Nocardial taxonomy, like that of other actinomyeete genera, has undergone all the viscissitudes of doubt associated with years of inadequate and sometimes inappropriate methodology. This is reflected by the great expansion of species names allocated to the genus over the first two thirds of this century (Buchanan, Holt & Lessel, 1966), followed by the contraction achieved in the list of approved bacterial names in 1980 (Skerman, McGowan & Sneath, 1980). Created by Trevisan in 1889 for five species, modern taxonomists would now allocate some 20 species to the genus, including only one of Trevisan's original five species (Table 1). As an actinomycetc genus Nocardia has suffered from two particular disadvantages. First in having an apparently distinctive nocardioform morphology on the basis of which organisms were referred to this genus long after crude morphology lost its importance in many other genera. Second in having an unfortunate type species, Nocardia farcinica, with no really well accredited type strain.

Co-cultivation of Legionella pneumophila serogroup I and Acanthamoeba palestinensis in Neff's medium at 35 °C resulted in the intracellular multiplication of the bacteria as demonstrated by electron microscopy and immunofluorescence. In the closed experimental system used, the number of legionellae rose from 107 colony forming units (c. f. u.)/ml initially to a maximum of 1010 c. f. u. /ml on day 5. Legionellae were seen in expelled phagosomes, in some amoebae filling the cytoplasm and in others in which the process of encystment appeared to have commenced.

It is nearly 100 years since Noeard described bovine farcy, a granulomatous disease of cattle characterized by abscesses, sinus tracts and pulmonary disease (Nocard, 1888). This was the first disease due to Nocardia species described. Since that time our knowledge of the role of the nocardiae in human veterinary diseases, of the epidemiology of these diseases, and their diagnosis and treatment, has enlarged rapidly. This paper attempts to give an overview of our current understanding of these diseases.

The genetics of gentamicin resistance in Staphylococcus aureus strains isolated during an outbreak of infection in a dermatology department have been studied. The predominant strain of S. aureus did not appear to possess a plasmid mediating gentamicin resistance though one isolate yielded a plasmid coding for penicillin and gentamicin. Three distinct plasmids were isolated from other phage types of S. aureus which appeared towards the end of the epidemic. There appeared to be a stepwise loss of gentamicin resistance in the predominant strain.

The detection of Legionella pneumophila in water via guinea-pig intraperitoneal injection has been compared with direct isolation of these organisms on semiselective BCYEα media. Both techniques were of similar sensitivity, detecting as few as 102L. pneumophila in 10 ml of water. However, at this concentration, detection using guinea-pig intraperitoneal injection was via the indirect parameters of antibody production and immunofluorescent microscopy. Isolation of L. pneumophila from guinea-pig tissue for further investigation required > 107 organisms injected and in this respect is relatively much less sensitive than direct isolation on semiselective BCYEα media. Since L. micdadei and L. longbeachae are inhibited by one of the selective supplements used, other supplements and techniques need to be developed in order to avoid possible inhibition of other Legionella species.

This paper describes the relationship between different forms of nocardia infection involving the skin. The best recognized of these conditions are systemic nocardiosis and actinomycetoma due to Nocardia spp. in which the organisms are present in either filamentous or grain form respectively. Attention is drawn to evidence for the existence of a primary cutaneous form of nocardiosis which follows inoculation and in which the organisms grow in filamentous phase. It is not clear whether such cases can progress to mycetoma formation if left untreated, or whether other factors are involved. The clinical and laboratory features of these different clinical varieties of nocardia infection are discussed.

Records of 171 cases of bacterial meningitis admitted to Nottingham hospitals from January 1974 to June 1980 were reviewed. The distribution of organisms producing meningitis and the factors influencing mortality in different age groups were assessed. Neisseria meningitidis, Haemophilus infiuenzae and Streptococcus pneumoniae accounted for 69% of all proven cases. The overall mortality was 26% being lowest in patients with meningococcal meningitis (0%) and highest in those with pneumococcal meningitis (53%). The following factors were associated with a poor prognosis: age more than 40 years, or less than 2 months; state of consciousness on admission; high CSF protein concentration; and a positive blood culture. There was no evidence that antibiotic therapy prior to admission affected prognosis. Although many laboratory findings were altered by prior treatment with antibiotics, this did not prevent the establishment of a diagnosis in the individual patient.

An estimate of the benefits which would result from a ban on the sale of non-pasteurized milk in Scotland has been assessed by costing a recent outbreak of milk-borne salmonellosis in the Grampian Region.

The cost of such a ban would not exceed the benefits under any but the most severe assumptions about the values attached to intangible benefits

There are three commonly recognized species of Nocardia that cause disease in a large variety of animals including humans. In the United States, pulmonary or systemic nocardiosis in humans caused by N. asteroides is most frequently diagnosed. It should be noted that N. brasiliensis can cause nocardiosis also. In Central and South America mycetomas induced by N. brasiliensis appear to be more prevalent even though N. asteroides can be seen in this type of infection. Sporadic cases of both mycetoma and nocardiosis caused by N. caviae have been reported. These three species of Nocardia appear to be present in the soils of most countries; but N. asteroides is more frequently isolated in the temperate climates whereas N. brasiliensis predominates in the tropical and subtropical regions of the world. No specific geographic distribution has been noted with N. caviae. Infections involving these three species of bacteria are, therefore, worldwide in occurrence. Even though nocardial infections in both humans and animals were recognized as early as 1888, traditionally, they have been considered rare in frequency as compared to many other infectious diseases. This is a perception that probably is not accurate. During the past twenty years, greater diagnostic awareness combined with more intensive medical and veterinary surveillance have resulted in significantly increased recognition of disease caused by these aerobic actinomycetes.

A short study of salmonella isolation from reptile faeces is described. The samples came from the reptile house at the Bristol Zoological Gardens. The wide range of salmonella serotypes present in the material, including representatives of all four subgenera, is noted. The important factors in the technique of isolation are discussed. These are choice of inoculum, enrichment medium, and plating Medium, useof multiple subculture and serological isolation. If the number of samples available is scanty, an extended technique will probably yield more information than a simple method.

This study attempted to find the incidence of scarlet fever in the Oxford region, including the proportion of patients from whom Streptococcus pyogenes could be isolated. General practitioners collected throat swabs from patients with suspected scarlet fever. The swabs were examined for viral and bacterial pathogens. Children admitted to hospital were used as controls. Twenty-five of 105 patients with suspected scarlet fever grew Str. pyogenes; M type 4 was the commonest type. The clinical diagnosis of scarlet fever was not always confirmed by throat culture. The annual incidence of scarlet fever was estimated to be 0·3 cases per 1000 per year.

A Staphylococcus aureus strain grown once in vivo in rabbit pleural cavities was appreciably more resistant to killing by soluble polymorph bactericidins than its broth-grown counterpart. Parallel results were obtained using an adaptation of Solberg's drug-inhibition method of measuring intracellular killing after critical aspects of it had been evaluated and modified. Individual in vivo grown organisms were surrounded by a layer of less dense material which was not capsular in nature but which made the organisms clump together. This surface coating was lost on subculture in vitro, resulting in a reversion to broth-like susceptibility, thus indicating that the surface coating was largely responsible for the enhanced resistance and that it resulted from interactions between staphylococci and components of the rabbits' natural body fluids. Consequently, growth in plasma produced organisms which mimicked in vivo grown organisms in clumping, surface coating and in resistance to killing. The use of plasma-grown staphylococci in further studies of likely resistance mechanisms in vivo is discussed.

Cefotaxime, a cephalosporin drug, has been shown to be active in vitro against nocardiae, a finding confirmed in this study. Pharmacokinetic studies were performed in mice to define regimens which provided peak serum levels comparable to that achieved in man with currently used doses. These regimens were shown to be effective with only short courses of therapy of rapidly progressive and highly lethal N. asteroides infection, produced by pulmonary challenge of mice. This suggests the possible utility of this drug in human nocardiosis.

A total of 495 calves in 16 batches were examined (117 calves in 4 batches in 1979 and 378 in 12 batches in 1982). They were purchased in markets, transported by road to a farm in Somerset and reared ona milk substitute diet for a period of up to five weeks. Salmonella typhimurium phage type DT 193 was endemic in 1979 and phage type DT 204c in 1982. The mortality rates in the two years were 9·4% and 1·9% respectively. The causes of death were not investigated although the majority were probably due to salmonellosis.

The rate of isolation of S. typhimurium from the rectal faeces of calves in all groups was either zero or relatively low on arrival. It rose to a peak (which was higher in 1979) in the second or third weeks before declining to low levels by the end of the fourth week of residence on the farm.

Data from 162 calves, examined twice weekly for four weeks in 1982, indicated that the distributionof infected calves, based on the number of times that S. typhimurium was isolated from each, was not random. The calves could be assigned to two main categories; those from which the organism was never isolated and those from which it was isolated at least twice. This suggested that salmonella infected calves actively excreted the organism.

The association between salmonella excretion and medication of sick animals with antibacterial drugs was strongest during the second week. Over the four-week period nearly 40% of the calves found to be excreting S. typhimurium were not treated, indicating a high incidence of subclinical infection.

Salmonella excretion by the calves followed a regular pattern and infection was self-limiting within five weeks. The peak in the salmonella excretion rate and the mortality rate were higher in 1979 when phage type DT 193 was the endemic strain. However, in 1982 the calves received 100 p.p.m. furazolidone in their milk ration during the first week of their stay on the farm, and this may have contributed to the differences noted between the two years.

The proportion of salmonella carriers among town-nesting herring gulls increased significantly from 2·1% in 1975–6 to 8·4% in 1979. The range of serotypes carried by herring gulls was similar to that causing infection in man, and it is likely that the gulls ingest these serotypes when feeding at untreated sewage outfalls on the coast. This is supported by the proportion of salmonella carriers being higher among first-year birds (9·7%) than among older birds (2·0%), as it is known that higher proportions of immature herring gulls feed on the coast. Herring gulls carrying salmonellas appeared healthy at the time of capture and at a later date it was assumed that they were not themselves infected. However, their habit of congregating in large numbers on reservoirs and rubbish tips and also at resting sites on farmland often far from feeding and roosting areas, multiplies thejpollution problem and increases the potential health hazard for both man and farm stock. Herring gulls feed at a variety of sites and fly many miles from food source to food source and from feeding areas to the roost. Thus, even within the same day, there is the possibility of the transfer of salmonellas over a much wider area than previously considered.

Living flies collected from three locations were cultured on selective medium for Campylobacter spp. Campylobacter jejuni was isolated from five (2·4%) of 210 flies examined.

These results suggest that the potential hazard to health from the transmission of campylobacters from animals to human food by flies is small.

The numbers of salmonella serotypes in raw sewage, sewage sludge and final effluent at a sewage treatment works were determined. Resting gulls which had previously been feeding on the sewage were disturbed and individual faecal samples tested for the presence of salmonellae. The serotypes were compared with those in the sewage. Six serotypes were isolated from the sewage, Salmonella Stanley being found in all types of sample. Eleven of the twenty gull faeces samples were positive for salmonellae, a carriage rate of 55%. Seven serotypes were found, S. stanley being most frequent. Three serotypes were found in both sewage and gull faeces.

The relative efficiency of three selective enrichment broths (Muller-Kauffmann tetrathionate, Rappaport's and selenite F) was investigated for the isolation of salmonellae from seagull cloacal swabs. Pre-enrichment in buffered peptone water was employed throughout the study, which involved the examination of 560 gulls, sixty (10·7%) of which were found to be carrying salmonellae. Rappaport's broth as modified by Vassiliadis for incubation at 43°C (Vassiliadis et al. 1976) yielded the highest number of positive swabs (57) and the widest range of serotypes. It was significantly more efficient that either selenite F or tetrathionate broth, although the results obtained with Rappaport's broth incubated at 37 and 43°C did not differ significantly (P > 0·5). Eleven serotypes were isolated during, the study, the most prevalent being Salmonella virchow.

This paper presents evidence for the involvement of herring gulls (Larus argentatus) as vectors in the recent outbreaks of Salmonella montevideo in sheep and cattle in Scotland and suggests that the transfer can take place over considerable distances. The breeding area in Scotland of herring gulls which overwinter in N.E. England is remarkably similar to the geographical distribution of the outbreaks. This pattern, together with the feeding behaviour of herring gulls on farmland, the presence of S. montevideo in herring gulls just before their departure from the wintering area and the timing of the return just before the peak of outbreaks are all circumstantial evidence implicating this gull in the outbreaks. The rapid return of these gulls to their breeding areas means that S. montevideo can be transported long distances in one day and raises the possibility that the original source of S. montevideo could have been in N. E. England rather than in Scotland.