The death-rates have been determined of strains of Staphylococcus arueus dried on lint and other textiles and stored at room temperature in the dark.

Two categories of strains were distinguished, those that showed no significant loss of viability after 15 days storage and those whose death commenced from the beginning of the drying period.

There was some degree of correlation between survival for 15 days and implication in epidemics of hospital infection.

Two strains of myxoma virus isolated from naturally infected wild rabbits have been described. They showed similar clinical symptoms, the most striking feature being the flat primary lesion. In fifteen rabbits the Brecon strain gave a mean survival time of 26·3 days with a range of 17–47 days, and one recovery. Of 5 rabbits inoculated with the Durham strain three became clinically infected and two remained healthy, the latter were found to be fully susceptible on reinoculation. The three infected rabbits all recovered.

The field strains of virus were compared with known myxoma strains—Glenfield, Cornwall, A & H and KM 13. The first three are fully virulent, while KM 13 is attenuated and characteristic of the field strains now being isolated in Australia. The similarity between the clinical symptoms and mean survival time of KM 13 and the Brecon strain was noted.

It is suggested that the two field strains described in this paper may be the first indication that the evolution of myxoma virus in this country is similar to the evolution of the virus in Australia.

One of us (P.J.C.) is greatly indebted to the Director for allowing this work to be carried out at M.R.E. whilst laboratory accomodation was being prepared at Worplesdon. We are grateful for the help and encouragement of Dr J. C. N. Westwood of M.R.E. To Prof. F. Fenner who gave us the Australian strains of myxoma virus and to Mr J. H. Darbyshire of the Central Veterinary Laboratory who supplied the tissues from which the two attenuated strains were isolated, we extend our thanks.

1. The virulence of S. typhi strains Ty 2 and O 901 injected intra-abdominally into white mice was examined. The lethal dose of strain Ty 2 grown in vivo was lower than that of the corresponding culture grown in vitro, while the lethal doses of the in vivo- and in vitro-grown strain O 901 were almost identical.

2. The extracts of infected livers and spleens were not toxic, but acted as infection-promoting substances. Extracts from normal organs exerted similar but weaker effects. Glycogen together with sodium taurocholate were powerful infection-promoting substances. The highest increases of virulence were observed when spleen-grown bacteria together with extracts from infected organs were injected.

3. Extracts from infected spleen, pancreas, stomach and lower intestines, spleen-grown S. typhi, other spleen-grown Enterobacteriaceae, and bacteria taken from the intestines of normal and diseased animals, exhibited high DPN-ase activity.

4. The extracts from infected organs contained soluble bacterial antigens. Their presence was demonstrated by tube precipitation, agar gel precipitation and antibody production after immunization of rabbits with extracts from infected organs with addition of adjuvants.

5. Spleen-grown bacteria were able to absorb host antigens and to produce, in addition to the known agglutinating antibodies, anti-spleen precipitins.

Values are given for the median equivalent diameters and for the inter-quartile range, of airborne particles carrying a variety of micro-organisms.

Organisms associated with human disease or carriage were usually found on particles in the range 4–20 μ equivalent diameter.

Many fungi appeared to be present in the air as single spores.

We should like to acknowledge the collaboration of Dr Yvonne M. Clayton in the collection and examination of the fungal samples.

1. The respiration of bacteria normally present in thin slices of skin taken from pig's trotters has been measured in Cruickshank micro-respirometers.

2. These respirometers were originally designed to measure skin respiration in a medium including antibiotic to inhibit bacterial activity. The skin and medium was contained in a glass well. Using pig skin in a similar system without antibiotic, bacteria left the skin to grow independently in the medium. To overcome this difficulty polythene cups were made with an inner compartment for the tissue without medium and an outer compartment for water which provided a humid atmosphere.

3. Using polythene cups without medium the skin respiration was neglible, bacterial respiration was measured in situ and correlation between oxygen uptake and bacterial counts was obtained. This correlation was valid within experiments only.

4. Respiration of skin samples taken from adjacent areas of pig's trotter was similar, but there was large variation between samples from different trotters.

5. Respiration due to bacteria increased rapidly for 6–9 hr., followed by a slower increase. This may be due to a limited food supply or to a state of equilibrium.

A comparison was made of three methods for the removal of dust from a hospital ward floor.

The viable bacterial counts of impression plates from the floor showed a mean reduction of 51 % after cleaning with fresh oiled mops, 40 % after cleaning with a tank model vacuum cleaner and 5 % after sweeping with a broom.

The mean proportions of airborne bacteria, compared with viable counts before cleaning, were 192 % during and 194 % after cleaning with a broom; 128 % during and 103 % after cleaning with an oiled mop, and 82 % during and 48 % after cleaning with a vacuum cleaner. On repeated use the oiled mop dispersed almost as much dust as a broom, but this settled rapidly from the air.

The effects of these cleaning methods on counts of presumptive Staph. aureus on the floor and in the air were similar to those found in counts of total organisms.

We wish to thank Mr M. D. Wilkins for valuable assistance, the Domestic Superintendent and staff and the nursing staff for their co-operation, and Messrs Leeming Brothers Limited for supplies of ‘Kex’ mops and equipment.

1. A food store infested with house-mice (Mus musculus L.) was treated with boxes containing 1% warfarin dust surrounding a solution containing 0·005% sodium warfarin and 5% sugar. The boxes were placed around the bases of the stacks, in the air-lanes between stacks and at eaves height at the gable ends of the store.

2. Census baitings with counted numbers of wheat grains showed that the mouse population was reduced by about 95% after 5 weeks of the poison treatment and by almost 100% after 20 weeks.

3. The field trial suggests that the combined use of rodenticidal dust and poison solution baiting points is a useful additional method of controlling M. musculus and is most effectively employed against mice living in warm, dry environments.

Thanks are expressed to our colleague Mr E. W. Powell, who co-operated in this work.

The sensitivity of in vitro- and in vivo-grown strains of S. typhi was determined not only by their antigenic structure, but also by their growth phase. An increased vulnerability to the antibody-complement system has been found in cells during the lag and the early exponential phase, while non-multiplying cells devoid of any nutrient were almost invulnerable to the antibody-complement system. Extracts of organs from normal and infected animals may promote the growth of S. typhi, and, therefore, increase its vulnerability to the antibody-complement system. The majority of extracts from organs of mice infected with strain Ty 2 inhibited markedly the bactericidal action of serum on this strain and, to a lower extent, on strain O 901.

Optimum forms of moving averages are derived for the estimation of LD 50 in the situation where no assumptions are made about the form of the dose-response distribution. The theory is also applied to uses of moving averages based on the angular transformation of the percentage response. A numerical example illustrates this application, and its results are compared with the corresponding probit and logit estimates.

A milk-borne outbreak of Salm. heidelberg infection due to unpasteurized milk from tuberculin-tested cows is described. There were 77 cases and 46 symptomless excreters of the organism. There were no deaths. The infection was traced to a cow with a symptomless salmonella mastitis and the origin of the organism was thought to be contaminated animal feeding stuffs.

We are indebted to Dr Joan Taylor, Colindale, for the identification of the salmonella serotypes; to Dr H. R. Cayton, Bristol, for assistance in the examination of animal feeding stuffs; to Dr E. N. Davey, Gloucester, for permission to include his results of milk examinations; and to the following public health inspectors who assisted in the field investigation—Mr A. Furniss, Mr G. Baillie and Mr M. V. Smith, Cirencester Urban District; Mr H. Wheeler and Mr R. Hayward, Cirencester Rural District; Mr P. Hayter, Bristol; and Mr D. S. Smith, Trowbridge.

Semi-purified elementary body suspensions, prepared or preserved in three different ways, have been compared as to their species-specific CF-activity as revealed both against Bedsonia-negative sera of trachoma patients (of known stage), and against psittacosis-positive sera of men and animals.

Two of these antigens (the SPG-formol-antigen and the SPG-glycerol-antigen) revealed a positive species-specific CF-reaction in 61–100% of the Bedsonia-negative serum samples obtained from trachoma patients, with titres ranging from 1/2 to 1/128.

Various parameters (as percentage proportion of positive reactors, CF-titre, duration of the disease, clinical stage of trachoma, etc.) have been analysed for correlation, and the results obtained with the different antigens are compared with one another.

The most useful of the three antigens, showing a good species-specific CF-reactivity and no anticomplementary activity, was found to be the ‘SPG-formolantigen’. However because of its group-specific reactivity, for detection of species-specific trachoma antibodies in routine, it could be used only with Bedsonia-negative samples of sera. Because of the thermolability of its species-specific component, the SPG-formol-antigen proved to be liable to deterioration.

We have pleasure in thanking to Dr F. B. Gordon for help in procuring bibliography and the viable strain of trachoma virus. We wish to express our thanks also to Mrs R. Šepetavic and to Miss M. Krečo for their technical assistance.

1. A technique for measuring the antibacterial activity of germicides on pig skin in micro-respirometers has been described.

2. Using this technique the activity of the following germicides has been estimated: 2,2-dihydroxy, 3,5,6,3′,5′,6′-hexachlorodiphenylmethane, 3,4,4′-trichlorocarbanilide, 3,5,4′tribromosalicylanilide, 2,2′-thiobis (4,6-dichlorophenol), Irgasan CF3*, 3,5,3′,4′-tetrachlorosalicylanilide.

3. The results obtained have been discussed in relation to hand-washing test results (Hurst et al. 1960).

Data are presented which amplify previously published work on the additive antiseptic effect of Roccal as follows:

1. Smallpox calf pulp derived from animals whose skin is treated with Roccal during the viral incubation period apparently adsorbs sufficient concentration of the reagent to produce a measurable bactericidal effect.

2. The adsorbed antiseptic induces a significant additive effect well beyond that of the action of stated concentrations of phenol and ether alone, as evidenced by the fact that this combination produces rapid diminution of the bacterial flora in highly contaminated calf lymph.

3. Vaccines containing the combination of antiseptics described caused no dermatitis or other untoward reactions in rabbits and humans and their viral potency was not affected.

4. The use of the antiseptic reagents makes possible the rapid preparation of smallpox vaccines for human use. Acceptable bacterial flora levels are obtained usually within 24 hr. of preparation.

The author is indebted to Magda Gagliardi, Marie Mazzochi and Izola Prunty for their assistance in the performance of the many bacterial counts, toxicity tests and other technical work.

1. Experimental typhoid vaccines, treated by 1% formalin and 0·02% chrome alum (KCr(SO4)2.12H2O), as suggested by Japanese workers were prepared and tested by the usual in vitro and in vivo tests.

2. Agglutination tests, antibody production in rabbits, active and passive mouse protection tests confirm the stability of the Vi antigen of the vaccine, if properly stored, and the good protection afforded to laboratory animals in both the active and passive mouse protection tests.

3. It is suggested that only a full-scale field trial in a typhoid endemic area can give the answer as to the real efficacy and/or superiority of the chrome vaccine over other typhoid vaccines.

Twenty-eight adult volunteers were inoculated intranasally with para influenza 2 virus and eight developed illnesses; twenty-eight volunteers were given flanks' saline and one became ill. The illnesses occurred in volunteers given between 2 × 104 and 2 × 106 TCD 50 of virus. The most prominent symptoms were sore throat, nasal stuffiness and coryza; four of the eight volunteers had sufficient coryza to be regarded as having mild colds. Although only eight volunteers had clinical evidence of infection, twenty-four had laboratory evidence of infection as judged by virus re-isolation or antibody response. Neutralization, haemagglutination-inhibition and complement-fixation tests on paired sera showed that sixteen individuals had a fourfold or greater antibody response by one or more tests including five of the eight volunteers who were ill. Twenty volunteers, including seven who were ill, had reciprocal neutralizing antibody titres of eight or more before inoculation of virus so it seems that the illnesses were due to re infection in the presence of antibody. Evidence is presented which suggests that although illnesses occurred in the presence of antibody they were due to the para influenza 2 virus and not some other agent in the inoculum. The results of these experiments seem to fulifi the third of Koch's postulates for para-influenza 2 virus as a cause of respiratory disease in adults.

We wish to thank Dr P. A. J. Tyrreil for his advice during the course of this work and in the preparation of the manuscript. We also thank Dr K. V. Shah (The Johns Hopkins University, Baltimore, U.S.A.) for help in some of the early experiments and Miss B. Ridgwell for valuable technical assistance.

The nature of the ringworm flora of Birmingham over 17 years was investigated and tabulated by species, geographical distribution and concentration.

The rise and fall of such outbreaks as occurred has been charted.

The persistent preponderance of zoophilic species, in spite of the recent decline in over-all incidence, has been demonstrated.

1. That the number of toxoplasma organisms in the dye test affects the titre of the test serum is confirmed.

2. Evidence is presented that the toxoplasma cytoplasm-modifying substance present in the peritoneal exudate of mice injected with large numbers of toxoplasma organisms more than 3 days previously is probably antibody.

3. Various chemicals including detergents were capable, in high concentration, of producing a false positive dye test.

4. Evidence is presented that there are two complement-fixing antigens, one with a low sensitivity and a low density, the other with a high density and a sensitivity close to that of the dye test. Heating the heavy antigen to 56° C. for 1 hr. removes most of its anticomplementary effects without altering its complement-fixing activity.

5. The light antigen mentioned above was capable of producing a precipitation line in agar double-diffusion experiments, but did not inhibit dye-test antibody when mixed with an equal volume of antiserum in the dye test.

6. The heavy antigen inhibited dye test antibody when mixed with an equal volume of antiserum in the dye test.

An autoradiographic method to trace the dissemination in water of radioactive bacteria is reported. It has been tested for accuracy and sensitivity and a field trial has been performed. The method appears suitable for practical use and permits the study of environmental effects on the spreading bacteria.

An outbreak of food poisoning caused by Salmonella typhi-murium, phage-type 12, in which 122 persons were infected, is described.

Isolation of the same organism from the drains of two large abattoirs, from pig faeces at one abattoir, and from the floor drains of fifteen out of fifty-four butchers' shops and bakehouses examined, together with records showing that this organism had been recently isolated from pigs and cattle in the neighbourhood, suggested that meat was the vehicle of infection.

We are indebted to Dr E. S. Anderson and the staff of the Central Enteric Reference Laboratory and Bureau, Colindale, for phage-typing the cultures of S. typhi-murium.

We should like to thank Prof. Scott Thomson for his encouragement and advice in the preparation of this paper and Mr J. Morgan and Mr B. M. Jones of the Public Health Laboratory, Cardiff, for their technical assistance.

A survey embracing all cattle vaccinated in Kenya with lapinized rinderpest virus between 1 January 1954 and 30 June 1956 revealed an over-all incidence rate of adverse reactions of 0·09 %. The real incidence rate approached 3 per 1000 cattle because only susceptible cattle were at risk. Four types of reactions were recognized. The most serious, because of the attendant high case mortality rate, was aggravation of latent protozoal infections. The commonest reactions were rinderpest-like without serious sequelae except in pregnant cows, many of which aborted. Transient spermatolysis was observed in six vaccinated bulls and anaphylaxis was recorded rarely.

The incidence of rinderpest-like reactions was primarily influenced by the innate resistance of the cattle. The viral strains recovered from reacting cattle were genotypically unchanged.

I am grateful to Mr J. W. Macaulay, B.Sc., F.R.C.V.S., D.V.S.M., Chief Veterinary Research Officer, Kabete, Kenya, for his encouragement, advice and counsel. This paper is published by permission of the Director of Veterinary Services, Kenya.