In this study we examined 730 faecal samples of offal (mainly liver), mince-meat and sausage meat collected from abattoirs and retail butchers' shops for campylobacters. Campylobacter jejuni or C. coli were isolated from 30·6, 10·5 and 6% of sheep, cattle and pig offal samples respectively. Specimens collected from abattoirs were, in general, more often contaminated than material obtained from retail butchers' shops. Only 1·4% of minced meats and sausage meats contained campylobacters. Most isolates (89·5%) were C. jejuni biotype 1 (Skirrow & Benjamin, 1980) of serotypes 1 and 2 (Penner & Hennessy, 1980). This study shows that animal offal is frequently contaminated with C. jejuni of biotypes and scrotypes commonly isolated from human beings with campylobacter enteritis.

Mice of strain TO, in groups of ten, were inoculated intravaginally with Mycoplasma pulmonis organisms. Seven mice became infected after inoculation of organisms with strong haemadsorptive capacity, four after inoculation of organisms with diminished adsorptive capacity following ten passes in medium, and none after inoculation of apparently non-adsorbing organisms which had been passed 50 times. There appeared to be a correlation, therefore, between the ability to infect and the cytadsorptive capacity of the organisms. There was only a minimal vaginal polymorphonuclear leucocyte (PMNL) response in the infected mice and most of them had ceased to be infected by 35–42 days. In contrast, mice treated with progesterone had enhanced infections; all those given strongly haemadsorbing organisms, and organisms passed ten times, became infected and remained so for at least 42 days. Furthermore, at least ten fold more organisms were recovered from progesterone-treated than from untreated mice, and the PMNL response was much greater. Most of the progesterone-treated mice given organisms passed 50 times did not become infected, but some did, and the organisms recovered from them were fully cytadsorptive. It is postulated that a few cytadsorbing organisms in this inoculum were induced to infect under the enhancing effect of progesterone.

A survey of Campylobacter species in the faeces or rectal contents of domestic animals was carried out using direct and enrichment culture methods. Campylobacters were isolated from 259 (31 %) of 846 faecal specimens. The highest isolation rate was found in pigs (66%); lower rates were found in cattle (24%) and sheep (22%). In pigs all the isolates were C. coli, in sheep and cattle about 75% were C. jejuni. Only five isolations of C. fetus suhsip. fetus were made, all from cattle. More pigs with diarrhoea had C. coli in their faeces than healthy pigs (77% vs 47 %), but such a clear difference in isolation rate between sick and healthy animals was not seen in cattle or sheep.

The enrichment method increased the total isolation rate of C. jejuni and C. coli by 33%, but for cattle specimens it increased it by 69% (from 6·5% to 21%). However, the enrichment method failed to detect 16% of positive specimens (mainly C. coli), so direct and enrichment methods should be used for the culture of campylobacters from animal faeces. The results show that cattle, sheep and pigs constitute a large potential source of campylobacter infection for man.

Among 622 cultures of Salmonella montevideo, 27 biotypes belonging to two biogroups were recognized. One biogroup (10di) was predominant in all animals in Scotland but only in sheep in England and Wales. The other (biogroup 2d) was responsible for almost all human, cattle and poultry infection in England and Wales, but only 24% of human infection in Scotland.

Two strains of Legionella pneumophila (LP) serogroup I, of differing virulence, were examined in terms of numbers of viable organisms in tissues, pyrexia and mortality following aerosol infection. The Corby strain was the more virulent, with Pyrexia and deaths of guinea-pigs 3 to 6 days after infection. This strain multiplied very rapidly in the lungs to reach a peak of 5 × 1011 viable organisms/lung. Organisms were present in the blood, liver, spleen and kidney. The Philadelphia-1 strain (NCTC 11192) was unable to replicate in the lung and was cleared between 14 and 21 days after infection. Pyrexia was not observed. No guinea-pigs died and viable LP was not found in any organ other than the lung.

Lung lavages on aerosol infected animals were performed and the virulent Corby strain was found to be mainly intracellular. The avirulent Philadelphia-1 strain was found predominantly in the extracellular location. There were approximately 10 times the number of viable virulent LP in the lung macrophage fraction than in the lung PMNL fraction. In comparison, there were approximately equal numbers of the viable avirulent strain in the macrophages and the PMNL. Experimental evidence suggests that the macrophage preferentially supports the growth of the virulent Corby strain compared with the PMNL. The avirulent strain on the other hand appears to be destroyed by both the macrophages and the PMNL.

Pour thousand two hundred and forty six samples of retail processed food were examined for the presence of Escherichia coli. Overall 12% of samples contained this organism, cakes and confectionery being more frequently contaminated (28%) than meat and meat based products (9%). Contamination was more frequent in the summer months than in the colder weather and 27% of the contaminated foods contained > 103E. coli/g. E. coli from meat and meat based products were more commonly resistant to one or more antibiotics (14%) than were confectionery strains (1%). The significance of these findings in relation to the E. coli population of the human bowel is discussed.

Lancefield group A streptococci isolated from recent outbreaks and sporadic cases of scarlet fever were restricted to the following M types 1, 3, 4, 6, 12, 18, 22 and 66. These strains were examined for the presence of streptococcal pyrogenic exotoxins (SPE) types A, B and C by isoelectric focusing in polyacrylamide gels and by immunoprecipitation in agar gels. SPE B was produced by 70% of the strains and SPEC by 40%. SPE A could not be detected in these strains. In contrast, SPE type A was found in 4 of 10 strains, held by the NCTC, that had been isolated before 1940 from patients with scarlet fever. Nine of 12 recent isolates from patients with sore throat uncomplicated by a rash produced SPE C and 4 of these also produced SPE B.

Three of four extreme methods of immunization completely failed to protect mice against challenge with the homologous strain of Fusobacterium necrophorum. Unsuccessful vaccines included (1) broth culture killed by mild heat and emulsified with Freund's complete adjuvant, and (2) a homogenate of heavily infected mouse brains, inactivated by mild heat and given in two doses. Also unsuccessful as a method of immunization was the production of a severe subcutaneous infection with F. necrophorum, followed by curative treatment with metronidazole. Slight but significant protection against subcutaneous challenge resulted, however, from two such infections given in rapid succession. It would appear that the main virulence factors of F. necrophorum are onty weakty immunogenic, and the experiments give little encouragement to the prospect of an effective necrobacillosis vaccine.

Caecal samples from 350 Danish bacon pigs were investigated for salmonella using three methods of isolation. (1) Direct inoculation of 1 g of faeces into 10 ml of Muller–Kaufmann medium (MK medium) with addition of 0·3% Teepol 610 and subculture on Brilliant Green lactose sucrose phenol-red agar (BLSF agar) with 0·3% Teepol G10. (2) Pre-enrichment of 5 g of faeces into buffered peptone water with addition of 1% Teepol 610 followed by enrichment of 1 ml in 10 ml MK medium with 1% Teepol 610 and subculture on BLSF agar with 0·3% Teepol. (3) Incubation of 0·1 ml of the pre-enrichment (2) into 10 ml Rappaport-Vassiliadis medium (RV 10 medium) incubated at 43°C, subculture on BLSF agar.

The MK media with and without pre-enrichment yielded higher findings than the RV 10 media. In total, 28 (8%) of the pigs were found positive, representing 11 (7·4%) of a total of 142 herds investigated. Lymph glands were collected at a later date from six of the positive herds. Five of the herds were found positive.

The number of salmonellas in the glands was low, probably less than ten per gram.

Four recently weaned pigs were dosed orally with oxytetracycline. This caused a rapid increase in the incidence of tetracycline resistance (TcR) among Escherichia coli isolates from the faecal flora. The isolates were differentiated further on the basis of O-serogroup, biotype and resistance pattern. There was no evidence that the administration of the antibiotic selected for a few TcR clones, but rather a relatively large number of TcR strains were identified during the dosing period. Using selective isolation media a proportion of these strains were demonstrated in the minority faecal Esch. coli flora before dosing, while the remainder were recognized for the first time after dosing commenced.

The incidence of TcR among Esch. coli isolates also increased after weaning in other pigs which were not dosed with oxytetracycline or any other antibacterial agent. In a proportion of these animals this increase was associated with the dominance of a TcR enteropathogenic serotype (0149:K 91, K 88a, c) in the faecal Esch. coli flora which was probably ingested in small numbers before weaning. The source of other TcR strains was probably the environment in which each pig was placed after weaning.

A review of campylobacter infection in Scotland over five years (1978–82), during which 7808 human isolates were recorded (mean annual isolation rate of 30 per 100000), revealed differences in the epidemiology of the disease between rural and urban populations which were not apparent in the national data. The incidence of infection in the two rural areas studied was greatest in the early months of the year, whereas that in the two urban areas showed a third-quarter predominance. In both urban and rural populations, age-specific infection rates were highest in children less than 5 years old, but this trend was more pronounced in rural than urban populations. Conversely, secondary peaks in age-specific infection rates observed in young adults were more pronounced in the urban than rural populations.

It is postulated that rural children were being infected by campylobacters at an early age by drinking contaminated raw milk which was not normally available to city residents. The lower incidence in adults in the rural populations is interpreted as indicating more widespread immunity, resulting from greater exposure to infection during childhood. The effect of compulsory heat treatment of milk sold in Scotland, introduced in August 1983, is currently being studied.

The ELISA and an immunoblotting technique were used to study F38-type mycoplasmas – an important cause of contagious caprine pleuropneumonia – and a number of related mycoplasma species, subspecies, types or serogroups.

Two-way ELISA cross-reactivity was demonstrated between five mycoplasmas, namely strain F38, Mycoplasma mycoides subsp. mycoides (LC strain), M. equigenilalium, M. primatum and bovine serogroup 7. In addition one-way cross-reactivity was demonstrated between F38 and each of the following mycoplasmas: M. mycoides subsp. mycoides (two SC strains), M. mycoides subsp. capri, and bovine serogroup L. F38 and M. capricolum did not cross-react.

Immunoblot analysis, unlike ELISA, revealed that F38 and M. capricolum were closely related. At least four major protein antigens were shared between F38, M. mycoides subsp. mycoides (SC and LC strains), M. mycoides subsp. capri and bovine serogroup 7. The ELISA cross-reactions (above) shown by M. equigenitalium and M. primatum with each other, with F38 and with other mycoplasmas were not apparent by immunoblotting.

d-Tartrate dehydrase of Salmonella java is an oxygen-sensitive enzyme active in cultures incubated under the poorly aerated conditions of static culture but not in fully aerated shaken cultures nor on plates incubated aerobically. On plates of d-tartrate minimal agar incubated anaerobically the enzyme or the degradation products of d-tartrate are exported from d-tartrate-positive cells and are available to d-tartrate-negative bacteria. This may give misleading growth results when d-tartrate-positive and d-tartrate-negative strains are tested for growth on the same plate of d-tartrate minimal agar.

The lead-acetate test terminated at 24 h, the 24 h turbidity test and the ability to grow on d-tartrate minimal agar within 48 h differentiated 53 S. paratyphi B strains that were negative in each of the three tests from 76 S. java that were positive in each of the tests. An intermediate group of eight strains utilized d-tartrate in Difco bacto-peptone water to give a positive lead acetate reaction at 2 days, were stimulated to a varying degree by d-tartrate in Oxoid peptone water within the same period of incubation and grew poorly on d-tartrate minimal agar. These latter strains may be deficient in a permease controlling uptake of d-tartrate or export of d-tartrato dehydrase.

Inability to utilize d-tartrate is unlikely to be the single character accountable for the reputed enhanced pathogenicity of S. paraptyphi B when compared with S. java. Indications for the existence of an enzyme, complementary to and mutually exclusive with d-tartrate dehydrase, that has a positive correlation with pathogenicity are discussed.

The actiology, epidemiology and clinical presentation of 137 bacteriologically confirmed cases of superficial mycobacterial lymphadenitis identified at the Liverpool Public Health Laboratory between 1969 and 1984 were reviewed. Despite a fall in pulmonary and total extrapulmonary isolates, the annual recovery of mycobacteria from lymph nodes remained relatively constant. Mycobacterium tuberculosis was the cause of infection in 121 patients (88·3%), M. bovis in 6 and the remaining 10 isolates were atypical mycobacteria. In European patients (68·6%) the highest incidence was in the elderly, whereas in non-Europeans (31·4%) the disease almost exclusively occurred in the third and fourth decades. The proportion of isolates from non-Europeans reflected the size of the immigrant population and increased during the period of study.

A dispersal chamber (body box) technique has been used to compare bacterial dispersal from the skin of subjects carrying out a stepping test under controlled conditions while wearing four differing garment systems namely:

(a) basic underwear, cotton ‘blues’ (standard pyjama style jacket and trousers for men or dress for women), ankle socks, boots for men and shoes for women, mask and theatre hat; (b) the basic set covered with a cotton gown; (c) the basic set covered by a gown with a front made from GORE-TEX fabric in which an expanded polytetrafluoroethylene membrane is sandwiched between layers of woven or knitted polyester; (d) the basic set covered with a fully enclosed suit of the same fabric.

A slit sampler was used to measure the number of bacteria liberated in a downward current of air. Six subjects (three female and three male) were studied. Males liberated more bacteria. Covering the ‘blues’ with a cotton gown increased the bacterial count; a gown of the new material reduced the increase by 50%, and the suit cut the dispersal to virtually zero.

Preliminary work suggests that GORE-TEX garments survive laundering better than cotton, and may be cost-effective, but are not yet as comfortable. Research is presently in progress to improve this aspect.

Foods from outbreaks of food poisoning were examined for the presence of staphylococcal enterotoxin A (SEA) by a sandwich ELISA using microtitre trays as the solid phase and SEA antibodies raised in sheep.

The presence of SEA was confirmed by neutralization tests. The toxin was detected in 12 of 15 foods from separate outbreaks of staphylococcal food poisoning; all 15 foods contained a strain of Staphylococcus aureus which produced SEA. For most foods a simple extraction procedure without a concentration step was sufficient to detect the toxin. The method was semi-quantitative and recoveries of SEA added to control foods varied from 30 to 80%. The foods from outbreaks contained between 1 and 10 μg of SEA/100 g. SEA was not found in foods from 21 outbreaks in which an SEA-producing strain of Staph. aureus was not isolated.

A total of 574 samples, of seven different types, were examined for the presence of salmonellas. All the specimens were pre-enriched in buffered peptone water and enriched in Rappaport–Vassiliadis medium (RV medium). In one trial 0·1 ml of pre-enrichment culture of 497 samples (79 chicken carcasses, 228 specimens of minced meat, 100 pork sausages, 19 samples of dried powdered chicken meat, 11 specimens of faeces of healthy pigs and 60 samples of sewage polluted natural sea water) was seeded to 10 ml as well as to 100 ml of RV medium. With the first inoculum (ratio 0·1:10), 111 samples were found to contain salmonellas, while with the second inoculum (ratio 0·1:100), only 102 positive samples were detected. This difference is marginally significant (P < 0·05). In another trial, 0·1 ml, 0·2 ml and 0·5 ml of pre-enrichment culture of 162 specimens (71 chicken carcasses, 40 samples of sewage polluted sea water and 51 samples of sewage polluted river water) were in turn introduced to 10 ml of RV medium. With the 0·1 ml inoculum 93 positive samples were detected, while with the 0·2 and 0·5 ml inocula 93 and 88 positive samples were found. The differences are not statistically significant. In these trials the growth of competing organisms was minimal with ratios of inocula 0·1:10 or 0·1:100.

An outbreak of bluetongue in sheep started in the Menderes valley, Aydin Province, Western Turkey, in October 1977. The severity of the disease indicated that it had not been there before but had been introduced into the area. Anatysis showed that, while it was possible for the virus to have been brought into the area by movement of infected animals, there was also a period of south-easterly winds which could have carried infected midges from Cyprus, where bluetongue was present. During the night of 14–15 October 1977, south-easterly winds could have brought midges infected with bluetongue virus for the 15 h flight at a height possibly of 500 m and at temperatures of about 20 °C. A depression moving north-eastwards accompanied by rain may have affected the landing of midges in the Menderes valley on the morning of 15 October.

An outbreak of arthrogryposis–hydranencephaly in newly born calves occurred in March–May 1980, also in the Menderes valley, Aydin Province. The severity of the outbreak indicated that Akabane virus had not been in the area before but had been introduced in September–November the previous year. While infected animals could have brought the virus into the area, analysis based on the probable time of infection of pregnant dams showed that easterly winds at the end of September or beginning of October 1979 could have brought insects infected with Akabane virus into the Menderes valley from eastern Turkey or northern Syria.

These analyses illustrate the use of meteorological data to backtrack to possible sources and to identify the time of infection.

Influenza A antibodies in serum samples obtained in 1980 from two Indian populations in Central Brazil were compared. The Kren-Akorore, who were first contacted in 1973 and two years later transferred to the Xingu Indian Park (PIX), were compared with Indians from other tribes already living in the PIX before 1975. An analysis was made of the prevalence and distribution of antibodies against the influenza A viruses which have circulated in the civilized world since 1918. Antibodies to the early influenza A viruses were absent in both Indian populations, but A/Hong Kong/1/68 (H3N2) virus apparently circulated in the PIX. No antibody to influenza A/Bangkok/I/79 or to A/Brazil/11/78 (H1N1) was found in any of the sera, whereas antibodies to these viruses were commonly found in urban populations in Brazil. The evidence from influenza antibodies agrees with the information that the Kren-Akorore Indians had been living in complete isolation until 1973, when they were first contacted.

There is recent evidence of bluetongue (BT) and epizootic haemorrhagic disease (EHD) virus infection of cattle in the American tropics, including BT group reactive antibody in Colombian cattle. These observations prompted a study to determine serologically the specific BT and EHD virus types present, and time of infection and to collect Culicoides spp. as potential vectors. A prospective study of BT and EHD virus infection was done on two farms in the Colombian department of Antioquia. Sequential sampling of young cattle indicated acquisition of neutralizing antibody to BT virus serotypes 12, 14 and 17, and EHD virus serotypes 1 and 2. Insect captures showed a high association of Culicoides insignis with infected cattle.