A Factor Analysis has been made of the co-variation between the mortality rates from cancer of ten male body sites and of eight female body sites, in thirty large towns in England and Wales from 1921 to 1930. The method of analysis adopted is Hotelling's method of Principal Components.

Four male and four female Factors are obtained, which together account for approximately three-quarters of the total variance.

A Factor is found to be associated with cancer of the larynx, oesophagus, stomach and tongue in men and with cancer of the stomach and negatively with cancer of the breast and ovaries in women. In both sexes, the Factors are associated with an index of adverse social conditions.

Another Factor is found to be associated with cancer of the rectum and bladder in men and this Factor is associated with good social conditions.

A special Factor associated with cancer of the colon is unrelated to the mortality from cancer of other sites, save that in women it is negatively associated with cancer of the rectum.

A Factor for cancer of the lung in men is unrelated to cancer of the larynx, and is inversely related to cancer of the tongue.

We are most grateful to Dr Percy Stocks, late of the General Register Office, and to Dr W. P. D. Logan, Chief Medical Statistician of the General Register Office, for their help in providing us with the relevant basic material; and to Miss E. M. Hines, Miss A. H. Huntley and Miss M. Rogers for assistance in the calculations.

1. The susceptibility of the Henzerling strain of R. burneti to phenol, formalin, ‘Lysol’, ‘Dettol’, ‘Cetavlon’, and ether was investigated.

2. Preliminary investigation of techniques showed that, with one passage, guinea-pigs, mice and fertile hens' eggs were of similar susceptibility to small numbers of viable organisms, but the greater feasibility of sub-passage from the eggs increased the sensitivity and made this the technique of choice.

3. Under the condition of testing, ‘Lysol’ was the most active disinfectant. Treatment of the rickettsial seed with 1% ‘Lysol’ at 37° C. was in fact the only effective method for destroying completely the infectivity in 3 hr.

4. Some living organisms survived contact for 3 hr. with each of the other chemicals in concentrations up to 1%. A rough quantitative comparison of these agents was made.

5. The failure of 1% phenol to destroy R. burneti raised the possibility of this organism surviving as a contaminant in vaccine lymph. This problem was investigated and is discussed.

1. The aerobic faecal flora of healthy infants was found to comprise coliform organisms, streptococci and occasionally yeast and spore-bearing organisms. Coliforms and streptococci occurred more frequently in the stools of artificially fed than of breast-fed infants in whom micrococci occurred more frequently.

2. Of 191 strains of streptococci isolated from healthy infants' faeces and tested serologically, 8·90% fell into Lancefield's group D. Of the stools examined 93·7% contained group D streptococci.

3. A comparison of the group D streptococcal flora of the faeces of healthy infants and of infants associated with outbreaks of neonatal diarrhoea showed that the percentage of these organisms found in the latter was much higher than in healthy infants, the modes being 90–100 and 10–20% respectively.

4. The same species and varieties of group D streptococci were found in both healthy and sick infants. Streptococcus faecalis was the predominant species.

5. Serological typing of strains of group D streptococci from healthy infants from three maternity hospitals showed that particular types tended to predominate in each maternity unit and to recur over a considerable period of time, in each instance suggesting a reservoir of dissemination to successive infants as they arrived. With artificial feeding an increased diversity of serological types was found.

6. Serological types of group D streptococci isolated from sick infants during ten outbreaks of neonatal diarrhoea showed a considèerable variety. Apart from one outbreak (A), where type I was isolated with great frequency, one or two types tended to predominate in each outbreak, but not to any great extent, and not always the same one.

7. Types 1, 5, 8 and 9 were the types of group D streptococci most commonly found in both healthy and sick infants, but type 1 occurred in a significantly greater percentage of sick than of healthy infants.

8. Tyrosine decarboxylase activities of group D streptococci were usually associated with serological types. The distribution of activities of these organisms was similar in healthy and sick infants.

9. The possibility that neonatal diarrhoea may consist of more than one syndrome is discussed, and several of the outbreaks described here are classified into two groups.

The author wishes to thank Dr A. T. R. Mattick for his constant interest in this work; Dr P. M. P. Shattock for advice and many valuable suggestions; Dr E. F. Gale for advice and material; Dr E. H. Gillespie, Dr A. M. McFarlan and Dr J. Wright for material and clinical data; Dr P. White and Miss Z. D. Hosking for assistance with the statistical computations; Mr L. G. Newland for technical assistance; and to express her appreciation to the Medical Research Council for a grant which enabled this work to be carried out.

A vertical closed circuit wind tunnel is described in which a spinning top sprayer is used as a source of homogeneous particles of pathogenic organisms. The dry particle size may be set at any figure between 1 and 20 µ, and the cloud concentration can be varied over wide limits. Provision is made for humidity control. Animals may be exposed to the cloud, and samples taken from it. The apparatus can be sterilized with flowing steam and formalin.

An appendix giving details of the operational technique is available on request.

Influenza virus elementary bodies can be disintegrated by treatment with ether with the liberation of two types of smaller particle, soluble antigen and red cell agglutinin.

Soluble antigen derived from the elementary body is identical in serological and chemical behaviour with soluble antigen recovered from infected tissues. Its chemical properties are those of a ribonucleoprotein.

The agglutinin is an enzyme, with a protein part carrying the enzymic activity and a combining group with affinity for red cells. The agglutinin does not react in complement-fixation tests by the short fixation technique, but by the use of prolonged fixation or by indirect complement fixation it can be shown to contain a strain specific antigen and also a non-specific antigen. The specific antigen is identical with the specific complement-fixing antigen demonstrable in the intact elementary body, and has combining affinity for red blood cells. The non-specific antigen is probably similar to the protein part of the soluble antigen. No evidence has been found that the agglutinin contains either carbohydrate or nucleic acid.

From the ether used to disintegrate the elementary body a serologically active lipid can be recovered which has properties suggesting that it is derived from the host cell.

The author is greatly indebted to Dr L. Dmochowski for assistance in experiments involving high-speed centrifugation, to Sir Macfarlane Burnet and Dr G. L. Ada for supplies of receptor-destroying enzyme, to Dr E. S. Duthie for crystalline trypsin and chymotrypsin, and to Dr R. R. Porter for a sample of crystalline ribonuclease.

Experiments were carried out in a test chamber of 16,500 cu.ft. capacity with air conditioned at 68° F. and 54% relative humidity. Ventilation was controlled at 3·0 changes/hr., air being recirculated with 0·5 air change/hr. of fresh air admitted. By spraying into the air ducting a concentrated suspension of Chr. prodigiosum a bacterial cloud was introduced into the chamber of over 5 × 106 organisms/cu.ft. of air as l μ particles. Viable organisms were still present in the atmosphere after 6 hr.

When a heat-generated vapour of hexylresorcinol was introduced in a concentration of between 6 and 7 μg/cu.ft. of air the bacterial count dropped from nearly one million organisms per cu.ft. to fewer than ten in 35 min.

When the bacterial suspension was sprayed into the chamber with hexylresorcinol vapour present, the peak concentration reached was about one-quarter and on stopping the spray, the bacterial count fell from one million per cu.ft. to fewer than ten in less than 25 min.

Hexylresorcinol is also effective when the bacterial aerosol is in particles of 6–7 μ in size (approx. 100 bacteria per particle).

Hexylresorcinol has no effect against a bacterial aerosol of B. subtilis spores in 1 μ particles.

No attempt has been made to analyse the mode of action of hexylresorcinol in diminishing the bacterial content of the air. The evidence collected clearly indicates the necessity for further study of this problem.

The labile Vi antigens of Salmonella typhi, Salm. paratyphi A and B, and of Salm. typhi-murium were compared, especially in their response to various chemical and physical agents.

1. Inactivation by dilute acid:

(a) The TVi antigen is the most resistant to acid treatment, the BVi antigen is most readily inactivated and the AVi antigen holds an intermediate position.

(b) Pure Vi sera may be obtained by absorption with acid-treated bacilli, but this technique gives irregular results.

(c) Contrary to Kauffmann's statements, the TVi antigen is not ‘destroyed’ by acid treatment.

2. Inactivation by dilute alkali:

(a) All the Vi antigens so far tested are susceptible to dilute alkali.

(b) Complete inactivation, or extraction, of the Vi antigens is ensured when the yield of alkali-treated bacteria represents no more than 25% of the original bacterial count.

(c) The most reliable method of preparing a pure Vi serum is by absorption of the Vi + O serum with alkali-treated organisms of the same strain as that employed in immunization.

(d) Immunization with alkali-treated bacilli results in the elaboration of pure 0 antisera entirely devoid of Vi antibody.

3. Inactivation by heat (in aqueous suspensions):

(a) The TVi antigen is the one most readily inactivated, the BVi antigen is the one most resistant and the AVi antigen holds an intermediate position.

(b) These differences are only of minor significance, since similar differences have been established for the TVi antigen itself when it is contained in the two varieties of Vi strain of Salm. typhi.

4. Treatment with alcohol:

(a) Most of the properties of the three Salmonella Vi antigens remain unimpaired by alcohol treatment.

(b) The physico-chemical behaviour of alcohol-treated or heated bacilli, as exemplified by agglutinability by pure Vi antiserum, depends on the presence of other constituents of the bacterial cell.

5. Treatment with formalin:

(a) The Vi agglutinability of the different Salmonella is preserved undamaged for long periods of time. The 0 inagglutinability of Salm. typhi is also well maintained.

(b) It is not yet known whether formolized AVi and BVi antigens undergo an alteration that leads to ‘functional deficiency’ of the corresponding antibody similar to that of the altered T Vi antibody.

6. Treatment with phenol produces a reversible inactivation of the agglutinogenic activity of the T Vi antigen but not of the A Vi and B Vi antigens.

7. The inadequacy of the symbols employed by Kauffmann for expressing the changed reactivity of differently treated antigens is emphasized.

1. It has been found possible to detect traces of the virus of foot-and-mouth disease in mixtures of this virus and that of vesicular stomatitis by cross-immunity tests in cattle. A quantity of the virus of foot-and-mouth disease as small as 1·3 I.D. 50 was revealed by the inoculation of such mixtures into cattle immune to vesicular stomatitis.

2. It is suggested that such a method would be useful in the analysis of possible mixed infections in field outbreaks.

3. The usefulness of complement-fixation tests to confirm the resolution of mixtures has been demonstrated.

The author wishes to thank the Director of this Institute, Dr I. A. Galloway, for his advice and encouragement. The technical assistance of Messrs E. Scoates and P. M. Mitchell is also gratefully acknowledged.

Twenty-four selected strains of Pseudomonas aeruginosa have been classified on the results of precipitin tests with trichloracetic acid extracts into six serological groups.

This differentiation was confirmed by O agglutination.

Four of the six serological groups can be further subdivided according to their sensitivity to one or more of nine phages.

Alcohol treatment reduces agglutinability of some strains, particularly if alcohol is added to very concentrated suspensions of bacteria in the preparation of O suspensions.

The precipitation method has been applied to eighty successive strains of Ps. aeruginosa isolated, and with the available sera seventy-two could be placed into one or other of the six serological groups.

I am grateful to Mr G. S. Turner and Miss D. L. Deeks for their able and enthusiastic assistance in this work, and to all the members of my staff for their constant helpful co-operation.

The Virus Research Unit was in receipt of additional financial assistance from the Nkana-Kitwe and Chingola Poliomyelitis research funds.

Antibacterial sera against C. diphtherias gravis (serological type I), intermedius and mitis were prepared in horses. The details of the immunization and the preparation of the sera are described. The sera contained a minimal amount of antitoxin.

The anti-gravis serum in the dose used, when it was given to guinea-pigs 1 hr.before the subcutaneous injection of ten lethal doses of the homologous organism, saved the life of 75 % of the animals. Those which died lived longer than the controls, and C. diphtheriae was recovered less often from the local lesion and the liver than at the same sites in the controls; it was not recovered from the heartblood of the animals which received the experimental serum, whereas cultures from the heart-blood were positive in about 40 % of the controls.

In these experiments the degree of invasiveness of the strain, as judged by the frequency of its recovery in cultures from the heart-blood and the liver of the control animals, corresponded with earner observations (Robinson & Marshall,1934), which showed that in routine virulence tests carried out by the subcutaneous inoculation of guinea-pigs C. diphtheriae could often be cultivated from the local lesion, the liver, the pleural fluid and the heart-blood, and that gravis and intermedius strains were more invasive than mitis strains. This was one of the main facts which suggested that an antibacterial factor might prove to be a useful adjunct in the treatment of diphtheria.

The anti-gravis serum, when given as a supplement to the routine doses of antitoxin to patients with severe diphtheria, had a definitely beneficial influence on the local lesion. Moreover, the serum improved for a time the general condition of some of the patients, including those who died later. The death rate of the group to which anti-gravis serum was administered was similar to that of the control group but, on the other hand, the patients who received the serum lived, on the average, longer than the control patients notwithstanding that treatment of the former group was begun later in the disease. The clinical evidence in favour of the anti-gravis serum cannot be assessed quantitatively, but, since it is based on critical observations, it should not be disregarded.

Consideration of the results as a whole leads us to conclude that the anti-gravis serum used in the tests contained a specific antibody which, within limits, had a beneficial action. We suggest, therefore, that the role of the antibacterial factor in the pathogenesis and control of diphtheria merits investigation.

1. Salmonella paratyphi A possesses a specific Vi antigen essentially similar to that of the Vi antigen of Salm. typhi.

2. O-insensitive strains of Salm. paratyphi A contain more Vi antigen, and are more virulent for mice, than O-sensitive strains.

3. The most suitable strains for the preparation of paratyphoid-A vaccine are those that develop both the Vi and the O antigens in maximum quantities. The methods of selecting the vaccine strain of Salm. paratyphi A and routinely examining the cultures are essentially the same as those applied to the virulent Vi+O form of Salm. typhi.

4. Pure A Vi antiserum is destined to assume importance in the identification of Salm. paratyphi A strains.

5. Kauffmann's erroneous conclusion that Salm. paratyphi A does not develop Vi antigen is critically reviewed.

1. An account is given of the design and conduct of a controlled trial of streptomycin by mouth in the treatment of infantile diarrhoea and vomiting.

2. The classification of cases is described. The streptomycin-treated and control series are shown to be comparable in respect of severity of illness on admission and in respect of other factors likely to affect progress.

3. The criteria adopted for assessment of progress are defined, viz. degree and duration of dehydration, duration of diarrhoea and tolerance of feeds.

4. Comparison of streptomycin-treated and control cases was based partly on progress during the first week of hospital stay and partly on subsequent progress.

5. No significant difference was shown between the streptomycin-treated and the control series in respect of progress during the first week. Severe relapses of diarrhoea were proportionately higher in the streptomycin-treated series, but the difference was not statistically significant.

6. The bacteriological investigations are described. Bact. coli O group 111 and Bact. coli O group 55 were used as ‘ indicator organisms’ in studying the immediate and delayed effects of streptomycin on the coliform flora of the intestine. The results show the inhibition, in streptomycin-treated cases, of coliform growth from the rectal swab, and the subsequent emergence of streptomycin-resistant strains.

7. Evidence of cross-infection with the ‘ indicator organisms’ is given. Problems of cross-infection and relapses of diarrhoea are discussed in relation to the assessment of the effects of a chemotherapeutic agent.

8. The difficulty of defining criteria for classification of cases and assessment of progress is discussed.

We wish to express our most grateful thanks to Sister A. Doherty and other members of the nursing staff of the hospital for their unfailing co-operation in a long and often tedious investigation; to members of the medical staff, in particular to Dr V. V. Tracey, for assisting in the work; and to the entire staff of the hospital's laboratory.

We are indebted to Prof. Robert Cruickshank for his help in initiating the trial and for his continued interest and encouragement; and to Dr Joyce Wright for valuable criticism and help in the bacteriological investigations. Acknowledgement is due to Dr R. Swyer, Physician (Infectious Diseases), St Ann's General Hospital, for facilities for carrying out the trial; to Dr Joan Taylor and Miss R. E. Hilton for their serological and biochemical investigations; and to Prof. A. Bradford Hill for helpful statistical criticism.

Naval ratings serving in a warship in the tropics answered a questionnaire concerning their sensations of warmth at meal times on the mess-decks during two investigations carried out at an interval of 6 months when the ship was proceeding between tropical and more northerly waters. Temperature measurements were recorded at the same times as the questionnaires were completed by the subjects.

It is proposed that the upper level of warmth to be accepted in the design of ventilation systems for accommodation spaces should be those conditions when either more than 80% of persons feel reasonably comfortable or less than 20% observe the presence of unevaporated sweat on their skin, which corresponds to effective temperatures of 79 and 78° F. respectively, according to the majority opinion in these two investigations. The latter figure of 78° F. is suggested as the upper level to be accepted for men living under these conditions.

The upper level of the optimum or ideal zone for comfort, when equal proportions of persons reported that they were either cool or warm, was observed to be at 74° F. effective temperature.

The proportions of men who were uncomfortably warm at different levels of warmth were related to the proportions who reported that their skin was wet with unevaporated sweat. Very few observed that they were sweating, whatever the air velocity or their clothing, when the dry-bulb temperature was below 83° F., and the wet-bulb temperature below 75° F.

Brisk circulation of the air on the mess-decks made an increasingly large contribution to the comfort of these very lightly clad men as the air temperature rose from 85 to 89° F., but as the temperature fell below 83° F. variations in air Thermal comfort in the tropics 431 movement had little effect on the proportions of persons who were reasonably comfortable under these warm conditions.

Thermal discomfort was reported by an increasing number of men within a temperature zone where others have shown by experiment that there will occur a statistically significant decrease in the skilled performance of men who are acclimatized to working at high temperatures.

This paper is published by the permission of the Admiralty and the Medical Research Council's Royal Naval Personnel Research Committee. The investigations described were made possible by the active interest of Captain H. S. Hopkins, C.B.E., Royal Navy, and Commander W. Elliot, D.S.C., Royal Navy, the excellent facilities which they placed at the disposal of the observers, and the wholehearted co-operation of the naval ratings who acted as subjects. I am indebted to Dr T Bedford for advice and criticism and to Mr Cheng Swee Guan and Sick Berth Attendant W. F. Montague, Royal Navy, for care in tabulating the data, preparing the illustrations and technical assistance.

This investigation has been concerned with the changes that occur in the coliform and faecal coli content of water samples on storage at room and refrigerator temperature for 20–24 hr.

On examination by a seventy-tube method, using twofold diminishing volumes, twenty-three out of 151 samples of water stored overnight at room temperature showed a significant increase in the presumptive number of conform organisms and twenty-nine a significant decrease. Of the same number stored in the refrigerator, ten showed a significant increase and twenty-six a significant decrease.

Of 111 samples examined for faecal coli, eight showed a significant increase on storage at room temperature and twenty-nine a significant decrease. Of the same number stored in the refrigerator, seven showed a significant increase and fifteen a significant decrease. The effects of the source of the water, the time of year and the original number of coliform organisms in the sample were examined.

The present investigation has shown that in a considerable proportion of samples a significant change in the conform and faecal coli content does occur on overnight storage at room or refrigerator temperature. It may, perhaps, be safe to store samples under some conditions for shorter periods, but this is a matter for future investigation.

A method of titrating influenza virus-neutralizing antibodies is described. This method is based on the technique for cultivating virus in fragments of chick chorio-allantoic membrane using a special plastic tray with cups for 100 replicates. The neutralization test cannot be made directly by adding the antiserum to the cups in which the virus is growing in the membrane fragments because there is a significant non-specific inhibitory effect. Therefore, use is made of the observation that virus added to the cups is rapidly fixed by the membrane so that, after a short time, the membrane may be washed and transferred to a fresh cup which will later be found to contain the new virus released from the membrane.

In the neutralization test the membrane fragments are exposed to virus-serum mixtures for 18 hr. and then transferred to other cups without serum to determine which pieces have become infected. The titre of neutralizing antibodies is defined as the serum dilution which will protect from infection with 100 ID 50 of virus, half of a large number of replicate pieces of membrane.

The neutralization technique has been applied to the antigenic comparison of six strains of influenza virus. The results support the relationships disclosed between the six strains by a complement-fixation technique. It is suggested, however, that the neutralization technique, because of its greater specificity, and because of its dependence on the infectivity of the strain, is not as satisfactory as the complement-fixation test for strain comparisons. Apart from its use for titrating neutralizing antibodies, the technique may also be of value for proving the virtual identity of two virus strains and in the selection of strains for a vaccine.

It is a pleasure to acknowledge the help given by Dr Isaacs of the WHO Influenza Centre. Not only did he provide many of the virus strains and antisera, but he also undertook the purification of the WS strain.

1. Salmonella paratyphi B possesses a Vi antigen essentially similar to the Vi antigens of Salm. typhi and Salm. paratyphi A.

2. The biological function of the Vi antigen of Salm. paratyphi B is to protect the O antigen against the action of the natural or immune O antibody, thereby increasing the virulence of the microorganism.

3. The most suitable paratyphoid-B vaccine strains are those that develop both the Vi and the O antigens in maximum quantities. The methods of testing vaccine strains of Salm. paratyphi B are essentially the same as those applied to the virulent Vi+O form of Salm. typhi.

4. Evidence is brought to show that Kauffmann's conclusion that Salm. paratyphi B does not develop Vi antigen is unfounded. The so-called O-factors V and V obviously cannot be classified as O antigens.

5. It is suggested that the symbol V of the Kauffmann-White schema be abandoned and this antigenic component be classed as the Vi antigen of Salm. paratyphi B.

1. Four different Bacteriaceae possessing the same Vi antigen showed, after exposure to heat, striking differences in the physico-chemical behaviour of the Vi antigen. The most noticeable differences observed were those in the changes in Vi-agglutinability and O-inagglutinability of the bacteria; lesser differences were noted in the agglutinogenic activity of the Vi antigen, and none in its agglutinin-binding capacity.

2. Treatment with alcohol altered the TVi antigen of one of the species (Bact. coli 5396/38) in a way different from that seen in the other three species.

3. On the other hand, dilute acid or alkali produced the same chemical changes in the TVi antigen of all four Bacteriaceae.

4. The TVi antigen present in the four Bacteriaceae appears to be one and the same substance; it cannot be differentiated by the customary serological methods. Its different physico-chemical state after exposure to heat or alcohol is, therefore, conditioned by other constituents of the bacterial cell, which may, or may not, be antigenic.

5. The simultaneous O- and Vi-inagglutinability resulting from heating at 75° C. is particularly impressive since it does not appear to be specially related to any one of the known antigenic components.

6. These findings invalidate the basis on which the L, A and B antigens of Bact. coli have been differentiated.

7. There is also no valid reason for designating the labile somatic antigens of Salmonella and other Bacteriaceae as K antigens. These antigens have the general characters of the Vi antigen of Salm. typhi, are demonstrated by methods developed in the study of the typhoid Vi antigen, and are not associated with typical capsules.

8. The M (mucoid) antigens of Salmonella and of Bact. coli are in many respects different from the Vi antigens and should be classified separately.

I gratefully acknowledge the valuable technical assistance received from Mr F. J. Flynn throughout the work reported in this and the preceding papers of this series.

1. A quantitative study has been made of the activity of C′1, C′2 and C′4 in the complements of the horse, cat, pig, man, guinea-pig and rabbit.

2. Evidence is presented which suggests that either there may be two types of C′4, the one haemolytic and the other conglutinating, or else that the manifestation of this complement component varies under conditions not yet understood.

3. In certain complements it appears that the serum proteins functioning as C′1 with the bovine anti-sheep red cell antibody are distinct from those functioning as C′1 with an antibody of similar specificity from an immune rabbit serum.