various serotypes and serogroups of meningococcal strains were investigated by restriction fragment-length polymorphism (RFLP) analysis using a number of random DNA probes and a probe containing a truncated copy of the meningococcal insertion sequence IS1106. The data were used to estimate genetic distance between all pairs of strains and to construct phylogenetic trees for meningococcal strains. B15: P1.16R strains isolated from cases of systemic meningococcal disease in two health districts with a high incidence of disease were clonal in contrast to similar strains from cases occurring in other parts of the UK. Strains from these areas, which contain a similar genomic deletion, were found to be derived from two distinct lineages within the B15: P1.16R phylogenetic group. RFLP data demonstrated that present serological typing systems for the meningoccus do not necessarily reflect true genetic relationships.

Following the occurrence of a case of systemic meningococcal disease in a military camp in Norway, throat cultures and blood samples were collected from 33 healthy individuals belonging to the same troop as the patient (troop A) and from 29 individuals from a different troop (troop B) in the same camp. Serological studies showed that 91% of the recruits had bactericidal antibodies against the disease-causing strain. The isolates of Neisseria meningitidis recovered from the throat cultures were serogrouped, serotyped, and assigned to a clone on the basis of an analysis of the electrophoretic mobilities of 14 metabolic enzymes. None of the 23 carriers in troop A harboured the clone responsible for the case of disease, but 6 carried isolates of the same electrophoretic type, ET-7, which was not identified in any of the 19 carriers of troop B. Individuals in troop A were resampled 2 and 17 weeks after the meningococcal disease episode. Five of the carriers had acquired different clones and one of them changed clone twice in that period. Four of the six newly acquired clones had previously been identified in other carriers of troop A, demonstrating transmission of clones among individuals living and working in close proximity.

Lipopolysaceharide from serostrains of Serpulina (Treponema) hyodysenteriae for serogroups A to I was characterized using sodium dodecylsulphate polyacrylamide gel electrophoresis and silver staining. All strains had lipo-polysaccharide components ranging from 10 to 16 kDa that represented lipid A-core polysaccharide regions, and short O-antigen side chain were also recognized in certain immunoblots.

Serological reactions between lipopolysaceharide and antisera against each of these serostrains were examined by Western immunoblotting. There was relatively little antigenic cross-reactivity between LPS from the nine strains, thus confirming their suitability as serostrains.

Using cross-absorbed sera, isolates within serogroups A and E were shown to possess unique epitopes on the core lipopolysaceharide, distinct from serogroup reactivities. These isolates were therefore identified as serovars within the serogroups.

This study confirmed the usefulness of the serotyping scheme for S. hyodysenteriae, in which the bacteria can be placed into serogroups using unabsorbed sera, and into serovars within these using cross-absorbed sera.

One hundred and three known sexual-contact pairs of patients with culture-proven gonorrhoea who attended St Mary's Hospital, London between May 1989 and February 1991 were identified. All isolates from these patients were serotyped and auxotyped and compared for type concordance within sexual-contact pairs. Serotype was concordant in 80 (78%) of 103 sexual-contact pairs, auxotype in 88 (85%) and auxotype/serovar (A/S) class in 66 (64%) on the first screening. All pairs of isolates showed concordance in both serotype and auxotype when typing was repeated using a single set of serotyping reagents and of auxotyping media. Seventeen serovars. 9 auxotypes and 36 A/S classes were found in this population. Our results suggest that both serotyping and auxotyping may be used as markers to allow tracing of sexual-contact pairs, but that a single set of reagents should be used to ensure maximum reliability.

The DNAs of Shigella sonnei or Shigella dysenteriae type 1 strains isolated in outbrealis of shigellosis or in sporadic cases were analysed by restriction fragment length polymorphism (RFLP). Southern blots of the DNAs of 36 S. sonnei isolates digested by 8 restriction enzymes were hybridized with an Escherichia coli rRNA probe. The S. sonnei strains were unexpectedly diverse in their RFLP. Antibiotypes of the same isolates showed clusters of strains corresponding to the various outbreaks. On the other hand, RFLP analysis suggested concomitant multiple sources of infection rather than a common source and thereby introduced a new insight in the epidemiology of shigellosis. RFLP was also used to trace S. dysenteriae type 1 transmission in a recent cluster of clinical cases. Although antibiotic resistance patterns indicated the presence of more than one strain, RFLP analysis showed that the six isolates were identical clones and suggested the loss of an R episome after one person-to-person passage.

Enterotoxigenic Escherichia coli (ETEC) were the most frequently identified enteric pathogens associated with diarrhoea in 0–5 year old Aboriginal children in tropical north-west Australia with an incidence similar to those from other tropical regions. Heat-stable toxin-producing (ST + ) strains were associated with diarrhoea throughout the year but heat-labile toxin-producing (LT + ) strains were more important in the monsoonal summer season. ST + strains were commonest in children with diarrhoea between 6 and 18 months of age while LT + strains were associated with diarrhoea in children aged 18–24 months. Verotoxigenic E. coli (VTEC) which produced VT1, but not VT2, and enteroadherent (EAF + ) E. coli were significant causes of diarrhoea, mainly in children below 18 months but without a seasonal pattern.

Sera from 50 patients with culture-proven Campylobacter gastroenteritis were examined for the presence of antibodies to Legionella pneumophila. Ten patients (20%) had a positive titre (≥16) as measured by indirect immunofluorescence. Antibodies were detected in only 1 of 36 acute sera but in 10 of 14 (71%) sera obtained more than 10 days after the onset of symptoms. All positive sera contained specific IgM antibodies but specific IgG or IgA could not be detected in any sample. No legionella antibodies could be detected in sera from 42 similar patients with salmonella gastroenteritis. These results were shown to be due to serological cross-reaction between L. pneumophila and Campylobacter.

Commercially reared chickens were challenged with Salmonella enteritidis phage-type (PT) 4 by aerosol, or via the conjunctiva. Inhalation of 2·9 × 102 or 4·2 × 103 S. enteritidis resulted in the production of IgG antibodies to the lipopolysaccharide (LPS) of S. enteritidis PT4. When the aerosol inoculum was increased to 2·4 × 105 bacteria per bird the antibody produced were predominantly of the IgM-class. Chickens challenged with 103S. enteritidis PT4 via the conjunctiva mounted only a poor immune response. Increasing the challenge dose to 108 S. enteritidis resulted in the production of high-titre serum antibodies of both the IgG and IgM classes. Results from this study suggest that aerosols containing small numbers of S. enteritidis PT4 might be responsible for intraflock infection of poultry.

The use of the polymerase chain reaction (PCR) for identifying serotypes of human and bovine rotaviruses was examined. In the identification of 115 human rotavirus samples in stools, results with PCR showed excellent agreement with results of serotyping by an enzyme-linked immunosorbent assay (ELISA) using serotype-specific monoclonal antibodies. Furthermore, the PCR showed a much higher sensitivity (93%) than the ELISA test (82·6%). The PCR method could also be applied for identifying the serotype of bovine rotaviruses.

The influence of temperature on the behaviour of mixed bacterial contamination of the shell membrane of the hen's egg