The antigenic structure of Bordetella pertussis varies continuously with environmental conditions of growth and, under suitable conditions, any of three forms (or modes) with markedly different antigenic surfaces may be developed. High temperatures and certain ions, such as sodium, potassium, halides, formate and nitrate, favour growth like that on Bordet and Gengou's medium at 35° C. Low temperatures and other ions, such as magnesium, sulphate and mono- and dicarboxylic acids with more than two carbon atoms, favour growth with a markedly different antigenic structure and character. These two extreme forms have been called X mode and C mode respectively. A third mode (called I mode) is induced within a narrow range of intermediate temperatures and ion ratios. Bacterial cells in C mode, unlike those in X mode, do not agglutinate red-blood cells, are not haemolytic or agglutinable by heavy metal salts, have no surface antigen related to Bordetella parapertussis or B. bronchiseptica and are serotypically homogeneous.

The influence of ions is a function of their species and ratios and not of their absolute concentrations or total ionic strength. A number of ions has been arranged in a pattern. In this the influence of any salt is represented at the centre of a line joining its component ions, and a measure of the influence of any salt mixture is given by the position of its imaginary centre of inertia. (For calculating this, mass is represented by the product of salt concentration and conductance ratio and radius of gyration by distance.) The influences of ions and temperature appear not to interact since their combined influence is proportional to the sum of their separate influences.

The change from one mode to another (modulation) occurs without mutation or selection, is complete within 7–15 cell divisions and appears to stem from altered rates of synthesis of at least three kinds of antigen molecule. During the process of modulation, antigens not detectable by agglutination in any of the three equilibrium modes (X, I and C) may appear transiently at the surface.

Mutants may or may not modulate. Modulatable mutants may be distinguishable from the wild type in only one mode. Almost all convalescent antibodies are directed against X mode, but C-mode antibodies have been found in several adults with persistent cough and infection.

I wish particularly to thank Dr E. K. Andersen for her generous gifts of sera, Drs P. L. Kendrick, H. Proom, A. F. B. Standfast and A. J. H. Tomlinson for gifts of cultures and Mr M. Breach for his invaluable help in the preparation of media and sera. I am also indebted to the University of London Central Research Fund and the Governors of Westminster Hospital for grants towards the purchase of materials.

The potency in man of eight different commercial vaccines and an experimental vaccine, as judged by the antibody response after the primary course of two injections and after a booster dose 9–12 months later, was compared. All the vaccines produced good responses to Types 2 and 3 but some were less satisfactory for Type 1. The relation of potency to protection in man is discussed and the use of one of the vaccines as a minimum standard for this country is suggested.

The problem of describing quantitatively the effective ventilation in a room when the air within the room is imperfectly mixed is discussed. It is suggested that the protection afforded by the ventilation to any given position against airborne contamination liberated at any other position can be best expressed in terms of the integrated exposure at the first point following liberation of a tracer substance at the second. This quantity is called the transfer index and its reciprocal the equivalent ventilation approaches numerically to the rate of supply of ventilating air as the mixing of air within the room approaches completeness. Nitrous oxide is a convenient tracer gas for making such measurements.

A method is also described whereby estimates of the transfer index can be made employing only such apparatus as is easily available in a reasonably well-equipped laboratory. This method employs acetone vapour as a tracer substance. The concentration of this vapour in the air is measured by the pH change produced in a dilute solution of hydroxylamine hydrochloride as it flows down a cotton wick exposed to the atmosphere.

Experiments performed on spore formation in B. anthracis, in a wide range of closed atmospheres of chemically controlled humidity, showed that the sporing time of B. anthracis is prolonged considerably as the humidity decreases, both at 26 and 37° C. The effect is less noticeable at the lower temperature.

Experiments performed on spore germination in the same organism showed that a temperature range of 20–44° C. permitted germination of the five strains tested. Above a temperature of 39° C. and below 30° C. germination time lengthened considerably.

Experiments performed on the influence of humidity on spore germination, both at 26 and 37° C. showed that no germination could take place at a relative humidity of 80% or below. Germination time was considerably prolonged when the humidity value fell below 100%.

This work formed part of a thesis for the degree of M.D. in Bristol University. I wish to thank Dr G. S. Wilson for his advice in preparing this paper.

Single and multi-jet liquid impingers and membrane-filters were found to be efficient sampling devices for aerosols generated from suspensions of the virus of foot-and-mouth disease (FMD). Concentration of the aerosol samples with an adsorbent, Attaclay, facilitated the detection of small amounts of virus. Sodium dodecyl sulphate could be used for elution as the virus of FMD is resistant to this anionic detergent.

The penetration of these aerosols through various air filtration media was determined using impinger samplers. A glass fibre paper was found to be the most efficient with a virus penetration of less than 0·001%.

We wish to thank Dr D. W. Henderson and Dr H. M. Darlow, of M.R.E. and Mr R. G. Dorman of C.D.E. Porton for valuable prefatory discussions and advice. We are also indebted to Miss S. Cartwright for inoculating mice, to Miss M. Shoobridge and Mr S. Taylor for technical assistance, and to Vokes Ltd. for supplying materials.

Immunization experiments in guinea-pigs with high egg passage Flury virus gave the following results:

1. Intracerebral inoculation of the vaccine protects against subsequent intracerebral or peripheral challenge with fixed virus.

2. Intramuscular administration results in the development of a good immunity status to peripheral challenge.

3. Intramuscular or intravenous inoculation of the vaccine fails to protect against intracerebral challenge.

Various explanations have been proposed to account for the increased incidence of moniliasis as a result of ‘broad spectrum’ antibiotic therapy, but it is generally believed that the suppression of the bacterial flora enables fungi to proliferate without competition. The pathogenicity of Candida albicans is, in any event, associated with the numbers of cells present and we have undertaken tests in vitro in an attempt to contribute more evidence on the mechanism of their proliferation in the presence of tetracycline-treated bacterial flora. The results are presented here as a preliminary communication.

A dose of 5 × 109S. typhi (strain Ty2) and S. paratyphi B (strain HB3) administered to white mice orally, caused an infection of various organs of the mouse, namely: spleen, liver, kidneys, lungs, gall-bladder, mesenteric lymph nodes, jejunum and of the blood stream. The percentage of infected spleens was higher than that of other organs.

With the above-mentioned dose of Ty2, the infection of the liver was found to be five times higher than that of the gall-bladder. The infection of the latter lasted for about 8–9 days, whereas that of the liver for 18–19 days.

The bacteria appeared in the various organs: liver, spleen, kidneys, and lungs in about 2 min. after oral infection, and in the blood stream after about 20 sec.

The speed and intensity of the invasion of micro-organisms from the gastrointestinal tract to the blood stream depends neither upon the volume of the dose (0·5 ml. or 0·1 ml.), nor upon the content of the mouse's stomach.

The M.I.D. for the blood stream of the mouse by oral infection is about 5 × 105 bacteria for Ty2, HB3 and AH6.

In blood samples collected by cardiac puncture 2–3 min. after oral infection, an average of 5 × 102 micro-organisms/ml. blood was found. The count of bacteria increased and reached its peak after 20 min. (about 3 × 103 bacteria). After that period, the number decreased: 1·6 × 103 after 1 hr., 8 × 102 after 2 hr. and 1·7 × 102 after 3 hr. In spite of this considerable decrease, the blood did not became sterile until after several days.

In the infected spleen (by oral infection) an inverse process was observed: at the beginning only a small number of micro-organisms was found (about 5 × 10/100 mg. spleen), but afterwards the number of bacteria increased (1·2 × 102 after 24 hr. and 5 × 102 after 4 days).

In mice infected by the subcutaneous route with a dose of 5 × 109 bacteria, the number of micro-organisms in the blood stream did not differ significantly from that found after oral infection, whereas infection by intraperitoneal route caused a severe infection. When 5 × 107 micro-organisms (LD50) were administered by intraperitoneal route, 107 bacteria appeared in the blood stream of the moribund mice while in the spleen about 4·1 × 107 were found.

The author wishes to express his appreciation to Prof. I. Gurewitch and to Dr R. Rozansky from the Department of Clinical Microbiology, the Hebrew University-Hadassah Medical School, Jerusalem, for their advice and criticism.

A complement-fixation test for acute poliomyelitis using unheated antigens derived from suckling mouse brain infected with poliovirus Type 1 or Type 2 is described.

The results of tests in 62 patients clinically diagnosed as cases of acute poliomyelitis in a recent epidemic and in 26 controls are recorded.

The CF tests were positive in 100% of 53 cases with poliovirus Type 1 and/or Type 2 in stool. A positive result was obtained in 23 (76%) of 30 cases whose sera were examined in the first 7 days of illness.

Negative tests of the initial serum samples were found in 15 (28·5%) of 53 cases, but all these became positive in titres of 40 or 80 on testing of convalescent serum.

In 31 (69%) of 45 cases whose sera were re-tested between the 3rd and 4th weeks of illness the CF antibody levels rose, reaching titres of 80 or 160 in most instances. Of the remaining 14 cases only one dropped in insignificant degree (from titre 320 to 160) and the 13 stationary results had been positive in titres of 40–160 on initial tests most of which were performed in the 2nd week of illness.

Homotypic CF antibody response without crossing was found in 37 (71%) of 52 cases with Type 1 or Type 2 virus in stool. In the cases of crossing the heterotypic antibody response was either transient, diminishing or stationary in all and in only low titre in most instances.

In 26 control cases there were seven positive CF tests, but one of these was nonspecific, five were in lowest titres, and one case appeared to have had recent poliomyelitis infection.

Heating the antigens did not broaden the reaction. It caused only slight loss of potency except in two cases in which the CF titre increased substantially.

The antigenic preparation described appears to be superior to antigens of other origin in the diagnosis of acute poliomyelitis by complement-fixation tests, as positive tests are recognized earlier in the illness and the titres are higher. Homotypic results were obtained in all cases and no instance of false negative occurred in this series.

I would like to thank the medical staff of the Cape Town City Hospital for Infectious Diseases for the trouble taken in collecting stools and paired sera, and Prof. Kipps for his interest in this work. I am indebted to Miss Karin Larssen for valuable technical assistance.

It was shown in this work that the electrophoretic mobility of the MEF1 strain of poliovirus was increased approximately 25% after interrupted inactivation with formaldehyde. This is in accordance with past observations on the behaviour of formaldehyde-treated proteins. The residual live virus had the same mobility as the ‘killed’ virus, thus indicating that the surfaces of all the virus particles have been affected to the same degree by the inactivating agent.

Gel precipitin tests performed on the different samples obtained by zone electro-phoresis showed, in addition to the main antigen, the presence of a small amount of a second component of slightly higher mobility, and, as it occurs in the main fraction and the sample thereafter it probably has a slightly higher mobility than the main component.

The antigenic properties of the different fractions as shown by the production of antibodies in guinea-pigs correspond with the presence of precipitating antigen in the samples in the region of highest antigen concentration. The faint antigenicities of the fractions obtained from the regions higher up in the zone electrophoresis column are due to small amounts of virus antigen similar to that found in the column on electrophoresis of untreated MEF1 virus, bearing in mind that the inactivated virus was concentrated approximately 2000 times for zone electrophoresis.

Polluted water from a stream caused three cases of infection with Salm. paratyphi B belonging to two phage-types. These phage-types and two others were isolated from the water.

One woman, aged 81, with a history of enteric fever 20 years previously was discovered to be a prolific carrier of at least three of these phage-types, and to be the source of the pollution.

We are greatly indebted to Mr G. K. Anderson, County Sanitary Inspector, Midlothian, and to his staff for the large volume of field work they carried out; to Dr E. S. Anderson and Dr A. Bernstein, Central Enteric Reference Laboratory and Bureau, Colindale, London for the phage-typing on which this survey is based; and to Mr James Sutherland, F.I.M.L.T., for his enthusiastic technical assistance.

1. By inoculating conjunctival scrapings into the chick embryo yolk-sac trachoma virus was isolated from twenty-four of thirty patients, whereas inclusions were demonstrated in only eighteen of them.

2. Bacterial contamination of the chick embryos can be greatly reduced by treating the scrapings with streptomycin 20,000 μg./ml.

3. In twenty-four of twenty-five isolations, virus was first detected in embryos inoculated directly with conjunctival material.

4. The viruses usually kill the chick embryos, which may show haemorrhagic lesions, but it is not unusual for embryos to survive a large inoculum.

5. The viruses isolated all appear in the chick embryo yolk sac as elementary bodies similar to other members of the psittacosis-lymphogranuloma group. Of the twenty-four strains, two also induced the formation of compact aggregates of elementary bodies containing a carbohydrate, probably glycogen.

6. It is anticipated that this method of isolation will be useful in studying the virology and epidemiology of trachoma, in assessing the influence of treatment, and as a test of cure.

We are deeply indebted to Dr I. A. McGregor for all the facilities made available to us at the M.R.C. Laboratories in the Gambia. Our thanks are also due to the medical officers and sisters at the research ward for their kindness and unfailing co-operation in the care of those trachomatous patients who had to be hospitalized. The excellent technical assistance of Miss A. M. Fischer is gratefully acknowledged.

A new dulcitol–salicin–sucrose–iron–urea medium is described which, on the basis of a single-tube culture, tests the fermentation of three sugars, H2S production and urea hydrolysis.

On this medium 91% of 228 Salmonellae gave easily recognized characteristic reactions, simulated only by some strains of the Bethesda-Ballerup group. Proteus strains and most paracolon strains gave equally characteristic pictures.

The reactions of some other Enterobacteria are described. It is suggested that the medium is good enough to deserve extended trial in routine work.

A carrier has been found of at least three phage types of Salmonella paratyphi B. Investigation has shown that phage-mediated conversion could account for this multiplicity of types according to the following scheme.

It is not possible, however, to exclude the actual occurrence of repeated infection with different phage types.

I am grateful to Dr E. S. Anderson of the Central Enteric Reference Laboratory, Colindale, for his interest and advice during the course of this investigation, and to Dr Helen A. Wright of the University of Edinburgh who provided the many cultures and so introduced me to this problem.

1. Pathogenic bacteria are removed from air by relatively simple filters having an efficiency of 99·9% for 5-μ particles.

2. To prevent ingress of contaminated air from other parts of the hospital an operating-room should be pressurized by a flow of filtered air. An input flow of 1200 cu. ft. (34 cu. m.) per min. will ensure this if mechanical exhaust fans are replaced by pressure-relief valves.

3. Other rooms of the suite should be ventilated in this way with flow rates depending on door area, and with pressure gradients adjusted by relief-valve loadings to cause air flow from dirtier to cleaner zones.

4. All openings between the rooms of an operating-suite should be fitted with doors to prevent airborne cross-contamination and high-level loss of ventilating air.

5. The relative merits of turbulent and downward-displacement ventilation have been studied and are discussed. The choice between them depends on the sites from which pathogenic bacteria are mainly disseminated during operations; these are not yet fully understood but the available evidence favours the displacement system. Methods for producing turbulent and displacement ventilation are described.

6. Discomfort from overheating of operating-rooms is largely due to faulty design and siting of sterilizing equipment. In Britain, if these faults are avoided, ventilation without cooling equipment can give comfortable conditions except on very rare occasions.

1. Faeces from rural, outwardly healthy Bantu school children from a Native Reserve region have been bacteriologically examined seven times at regular intervals over a period of one year.

2. Of 75 children, 44·0% experienced at least one salmonella infection and 29·3% one shigella infection; considering the infections together, salmonellae or shigellae were recovered from 72·0% of the individuals. The observations are discussed and the view is expressed that over a period of one year, practically all children experience one, and many of them several, attacks by these pathogens.

3. In the majority of cases, the infections appeared to be subclinical and of short duration. There was no evidence proving the existence of chronic carriers.

4. Salmonelloses showed a seasonal dependence with the highest incidence in early summer, i.e. December. Shigelloses was distributed evenly throughout the year.

5. The recovered salmonellae were of twenty types. Salm. typhi, Salm. paratyphi A, B and C were not isolated. Members of all shigella groups were encountered.

6. All the recovered strains from the Tlaseng area were sensitive to streptomycin, chioromycetin, terramycin, achromycin and neomycin, and most of them to aureomycin and erythromycin. In comparison, salmonella strains isolated from the Johannesburg urban area showed signs of increased resistance to the same antibiotics.

7. Water was probably implicated in the conveyance of the infections.

We should like to express our thanks to the Director of this Institute, Prof. E. H. Cluver, for permission to publish this paper; to Mr R. G. Robinson for assessing the micro-organisms' sensitivity/resistance to antibiotics; to Dr P. A. Christensen for statistical advice; and to Dr A. R. P. Walker (of the Human Biochemistry Research Unit, Council for Scientific and Industrial Research), who incurred the major portion of the expenses of transport by permitting us to accompany him during his investigation on bilharziasis and nutritional state in the Bantu school children.

Measurements to assess the performance of the ventilating system have been carried out in a series of twenty six operating theatres using nitrous oxide as a gaseous tracer to simulate the movement of airborne bacterial contamination. In order to prevent local clouds of contamination derived from the activities of the operating-team persisting in the neighbourhood of the operating-site it is desirable that there should be appreciable air movement in the centre of the room. Downward displacement, ‘piston type’, ventilation systems may be an exception to this but the indications for their use are not clear. Other points discussed include the necessity for the limitation of the volume of mechanical exhaust if adequate positive pressure is to be maintained in the theatre and the need to control heat dissipation from sterilizers, autoclaves and other sources.

A non-explosive mixture of ethylene oxide and Arcton 12 is described. This mixture is now available commercially in this country.

A sterilizing chamber and related equipment for use with this gas and a sterilizing technique are also described.

Some applications of ethylene oxide sterilization are briefly discussed.

A field experiment was conducted from July 1958 to February 1959 in the San Francisco Bay region to test the effect on flea populations of prolonged exposure of the native meadow vole, Microtus californicus, to 5% DDT powder in insecticide-bait-boxes. Previous studies had shown that two important vectors of sylvatic plague, Malaraeus telchinum and Hystrichopsylla linsdalei, could be controlled both on their hosts and in the host nests by this method. However, the question of possible resistance was postulated since the control of H. linsdalei on its host was inferior to control of other flea species.

Bait-boxes with DDT powder were maintained in the field for 6 months. Records of flea populations on Microtus during this period showed equally good control of all flea species. Observations on vole nests at termination of the study revealed deposits of DDT and good control of all flea species in them. During the period of study, the evidence suggested that none of the flea species involved exhibited DDT resistance.

The major portion of the field routine was done by Messrs A. R. Kinney and R. L. Martin. Flea identifications were made by Mr H. E. Stark. Analyses of nests and rodents for DDT were performed by Mr C. Cueto, Technical Development Laboratories, CDC, Savannah, Georgia.

The serological properties of the two strains of Trichomonas foetus (strain B and strain M) isolated from uterine infections in cows in Belfast and in northern England, respectively, and two strains isolated from the caecum of pigs were in vestigated with rabbit antisera to each of the four strains by direct agglutination and by precipitation of a saline extract of the dired organisms in tubes and in gel-diffusion plates.

There was considerable cross-reaction between the four strains. By the use of predominantly polysaccharide fractions from strains B and M and the predominantly protein fraction TF 4 from stratin B, the polysaccharide complex of lines could be distinguished from the protein lines.

The major protein antigens were common to all four strains, but the major polysaccharide antigens were only partially shared.

The cow strains B and M were readily distinguished from each other but had a variable though slight serological relationship. The two pig strains were very closely related to each other but were not identical. Cross-reaction revealed a relation between the cow and the pig strains. The cow strain B was more closely related than the M strain to the pig strains.

The serological distinctions do not justify the separation of the pig and the cow strains of T. foetus into two species.

I wish to thank Prof. W. T. J. Morgan, F.R.S. for preparations of diethylene glycol extracts and kind advice and Dr Winifred Watkins for ethanol fractionation of antigen and for enzyme preparations. I am also grateful to Dr A. E. Pierce of the Institute of Animal Physiology at Babraham for the isolation of clones from the S2 strain.