No evidence was found to support the view that coliphages as such play an important role in reducing the number of E. coli in sewage water when it is treated in oxidation ponds. Coliphages seem to have greater powers of survival than E. coli in oxidation ponds.

Test baiting has shown that rats underground are distributed in a series of discrete colonies, each one in close association with a group of men. Interplay between colonies is rare in large mines although evident in small ones. In the absence of stables the rats must depend almost entirely on miners’ food scraps.

These observations were made during work carried out under a research grant from Safety in Mines Research Establishment, Ministry of Power, and I am indebted to the Ministry for permission to publish the results. I should like also to express my gratitude to Dr E. T. B. Francis who supervised the work, Mr. C. S. W. Grice of the Ministry of Power and Miss B. B. Jones of the Ministry of Agriculture, Fisheries and Food for their help in various aspects of the work.

Investigations carried out to ascertain the ability of various strains of Bordetella pertussis and B. parapertussis to produce agglutinins have shown that the agglutinin response is considerably greater with B. parapertussis.

Children inoculated with a combined vaccine in which the parapertussis element contained B. parapertussis in only one-twelfth of the concentration of B. pertussis in the pertussis element showed agglutinins in their sera in titres well above 1:300 for both organisms. There were no cross-reactions and the serological responses were specific throughout. The vaccine used was the standard diphtheria-tetanus-pertussis (DTP) prophylactic to which had been added a vaccine prepared from recently isolated strains of B. parapertussis.

Agglutinin titres of both whooping cough components with the combined vaccine were somewhat lower in mice than was the case when monovalent vaccines were used, but they were considered to be satisfactory.

It is suggested that the agglutination production test in mice could be used for the assessment of protective power of B. parapertussis vaccines against infection.

I wish to thank Dr Ikić, director of the Institute of Immunology, Zagreb, who enabled me to perform all these examinations, further to Dr B. Mravunac and Dr Z. Radanov for having carried out vaccination in children and for the clinical examination of post vaccination reactions.

In experiments on survival of microbial aerosols, viability should be measured as early as possible in the life of the cloud, but this must not be used as the datum to which subsequent viabilities are related; all viabilities must be related to that of the material from which the cloud was generated. Mathematical models may be used as a convenience in expressing results, but observed viabilities must also be quoted.

I am indebted to the many colleagues whose work over the years has consolidated this approach, and particularly to Mr G. J. Harper whose experiments and thinking form the backbone of my discussion.

Among 309 cultures of Salmonella typhimurium, phage-type 27, fifteen, isolated from eight patients, were found to be resistant to the three drugs, streptomycin, tetracycline and sulphathiazole. This triple resistance could be transferred by growth in mixed broth culture to a strain of Skigella sonnei and back again to sensitive cultures of S. typhimurium. In whole cultures the resistance was stable, but spontaneous loss could be demonstrated in a small proportion of the organisms in such cultures. No elimination of resistance was demonstrated after treatment with acriflavine. Resemblances to the multiple drug resistance in enteric bacteria reported from Japan are noted.

The author is most grateful to Dr E. S. Anderson, Director of the Enteric Reference Laboratory, Colindale, N.W. 9, for phage-typing the cultures of Salmonella typhimurium, and to Dr K. Patricia Carpenter, Director of the Dysentery Reference Laboratory, for supplying cultures of Sh. sonnei which were essential for the experimental work.

An LD 50 dose S. typhi (strain Ty was injected intraperitoneally into white mice, and the number of micro-organisms was determined after 1 min. and subsequently, in the blood, and after 3 hr. and subsequently, in the spleen and liver. The course of the infection was observed for 34 days.

The number of bacteria gradually increased in the blood stream during the first post-infective hours. It was possible to differentiate by external signs from the sixth post-infective hour and onwards between sick animals and recovering ones. This difference corresponded to the bacterial concentration in the blood stream and internal organs.

In the sick animals the increase in the bacterial count of blood stream continued till 9–10 hr. after the inoculation, when it reached its first peak. During the following 8 hr. a clear decrease in the blood count was evident which was attributed to the ‘clearing mechanism’ of the blood. The onset of a new bacterial rise in the blood stream appeared 18 hr. after the inoculation and reached a second peak during the further 6–8 hr.

In contrast to the observed decrease in the bacterial blood count, a steady and gradual increase in the count of the internal organs was seen; it was steadily progressive till the exitus.

In the recovering animals a steady decrease in the bacterial blood count was recorded. Between 24 and 26 hr. the number of micro-organisms in the blood and spleen was found to be 400 times less in the recovering animals than in the moribund ones. From 48 hr. and onwards a slow and gradual decrease in the bacterial count of the surviving animals was observed. The blood became sterile after 25 days. In the spleen and liver of the surviving animals after 48 hr. a new increase in the bacterial count is evident, accompanied by hyperplasia of the spleen. The number of bacteria reached a new peak in the liver on the third day and in the spleen on the fourth day after the infection.

Subsequently a slow and gradual decrease in the bacterial count was observed in both organs, but even after 34 days a small number of persisting bacteria in the spleen and liver tissue may be found.

Coxsackie A 7 virus was isolated from thirty-seven patients during an outbreak in Scotland in 1959. Seven cases were paralytic, one of them fatal. Evidence is presented that Coxsackie A 7 virus caused these paralytic illnesses. The virus was also isolated from a paralytic case in 1956 and from a non-paralytic case in 1961. Serological surveys suggest that it has been active in the community for some years. Specific haemagglutination by Coxsackie A 7 virus was useful for rapid identification of viruses and for measurement of serum antibodies.

I am grateful to Dr A. D. Macrae of the Virus Reference Laboratory, Colindale, London, for prototype Coxsackie A 7 virus; to Dr K. Habel of the National Institutes of Health, Bethesda, U.S.A., for tissue culture-adapted ABIV virus; to Dr J. Wallace of the Blood Transfusion Service, West of Scotland Region, for samples of blood donor sera; to Dr M. Rentsch, Klinik für Kinderkrankheiten, University of Berne, Switzerland, for permission to quote the results of virological tests of his cases; to Miss R. McLelland, F.A.T.A., and to Mr C. McLean F.I.M.L.T., for technical assistance with animal experiments; to Mr H. G. Carson, F.I.M.L.T. and to Mr J. Kerr, A.I.M.L.T., for technical assistance with neutralization tests; and to the many clinical colleagues who provided specimens and information for this study.

In toxoplasma-antibody titrations with the dye-test the tolerances of the parasite Toxoplasma gondii are approximately normally distributed on the logarithm of the serum dilution. Log LD 50 (the logarithm of the titre), the variance of the tolerance distribution and the variances of the sample distributions were estimated by probit analysis of data obtained with titrations of a serum which contained the antibody. An empirical expression was derived relating the fraction of parasites attacked by the antibody to the time of incubation with the antibody. Approximately the same values for the titre and the variance of the distribution were obtained for every incubation period between 20 min. and 4 hr.

The author is indebted to Mr J. C. A. Zaat and to Dr E. F. Drion of the Statistical Division of the Central Organization TNO for helpful discussions and assistance with the statistical calculations. The technical assistance of Miss A. K. Fast is gratefully acknowledged.

Mechanical ventilation at 7 air changes/hr., air disinfection using vaporized hexyl resorcinol or α-hydroxy-α-methyl butyric acid, or irradiation of the air of the room above 8 ft. with ultraviolet radiation when applied to various clerical offices did not result in any observed alteration in the numbers of colds experienced by those working in them.

There were no differences in the numbers of colds experienced by those working in rooms of different sizes. The working population in these rooms varied from, about 20 to more than 80.

We should like to express 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. We should also like to thank the Shell Petroleum Company for the supply of α-hydroxy-α-methyl butyric acid, and the British Thomson Houston Company of Rugby for the loan of the ultraviolet lamp fittings.

Most air disinfectants fail at low relative humidity. Those which remain active appear to be the same compounds which, in other circumstances, can protect living cells against the lethal damage caused by freezing to low temperatures and then thawing. The compounds are either simple glycols, whose respective activities in the two fields are already known, or the weaker Lewis bases, whose air disinfectant properties have not hitherto been investigated. It is suggested that both freezing protection and bactericidal action are due to the special solvent properties of these compounds, which come into play when most of the water usually present has been removed, by freezing or by drying respectively.

I am indebted to Dr 0. M. Lidwell for the loan of air sampling equipment, and to Dr L. N. Owen and to the Shell Chemical Co. for the gift of various diols.

1. An attempt has been made to follow up a consecutive series of seventy-two patients for the presence of Phase 1 C.F. antibody approximately 2 years or more after the acute illness. Fifty-one of the series were tested.

2. Fifteen of the fifty-one patients had detectable amounts of antibody, generally in low titre.

3. The presence of Phase 1 antibody correlated well, in older age-groups, with the duration of convalescence following the acute illness. There was no correlation with the duration of fever.

4. All cases with Phase 1 antibody also had Phase 2 antibody, usually in slightly higher titre. In no case in which Phase 2 antibody was absent or present in low titre was Phase 1 antibody found.

5. It is suggested that the presence of Phase 1 C.F. antibody is an indication of past persistent infection. It cannot necessarily be concluded that it is an indication of present persisting infection.

We wish to thank Dr H. Silverstone, Senior Lecturer in Medical Statistics, University of Queensland, for statistical advice.

For a period of a year the total count, Coli-aerogenes count and turbidity were determined on samples of sewage and effluent taken every week from pilot-scale percolating filters treating domestic sewage. The eight filters contained four types of filtering medium in two sizes.

Numbers of coliform bacteria in the settled sewage were much higher during the warmer months but the total count showed no regular seasonal variation. Treatment in all the filters reduced the numbers of bacteria in both groups by over 95%. In general the reductions in numbers of bacteria and in numbers of all particles in suspension were proportional to the total area of filtering medium with which the sewage came into contact. The probability of discharging more than 10,000 coli-aerogenes per ml. ranged from 10% in the most efficient to 90 % in the least efficient filter.

Experiments in which Serratia indica was added to the influent sewage showed that only 0·4 % were recovered in the settled effluent after 24 hr. Treatment in the filters had little selective action on the different types of coliform bacteria.

This paper is published by permission of the Department of Scientific and Industrial Research.

Biological assays based on a quantitative response in particular (e.g. hormone assays) typically involve the estimation of (log) relative potency (= M) with respect to some standard preparation on the basis of one or more symmetrical determinations usually at equally spaced (log) dose intervals. It is the purpose of this paper to consider further statistical aspects of estimates of M, and especially confidence limits for M based on a combination of separate assays. These are given as examples of tests of linear hypotheses. Two applications are given on the combining of quantitative bio-assay data.

The environment of two contiguous surgical wards was examined over a period of twelve months by means of a slit sampler, settle plates and blanket sweep plates. At the same time, nasal swabs were taken each week from the patients and all cases of sepsis examined bacteriologically.

Phage typing of more than 3600 isolations of Staph. aureus showed that there was one predominant strain in the air, bedding, patients' noses and infected wounds.

There was no relationship between the total number of bacteria in the ward air and the numbers of Staph. aureus.

The recovery of large numbers of Staph. aureus from the air at certain periods was associated with a high contamination rate in the blankets and with an increased incidence of staphylococcal sepsis.

Not all nasal carriers of Staph. aureus contaminated their bedding. There was evidence that some patients became nasal carriers of strains of staphylococci previously isolated from their bedding.

Some evidence was obtained that blankets may play a role in the transmission of staphylococci from patient to patient.

This work was supported by a grant from the Australian National Health and Medical Research Council. Our thanks are due to Mrs Elisabeth Bradshaw for her technical assistance and to Prof. John Loewenthal for his interest and for permission to study his wards.

A planned experiment to gauge the effect of regular blanket disinfection on bacterial, especially staphylococcal, contamination and infection in an acute geriatric unit is described.

With formalinization of blankets, the total colony counts of sweep-plates of blankets in use fell to 43 and 34% of the control period counts on two test wards respectively. The staphylococcal counts fell to 20 and 25% of their previous levels on the two test wards. Ordinary laundering with soap and water did not significantly alter the degree of bacterial contamination of the blankets.

Colony counts of slit-sampler plates taken during bed-making fell, with formalinization, to about one-third and one-half of the control period counts on the two test wards respectively. Staphylococcal colony counts fell to about one-fifth and one-third of their previous levels.

Staphylococcal contamination of patients was measured by the nasal carrier rate, there being very little sepsis on the unit. The gross figures for nasal carriage of Staph. aureus showed little variation attributable to blanket disinfection. The number of patients acquiring a new strain of staphylococcus after admission tended to increase as the investigation continued. This trend was reversed on both test wards coincident with the introduction of blanket disinfection, and it would seem that this measure may have the effect of lessening the chances of the hospitalized patient becoming a nasal carrier of Staph. aureus, but the results are not formally statistically significant.

There appears to be a lack of parallelism between the degree of environmental contamination and cross-infection rates. It is considered unlikely that contaminated blankets may cause infection except in two special circumstances: (a) in the wards, by direct contact with the patient's skin or in-dwelling catheter, and (b) in the operating theatre suite, either directly by contact with the patient's skin, or indirectly via the air.

I am grateful to Professor R. E. O. Williams for the loan of a slit-sampler; to Dr Martin Hynes for advice and encouragement; to Mr M. P. Curwen for the analysis of statistical significance; to Mr F. Seage for much technical help; to Dr J. Sharkey, geriatrician; to Sister J. Cleary and the nurses of the geriatric unit; and to Mrs H. M. Page and Miss M. Hanworth for secretarial assistance.

The work was supported by a grant from the free funds of the Northern Group Hospital Management Committee.