An investigation is reported into sepsis due to Staphylococcus aureus in the maternity unit of a provincial hospital and in domiciliary midwifery in the same town.

In the hospital Staph. aureus infections of mother or infant occurred 123 times per 1000 live births in one ward and 116 per 1000 in another, but the use of hexachlorophane powder on the infants in one ward reduced the sepsis rate in that ward to 34 cases per 1000 live births, thus confirming the reports by two teams in Bristol hospitals. Sepsis was more frequent in both mothers and infants after Caesarean delivery than after normal labour. Reasons are given for thinking that ‘sticky eyes’ are usually not caused by the bacteria isolated when the eyes are swabbed.

The rate of infection in domiciliary cases was 39 per 1000 live births. As all these were normal deliveries the sepsis rate cannot be directly compared with that occurring among complicated cases in hospital.

No evidence was found that hospital Staph. aureus strains carried home by newly born infants were causing a significant amount of sepsis among their home contacts within 28 days of delivery.

The importance of nurses as sources of infecting strains of Staph. aureus in maternity work is discussed. The infant-to-infant route of transmission has been stressed in the literature but reasons are given for believing that nurse-to-infant infection in the labour room may also be important under certain circumstances.

I am greatly indebted to Dr V. P. Randall Martin, formerly Senior Medical Officer in the Hampshire County Health Department, who organized the follow-up of patients after leaving hospital and after home delivery, and also to the County Medical Officer Dr I. A. MacDougall for his help.

I am also grateful to Messrs P. R. Mitchell and G. T. Hammond for permission to investigate their wards and to Sister D. M. Carter and her staff, and to the County midwives and Health Visitors, for their co-operation.

Dr G. J. G. King kindly undertook the phage-typing of a large number of strains of Staph. aureus at the Bournemouth Public Health Laboratory, for which I am most grateful.

I am also grateful to Dr W. A. Gillespie for information in advance of publication, and for advice, about the hexachlorophane regimen of which he and his Bristol colleagues were the pioneers.

Bordetella pertussis instified by the intranasal route into the lungs of mice multiply without difficulty even from small inocula, although B. pertussis is not a natural pathogen for mice and mouse to mouse infection could not be demonstrated. When the initial dose was large the bacilli multiplied until the number in the lungs reached a critical level at which the mouse died. With smaller doses the critical level was never reached; a maximum count was achieved in 10–14 days, after which the number of viable bacilli declined. These smaller doses were consistently non-lethal and the figures for the viable counts when plotted gave curves of typical shape which were called ‘sublethal curves’. The decline corresponded in time with the development of specific immunity. Growth in the lung during the first 4 days of the infection was exponential, the rate of increase in the viable count depending on the size of the inoculum; the smaller the inoculum the faster the increase. With infecting doses at about the critical level, the numbers did not increase during the first 3–4 days, and with larger doses they decreased during the first 2–4 days of infection.

Antiserum given with the inoculum reduced the number of viable organisms in the lung at once. Its effect was short-lived, because after 24 hr. the lung count rose; nevertheless, the initial check on the bacteria had converted a lethal in fection into a sublethal infection.

A clinical trial to determine the effect of oral Humatin therapy on the course of the chronic enteric carrier state in a group of twenty-seven typhoid and paratyphoid B carriers is described. It is concluded that the drug when administered orally in this group of carriers fails in the dosage given to effect a cure of the carrier state.

Analysis by epidemiological methods of the spread of the common cold among office workers and their families has failed to account for more than a fraction of the infections as a consequence of known exposure to infected individuals. Infections too mild to be recorded as colds, survival of the virus in the environment and its subsequent dispersal are considered as possible explanations of this.

The distribution of serial intervals between colds suffered by members of the same family indicates that the median interval between the onset of an infection and the first symptom in infection apparently arising from it as cross-infections was between 21½ and 31½ days. This is only slightly longer than the median interval, 2·4 days, observed between nasal installation and the onset of symptoms in experimental infections.

There is some epidemiological evidence that susceptibility to the common cold is reduced during the weeks following a cold.

We should like to repeat, in this the second paper in this series, our thanks to the Ministry of Pensions and National Insurance and to the Shell Petroleum Company, to those members of their staffs and to the nurses and others who helped us in this investigation.

1. A 0·005% solution of the sodium salt of pindone was found to kill wild house-mice (Mus musculus) in 4–6 days. Mice offered a choice between this solution and water drank more water, but the difference in consumption was not statistically significant.

2. The addition of 1% sugar did not appreciably alter the palatability of a 0·005% solution, but a solution of the anti-coagulant containing 10% sugar was more readily accepted than water. This preference was maintained with solution 4 months old.

1. This paper confirms and extends several observations during the past 20 years that, despite many reports to the contrary, the rat is not unduly resistant to initial infection with tubercle bacilli provided they lodge in the lungs.

2. The pattern of pathogenesis in the rat is probably closest to the now classical picture in the mouse, i.e. the response of a species with a low hypersensitivity potential. The pathology of the lesions agreed closely with the descriptions of Wessels (1941) and Kumashiro (1958b) resembling the mouse in most respects but, unlike the mouse, including the production of giant cells.

3. When tested by footpad inoculation with 1/3·5 Old Tuberculin a positive reaction was demonstrated, commencing between 2 and 5 weeks after infection and persisting for several weeks. A fatal systemic reaction could often be induced with large doses of tuberculin given intraperitoneally.

4. In a few cases loss of allergy was shown to be associated with a terminal anergic flare of the type observed previously in mice and guinea-pigs.

1. The use of sodium glutamate as a medium for freeze-drying Lact. bifidus significantly enhanced the stability of the dried product, especially the heat stability.

2. The optimal concentration of sodium glutamate, as a drying medium, was found to depend on the bacilliary concentration; the higher the concentration of the bacilliary suspension used the higher was found to be the optimal concentration of the sodium glutamate.

3. It was possible to classify various drying media according to the degree of sublimation of the product during the process of freeze-drying. Two types were observed; in the first, which included sodium glutamate and glucose, the sublimation rate diminished with the increasing concentration of the medium, whereas in the second group, which included polyvinyl pyrrolidon K 90 (PVP), soluble starch and dextran, no such diminution was observed when the concentration of the medium was increased.

4. By the combination of sodium glutamate or glucose with a medium belonging to the second group sublimation was found to be promoted.

5. The stabifity-conferring effect of sodium glutamate on dried Lact. bifidus was markedly enhanced when the glutamate was combined with a drying medium belonging to the second group. This effect is attributed to a desiccation-promoting property of the latter medium.

6. Lact. bifidus as a freeze-dried preparation showed a somewhat better survival rate when sealed in air than when sealed in vacuo. In the case of other anaerobic organisms tested there was no significant difference in the survival rates under the two methods of storage.

The growth of virulent strains of Bordetella pertussis in the brains of mice was studied by carrying out viable counts on mice killed at various times during the infection. The results suggested that this system conformed to the general model which postulates that the organisms causing death multiply in vivo at a rate which is constant for all doses and that death is certain to occur when the number of organisms reaches a certain constant figure.

Perhaps the most important factor in this route of infection is the lodgement of the parasite in the host, for if this is accomplished a single organism grows until the lethal level is reached. There is no sublethal infection.

In actively and passively protected mice, the growth of the organism is approximately the same as in unprotected controls for the first 4–5 days. At this time there is a striking change in protected animals and the viable count falls rapidly and progressively and the animals survive. At the same time the blood-brain barrier becomes permeable and circulating antibodies diffuse into the brain. In vitro, specific antisera plus complement are highly bactericidal.

1. The main differences between intracerebral and intranasal infections in mice with virulent strains of Bordetella pertussis are in: (1) the responses to small in fecting doses (< 1 LD 50); (2) the action of antisera in controlling infection; (3) the action of toxin on brain and lung; and (4) the rates of increase of the viable count. The two infections can run concurrently in the same mouse without any demonstrable interference.

2. The terminal viable count in the lung and brain is c. 108 organisms.

3. In the brain there is no sublethal infection with virulent strains; probably even single organisms can grow up to the critical level and kill the mouse. In the lung sublethal infections are found in which the count rises to a figure below the critical level and then declines.

4. The action of ‘intranasal’ antiserum is to reduce a lethal infection to a sublethal one in the lung but there is no effect in the brain. ‘Intracerebral’ antisera cannot act until the blood-brain barrier becomes leaky, when they are able to reduce the viable count and eventually sterilize the brain. In the lung ‘intracerebral’ sera have no action against lethal infections but can control small infections (<I LD 50). The sublethal intranasal test measures this effect but it also measures the action of ‘intranasal’ sera and so cannot be used to distinguish the two types of sera and hence the two antigens.

A study was made of the effect of chlorine solutions on broiler chicken carcasses that had been artificially infected with salmonellae. One hundred and sixteen carcasses were infected with widely varying numbers of seven different salmonella serotypes.

Treatment of carcasses with 200 p.p.m. of chlorine for 10 min. usually prevented the subsequent recovery of salmonellae when fewer than 1000 organisms had been inoculated. When larger numbers of organisms were inoculated or when lower concentrations of chlorine were used, salmonellae were usually recovered from the treated carcasses.

It is suggested that the incorporation of chlorination with 200 p.p.m. for 10 min. into the routine processing of poultry carcasses would greatly reduce the number of contaminated birds leaving processing factories.

We wish to thank the management of a local poultry-packing factory for their co-operation, without which this investigation would not have been possible. We are also grateful to Miss P. E. Ellis for technical assistance.

1. Moore's gauze swabs have, in our hands, been shown to be a reliable instrument for the survey of salmonella infection in the environment of two food premises.

2. A large bakery was demonstrated to be regularly contaminated with salmonellae. Thirty-one floor gully swabs out of 111 (27·9%) were positive for salmonellae. Little spread of infection to the general public was experienced.

3. Staff infection with salmonellae was shown to occur in this bakery. Fifteen positive specimens of staff sewage out of 93 were obtained (16·1%). There was little evidence of prolonged contamination of the sewers with salmonellae. Staff infections were probably, therefore, of short duration. The staff probably became infected from contaminated goods in the bakehouse. Rat contamination of sewers was excluded by water trapping.

4. It is not thought that regular examination of staff excreta would be a profitable means of discovering carriers or of preventing contamination of the products. The frequency with which such examinations would have to be performed to discover short-term excreters would be administratively impossible.

5. Of 274 swabs from gullies in abattoirs, 111 were positive for salmonellac (40·5%). Twenty-one different serotypes were isolated. Seventeen of these serotypes are known to occur in animal feedingstuffs.

6. Parallel examination of abattoir swabs and excreta from human infections for S. typhimurium, coupled with phage-typing of the strains, provided information of the geographically local nature of much sporadic illness due to this serotype. Of a total of 30 typable strains, 23 belonged to phage-types found in the abattoirs, or on local farms. Sporadic incidents were represented by 25 of these cultures, 18 of which corresponded in phage-type to those found in the abattoirs or in local farm animals. The phage-types of S. typhimurium isolated from the slaughter houses not infrequently corresponded in timing with the strains isolated from human infections.

1. An H.A.I. test for louping ill in sheep is described and its potentiality investigated alongside a standard C.F. test.

2. Sheep free from infection with, or which have not been vaccinated against, louping ill do not possess serum antibody demonstrable by the C.F., H.A.I. or neutralization techniques as described here.

3. After the administration of commercial vaccine followed by infective virus, C.F., H.A.I. and neutralizing antibodies are detectable. The C.F. antibodies are the least stable and in many instances disappear entirely from the peripheral circulation within 4 months of their appearance; they also appear to be susceptible to fluctuation unassociated with the disease. H.A.I. antibodies elicited under the same circumstances reach a high level early, are maintained for longer periods than C.F. antibodies, and fall to a basic level which tends to be maintained.

4. H.A.I. antibodies do not appear to be elicited in response to inoculation with vaccine only, in contrast to C.F. and neutralizing antibodies which are so elicited. It is suggested in these circumstances, measurement of H.A.I. antibody is the most useful of the serological tests available for surveys and for diagnostic purposes in sheep.

We wish to thank Mr A. Sharman, B.Sc., M.R.C.V.S., formerly of the North of Scotland College of Agriculture, Aberdeen, under whose care and supervision the flock was maintained and who inoculated and bled the animals.

We are particularly indebted to Dr C. E. Gordon Smith, of the London School of Hygiene and Tropical Medicine, for valuable criticism and advice.

A total of 375 separate sampling experiments were performed on groupings of 100 samples each from distinct binomial populations in order to obtain estimates of the frequency of rejection of the hypotheses of homogeneity against alternatives of a trend by means of the usual χ2 or binomial index of dispersion test. These artificial realizations are presented in Figs. 1–5, and compared graphically with corresponding non-central χ2 distributions.

The authors would like to acknowledge the technical assistance of the Research Computer Laboratory of this University in programming and supervising the computational phases of this investigation, which was partially supported by the State of Washington Initiative 171 Funds for Research in Biology and Medicine and the Boeing Employees' Good Neighbor Fund.

1. The existing procedures recommended for the detection of carriers of enteric pathogens are reviewed.

2. The techniques employed in this work at present were re-examined.

3. An investigation of the sera of 67 chronic typhoid carriers and 343 normal persons, suggested that titrations for typhoid Vi, O and H agglutinins are useful screening tests for the detection of typhoid carriers. The examination may be usefully extended to the titration of H agglutinins against paratyphoid B and paratyphoid A, the former as a possible indicator of the carriage of Salm.paratyphid B, the latter as an indicator of previous inoculation with T.A.B. vaccine.

4. Inoculation did not appear to increase the proportion of positive reactions with TVi bacterial suspension, which in uninfected persons was about 1%. In contrast, the proportion of positive reactors with TO suspension rose from 8·8% in uninoculated to 25·3% in inoculated persons, at a serum dilution of 50; and with TH suspension the positive figure rose from 2·7% in uninoculated to 71·4% in inoculated persons at a serum dilution of 25. Inoculation thus impairs the value of TO agglutination and destroys that of TH agglutination as indicators of infection.

5. The investigation of 29 chronic carriers showed that a single bacteriological examination of excreta for the presence of Salm. typhi is not a reliable method of carrier detection. Increasing the number of specimens to three greatly raises the probability of detection.

6. The findings are discussed and a revised form of paragraph 5 of the Ministry of Health Memorandum on the safeguards recommended for adoption by water undertakings is suggested.

The Working Party is grateful to a number of pathologists for the supply of sera from chronic carriers and we should also like to thank Mr F. J. Flyn of the Central Enteric Reference Laboratory for his great technical help.

During a recent epidemic of poliomyelitis three viruses isolated (one Type 1 and two Type 2) elicited both Type 1 and Type 2 CF and precipitating antibodies in the patients whilst only the homotypic neutralizing antibodies were present. These viruses inoculated into animals produced specific CF and neutralizing antibodies, but precipitating antibodies to both types were present in a significant number of animals. These findings suggest that certain polioviruses have complement-fixing and precipitating antigens common to both Type 1 and Type 2 viruses.

We would like to thank our colleagues in the Virus Research Unit for their helpful advice and criticism and Dr J. Hampton (Poliomyelitis Inst., Johannesburg) for the concentrated Type 1 poliovirus vaccine.

1. An outbreak of food poisoning due to S. typhimurium and S. thompson is described.

2. These same two serotypes were found in trifle eaten by the victims.

3. S. typhimurium was found in the stools of four members of the bakery staff and in the stool of one member of the grocery staff handling luncheon meat eaten by those affected.

4. S. typhimurium was isolated from drains in the bakery.

5. The phage-type of S. typhimurium found in the food handlers, drains, trifle and victims of the outbreak was the same.

6. The phage-type of S. thompson found in the trifle was identical with that found in the victims of the outbreak.

7. The phage-type of both salmonellae suggested an egg product source.

8. It is surprising how infrequent these outbreaks are, having regard to the very frequent opportunity for the contamination of confectionery products in bake-houses.

A simple device by which the characters of an unknown organism can be com pared with those of named genera and species is described. The comparison is made by using ‘diagnostic tables’ of characters found to have differential value.

Separate tables for Gram-positive and Gram-negative bacteria were made; the first stage indicated the probable genus into which the unknown fell and a second (and occasionally a third) stage was needed to identify the species or subspecies.

We wish to thank Dr Ruth E. Gordon for much unpublished data which were invaluable in constructing Table 1 gg, Miss Helen E. Ross for helping with Table 1f, Dr Oleg Lysenko for letting us use data from a paper about to be published, and Mr W. Clifford for the photographs and figures.

1. Apart from strain identification of V. cholerae for epidemiological studies by phage-typing and determination of their S-R dissociation cholera bacteriophages may be utilized for other valuable diagnostic aids in the laboratory.

2. The universal susceptibility of V. cholerae strains to group IV cholera bacteriophage and universal lyso-resistance of El Tor vibrios to it provide a specific method for differentiating between the two species of vibrios. This test is simple and convenient for routine diagnostie use and should be of special value for identifying non-haemolytic El Tor vibrios, which are otherwise liable to be missed.

3. The marked insensitivity of the NAG vibrios to cholera bacteriophages, particularly their total insusceptibility to the phage of group II, enables these strains to be readily distinguishable from V. cholerae.

4. Cholera bacteriophages which are stable on maintenance and storage in the laboratory can conveniently be supplied absorbed on filter disks for the use of the smaller laboratories.

It has been found that antigens suitable for routine tests for complement-fixing or precipitating antibodies in the sera of suspected cases of chickenpox or zoster can be readily prepared from tissue cultures of human amnion infected with zostervaricella virus.

Useful antigens were obtained when infected cells were incubated at 36°–38° C. in bovine amniotic fluid diluted with an equal volume of Hanks' solution.

Virus strains gave a good yield of antigen after two or more passages in tissue culture but one strain in its fiftieth passage did not.

Harvested culture fluids require 5- to 20-fold concentration for complement-fixation tests and 100- to 200-fold for precipitation tests; concentration of culture fluids was readily effected by drying from the frozen state after removal of salts by dialysis. Tissue culture antigens gave results by complement-fixation tests which were comparable to those given by a good vesicle fluid.

Some evidence was obtained that the antigens responsible for precipitation were not identical with those fixing complement with convalescent sera.

The ‘ceiling temperature’ of a pox virus has been defined as the maximum temperature (to the nearest 0·5° C.) of incubation at and below which that virus will grow and produce pocks on the chorioallantois of 12-day-old chick embryos, and above which no pocks appear.

Ceiling temperatures have been estimated for: alastrim (2 strains), 37·5° C.; variola major (2 strains), 3·8° C.; ectromelia (3 strains) and monkey pox (1 strain), 39°C.; cowpox (2 strains), 40°C. Five strains of vaccinia and two of rabbit pox were all capable of pock formation at 40·5° C. Above this temperature difficulty was encountered because many embryos died. But the ceiling temperature for two strains of vaccinia and one strain of rabbit pox was probably 41° C. The Utrecht strain of rabbit pox produced some lesions at 41·5° C.—the highest temperature used.

The ceiling temperatures of the viruses used were not correlated with their thermal stabilities at 55°C. in vitro. Thus vaccinia strain, Lederle-7 N, had a high ceiling temperature and a low thermal stability, while variola major had a low ceiling temperature and a high thermal stability. For this reason ceiling temperatures and thermal stability are regarded as distinct characters.

In experiments with twelve of the seventeen viruses of which the ceiling temperatures had been determined, the virulence for the chick embryo was then measured. It was found that, in general, the higher the ceiling temperature of a virus the greater was its virulence for the chick embryo.

The presentation of these results is followed by a brief discussion of their significance and potential use.