1.Phase variation of the somatic antigen I in Bact. paratyphosum A and of the component XII2 in Bact. typhosum and British strains of Bact. enteritidis has been confirmed.

2. Strains of Bact. enteritidis isolated in India by blood culture from cases of invasive disease in man fall into two serological types, neither of which undergoes demonstrable somatic phase variation.

3. The probable mechanism of this type of variation is discussed in the light of experimental findings.

4. The effect which phase variation may have on the clarity of end-points in agglutinin titrations and on the serological standardization of Salmonella O-suspensions is demonstrated.

5. The variation is briefly discussed from the points of view of Salmonella philogeny and of human Salmonella infection.

I am indebted to the D.M.S. in India for permission to publish this paper.

The work began with efforts to isolate a type II Vi phage from local sewage. When this attempt failed, typing material was obtained from Dr J. Craigie and Dr A. Felix.

The typing method was applied to 495 strains of typhoid bacilli, the technique laid down by Craigie being followed throughout.

It was found that in South Africa the distribution of the various types of typhoid bacilli is different from that obtaining in other countries where this typing method has so far been applied.

During further investigations of type F strains the fact came to light that by means of a simple biochemical test type F strains can be subdivided into two sub-types, which subdivision will be useful in further epidemiological work. No such biochemical subdivision was found possible in any other type.

It did not prove possible to type all South African strains with the available phage preparations. Some of these strains must represent new types. Workers in other countries have had similar experiences and it will become necessary to compare all these unusual strains directly with one another in order to reach uniformity in nomenclature. It has already been possible to co-operate with workers in England, and they have been assisted by my finding that a strain, unique in England and supposed to have been brought there by a carrier, who had typhoid fever in South Africa 40 years ago, was endemic in South Africa.

During the course of the work, two outbreaks of typhoid fever occurred in which the typing method proved most useful, not only for linking up a particular carrier with a particular outbreak, but also for excluding an already known carrier who was under grave suspicion.

I have to thank Dr Craigie and Dr Felix for supplying material, and Prof. Pijper, of the Pretoria University, for placing laboratory facilities at my disposal. I thank all others for cultures received for typing.

Presence of thermolabile agglutination-inhibiting substances in strains of the genus Shigella has been demonstrated by numerous authors. Archer (1942) used as antigen for agglutination tests with S. alkalescens cultures which had been killed at 100° C., since a thermolabile factor which prevents agglutination by the usual technique had been found present in the organism. The same method was employed by Mendes Silva (1943) in agglutination tests with S. ambigua. Braun & Unat (1943) found in S. paradysenteriae (Flexner) a labile antigen designated ‘O1’ which inhibits O-agglutination of the living bacteria. Schuetze (1944) reported that the insensitivity of S. dysenteriae (Shiga) to agglutination is abolished by heating at 100° C. for 30 min. Weil, Black & Farsetta (1944) reported that boiling for 1 hr. renders inagglutinable strains of S. paradysenteriae (Flexner) fully agglutinable. Recently, Weil et al. (1946) reported that one of the Sachs types contains an inhibiting substance, a heat labile antigen and a heat stable antigen. He observed inhibition of agglutination of living bacteria in many Flexner types, S. alkalescens and S. ambigua. Olitzki, Shelubsky & Koch (1946) reported that there is present in S. dysenteriae (Shiga) a labile substance ‘I’ which inhibits the agglutination of thd living bacteria. They were able to demonstrate a labile antigen in toxic saline extracts but not in the intact bacteria. The present paper reports a method for isolation of this antigen in the form of a relatively concentrated preparation.

A review is given of some of the literature concerned with bacterial infections and immunity in invertebrates and cold-blooded vertebrates. Among the former, insects have been most fully studied, although a certain amount is known of the immunological reactions of marine invertebrates. Diseases of insects are mainly generalized, bacteraemic conditions, and the great majority are caused by Gramnegative cocco-bacilli. Diseases of cold-blooded vertebrates also are usually generalized, but bacteria from a very wide variety of groups have been isolated from them.

It is probable that this resemblance in the infections of these widely separated groups of animals is due to the fact that, unlike mammals and birds, with whose reactions we are more familiar, the regulation of their temperature and the constitution of their body fluids is not exact. This toleration of wide variations of their own physical and chemical constitution must reduce their sensitivity to the changes produced by infection, and hence decrease the likelihood of a local reaction, designed to confine the invading organism to the immediate region of its point of entry, i.e. a local inflammation.

The production of humoral antibodies appears to be almost universal in the animal kingdom, although greatly affected by changes in temperature.

The effect of temperature upon the balance between host and parasite is also discussed. This question bears upon the problem of occasional pathogenesis by saprophytes. Even among mammalian pathogens the borderline between parasite and saprophyte is an indistinct one, especially in such cases as Proteus and Pseudomonas pyocyanea, and even those species which are usually regarded as exclusively parasitic may readily be constrained to adopt a saprophytic existence on artificial culture. Where cold-blooded animals are concerned the boundary is even more difficult to draw, and it is possible that under suitable conditions, bacteria which are normally saprophytes may be capable of causing infection. This point will be impossible of proof until more sensitive means of definition of bacterial species are discovered.

1. Whole cultures of Bact. coli grown under carefully controlled conditions have been subjected to the action of heat at temperatures ranging from 47 to 55° C. and the survivor curves determined.

2. The occurrence of excessive variation between replicate plates of counts made when the mortality exceeds 95 % was very much less evident than when phenol was the lethal agent.

3. At temperatures of 35° C. and below, disinfection of the cultures was never complete, as a permanent population of cells became established. In some cases the numbers of heat-resistant cells reached nearly 0·01 % of the original population, but great fluctuations were observed.

4. The death-rate during the active part of the disinfection was not constant but, in general, increased with time. In the faster disinfections this increase was difficult to detect and the logarithmic death-rate appeared to be virtually constant, but the increase is nevertheless believed to have been present in all cases.

5. The data for all the experiments have been combined to give a composite disinfection curve from which it was concluded that the death-rate was low at first but rose to a maximum at which it remained constant until the mortality had reached at least 99·99%.

6. This finding led to the decision to treat all the experiments as if the regression of log survivors on time had been linear between mortalities of 90 and 99·99%. The standard errors of the calculated regression coefficients were small, so that this method of treating the experimental data appears to be justified.

The authors wish to express their gratitude to Messrs I.C.I. (Pharmaceuticals) Ltd. for their valuable financial assistance, which has enabled one of us (H. E. F. D.) to devote his full time to this work.

1. A study has been made of the probit-log survival-time relationship in a number of experiments in which whole standard cultures of Bact. coli were subjected to heat at temperatures ranging from 47 to 55° C.

2. It is concluded that the whole mortality up to 99·99 % could not be covered by a single straightline relationship since the graphs are curves concave upwards and to the left.

3. Between 95 and 99·99 % mortality the relationship approximated closely to linearity.

4. Several individual experiments showed ‘humps’ in their curves. These, although always occurring at the same stage in time in the disinfections, varied in their probit levels.

5. By combining the data for all experiments, an idealized probit-log survival-time curve was obtained in which the changes of slope were less marked than in the curve representing the disinfection of similar cultures by phenol.

The authors wish to express their gratitude to Messrs I.C.I. (Pharmaceuticals) Ltd. for their valuable financial assistance, which has enabled one of us (H. E. F. D.) to devote his full time to this work.

In this paper a fatal case of meningitis is reported in which a capsulated Diplococcus of the Neisserian group was isolated from the purulent cerebrospinal fluid. A capsulated Gram-negative Diplococcus growing readily on agar was isolated from the nasopharynx and the cerebrospinal fluid by von Lingelsheim (1906, 1908) and called by him D. mucosus. Similar organisms were also isolated from the nasopharynx by Elser & Huntoon (1909). Cowan (1938) reported two cases operated on for cerebral tumour in which D. mucosus was grown from the cerebrospinal fluid. These strains were regarded by him as contaminants setting up a low-grade infection in a brain which, exposed for operation, offered little resistance to infection. Two cases of meningitis due to a capsulated Diplococcus have been reported by McFarlan (1941) and Bray & Cruickshank (1943); both these patients responded well to sulphapyridine and recovered. Edwards (1944) described a fatal case of meningitis due to an atypical Neisseria which in many respects resembled D. mucosus, although it was not capsulated. The rarity with which capsulated Neisseriae are associated with pathogenicity is considered sufficient justification for reporting the following case.

1. We have investigated bacteriologically the conditions in a central meals kitchen from which about 3000 meals were issued daily to some 65 departments.

2. Plate counts showed that the containers in which the meals were sent out contained large numbers of organisms, from several thousand to several hundred thousand per container.

3. Counts of this order were obtained both before and after the routine method of washing.

4. These counts were reduced to the order of a few hundred or less per container by steam sterilization.

5. Exposure for 2–3 min. to steam from a simple jet sterilizer was effective for all except large insulated lids which required at least 5 min.

6. Treatment for 5 min. in a chamber filled with steam from a boiler was effective with all types of containers and lids tested.

7. If containers are to be sterilized at all they must be sterilized each time they are to be used.

8. A great improvement in the bacteriological counts can be achieved by attention to the hygiene of washing, e.g. by the use of plenty of hot water with soap and soda or perhaps preferably with detergents, with a second sink of very hot water for rinsing, and by using fresh sterilized dish-cloths each time the washing water is changed.

9. Counts almost as good as those given by steam sterilization were obtained by washing with soap and water at 48–50° C., wiping with a sterilized dish-cloth and then rinsing in a second sink of water at 75–80° C.

10. The arguments for and against steam sterilization are discussed.

This investigation was greatly helped by the co-operation of the Leicester Education Department.

1. A cow given by the mouth very large doses of virulent typhoid bacilli (275 and 125 thousand millions) on two occasions at an interval of 29 days failed to show signs of infection or to pass the bacilli in faeces or milk.

2. Of two-month-old calves, dosed with about 100 ml. ascites broth culture of typhoid mixed with milk, one became infected and typhoid bacilli were grown from the intestine and ileocaecal gland. Both, however, passed typhoid bacilli in the faeces on the first or second day after dosing.

3. When large numbers of typhoid bacilli (say, 1–3 thousand millions) were instilled into the udder through the teat canal, excretion of the bacilli in the milk usually ceased within 10 days (fourteen quarters of four cows). In two quarters of one of these cows excretion continued for 25 and 27 days, and in two quarters of another cow for at least 85 days. The procedure usually caused an acute mastitis lasting for a few days, and the ‘H’ and ‘O’ agglutinin titre of the blood rose.

4. When large numbers of typhoid bacilli were applied to the teat orifice, even when this had been grossly injured beforehand by slitting the sphincter, there was as a rule only a very transient infection. In eleven cases out of twelve, either the organism could not be found in the milk or was found for 1 day only; in the remaining case there was intermittent excretion for 7 days.

5. The strain of typhoid used for most of the above experiments was of high Vi content and was isolated from the carrier involved in the Bournemouth-Poole epidemic of 1936.

6. These experiments do not support the view that milk-borne typhoid outbreaks are brought about by an active infection of the cow and suggest that the main object of the epidemiologist must still be the location of the human carrier.

7. The chances of typhoid bacilli growing in the milk after its withdrawal from the cow are stressed.

This work was rendered possible by financial assistance from the Ministry of Health and from United Dairies Limited. Thanks are due to Dr A. Felix for providing a Vi-rich culture of typhoid, and to Dr V. D. Allison for valuable help in checking the draft paper and making suggestions.

1. Breeding records are given of fourteen female Dutch rabbits reared and maintained without green food, and of twelve of their female offspring.

2. Fertility, average litter size at birth, and the percentage of young weaned, were good for the breed.

3. No difference in growth rate was found between the first and second generations of young.

4. With the dry pelleted diet, supplemented by drinking water, fresh green food is unnecessary for breeding rabbits, as it has already been shown to be unnecessary for growing animals (Bruce & Parkes, 1946).

I should like to thank Dr A. S. Parkes, F.R.S., for his help in the preparation of the manuscript.

On the basis of experience in using D.D.T. as an insecticide in both liquid and powder forms, suggestions are made for improved spray and dust guns.

My thanks are to S./Sgt. Toop, workshop foreman, and Pte Carter, tinsmith, of my old hygiene section, for the months of experiment, trial and error involved in producing these apparently simple innovations. Also to Brigadier H. B. F. Dixon, D.D.M.S. West Africa Command, for permission to forward this article for publication.

1. A general survey has been given of the subject of repellents against mosquitoes. Results recorded in published literature prior to the war have been summarized and those from the greatly increased volume of unpublished research during the war briefly outlined.

2. Of the older literature it has been shown that results were often conflicting due to many of the substances tested not being pure chemicals but very commonly essential oils and other products having a variable composition and to the methods used in testing being insufficiently precise and standardized. Particularly, standardization was lacking in that most important respect, viz. the condition of the mosquitoes used in the tests. The whole treatment of repellency on a scientific basis depends upon the suitability of the methods used in testing and for this reason a full and detailed account has been given in § IV of the methods used in the present enquiry in rearing standardized cultures of Aedes aegypti and in § V of the methods of laboratory testing and of interpreting results.

3. Some remarks have been made on the proper sphere of field tests. Field tests, it is pointed out, are extremely valuable but less suitable than laboratory tests for determining with precision the properties of repellent substances owing to the impossibility of obtaining controlled conditions. The proper sphere of such tests would appear to be as an extension of the results of laboratory tests to cover the many qualifying circumstances which are introduced when repellents of known effectiveness are used in practice.

4. A description has been given in § III of the chief features of such repellents as have been in use, and in § VI are given the results of testing a considerable number of compounds and preparations, with data showing their comparative repellent power and duration of protection afforded. In this series three types of compound especially exhibit repellency, viz. (1) unsaturated long-chain alcohols, aldehydes and phenolic compounds such as are characteristic of most essential oils, oil of citronella, due to its active principle citronellal, being one of this type; (2) high boiling-point esters of the lower methyl series with various high molecular weight acids, such as phthalic, adipic, cinnamic, citric, etc., of which dimethyl phthalate is the best-known example, and (3) high boiling-point alcohols, especially diols, which need not necessarily be unsaturated or contain methyl series groups, of which Rutger's 612 or 2ethylhexane-l:3-diol is that best known. There are, however, undoubtedly many other groups which for some reason may possess this property as may be instanced the last entry on the list of compounds tested which belongs to none of these groups but was found as powerful as any compound tested.

5. Attention has been directed to the importance, in connexion with duration of effect, of the boiling-point. It is shown that to give protection for a period sufficient to qualify a substance to be considered a repellent, i.e. in the criterion adopted by the Unit a protection period of 2 hr., it must have the relatively high boiling-point of at least 250° C. at 760 mm. pressure. Compounds with lower boiling-point merely volatilize completely on the warm skin by the time this period is reached. To give protection for 6 hr. the boiling-point must be at least of the order of 280° C. Substances with boiling-point much over this, whilst they show lasting effect, are apt to exhibit a lower grade of repellency. In this connexion a method, sufficiently accurate for the purpose, has been quoted by which any boiling-point given as at a reduced pressure can be reduced to its theoretical boiling-point at 760 mm. The basis on which the necessary factors for this have been arrived at will be found in a note kindly furnished by Dr E. A. Moelwyn-Hughes given as an appendix.

6. The question of rub-resistant creams, of measurement of effect of repellency at a distance, of impregnation of clothing with repellents and of the testing of penetrability of clothing fabrics to mosquito bites are dealt with in their respective sections and any conclusions regarding these subjects there given. The last has been dealt with in some detail since no account of such testing has, so far as is known, been given in the literature. A note by Mr J. R. Whinfield, Ministry of Supply, giving technical methods of measuring structural features of fabrics has been given as an appendix and should be useful to anyone taking up the study of fabrics from this point of view.

7. It remains desirable, perhaps, to indicate in a general way what recent advances in regard to mosquito repellents amount to in practice, apart altogether from theoretical or scientific considerations. Results during the war have related almost entirely to the use of repellents for military purposes and the repellent almost universally employed in practice has been dimethyl phthalate. A number of reasons have combined to make this the repellent of choice for such a purpose. For the American Army the so-called 6: 2: 2 mixture, i.e. a mixture in the above proportions of dimethyl phthalate, Rutger's 612 and indalone, has been the official repellent. But Rutger's 612 has never been so freely available as dimethyl phthalate and neither it nor indalone has been in extensive use in this country. Dimethyl phthalate has very generally been found most suitably used ‘straight’ being carried in a suitable receptacle from which a little is shaken out into the palm of the hand and used to anoint exposed skin areas or even clothing. For impregnation of tropical uniforms it has usually been used suitably dissolved in a solvent or as an emulsion and the articles sprayed or dipped and wrung out. It is effective against many forms of biting insects and has been used against mites. In this last respect, however, it has been to some extent replaced by dibutyl phthalate, which when used for impregnating clothing is by reason of its higher boiling-point more resistant to washing.

8. For civilian use there would appear to be a wider field. One paramount requirement for military use has always been that the repellent should give very prolonged protection, i.e. that it should retain its effects for such periods as overnight, or as often specified for 6 hr. This has limited the choice of repellents excluding many substances which might be very effective for shorter periods. For the civilian such emphasis on duration of effect does not arise, for circumstances must be rare in which the application could not be renewed, say after 2 hr. if necessary. Again should it only be a matter of giving comfort against moderate numbers of mosquitoes something less drastic than anointing all bare skin with dimethyl phthalate might well suffice. For military purposes smell is a serious, indeed fatal, drawback. For the civilian a pleasant smell might even be an advantage. The civilian too would probably prefer a light pleasant cream to the repellent in liquid form. There is much scope here for the chemist and the cosmetician.

What then at present should the civilian be advised to use? This is dependent very largely on what he can now readily purchase, and there is some difficulty in saying under present circumstances what this might be. If a repellent is required to prevent attack where mosquitoes are a serious nuisance, and the same applies to midges, probably the best advice would be to try dimethyl phthalate straight as in military use. This might be obtained by order through a chemist to some wholesale firm such as Boake Roberts and Co. or Stafford Allan and Co. These or other firms would probably supply a moderate quantity, say 4 oz. on such an order. This could be used as stock from which to fill a small corked bottle for use. A corked bottle is better than a stoppered one as dimethyl phthalate is very liable to leak. It should be applied by pouring a little in the palm of the hand, rubbing the hands together and anointing any exposed parts such as the neck, behind the ears and even the face, but avoiding neighbourhood of the eyes. It might be more pleasant, especially if conditions were not so severe, to use a dimethyl phthalate cream such as Dr Hamil's very pleasant wax and arachis oil cream or the 70% magnesium stearate Porton cream as given in § VI(2). This could probably be made up to the formula given by a dispensing chemist. If it were possible to obtain it, a good Java citronella oil would probably be even better, for the great advantage here is that a good citronella oil will keep mosquitoes at a distance and complete covering of skin with the repellent is not so necessary as with dimethyl phthalate. Or if such an oil is not to be at present obtained it might be possible to get citronellal, its active principle, or even citronellol the corresponding alcohol. A little experimentation and careful preliminary trial is always desirable as to what can be done and as to how effective the substance seems to be. This applies especially to the case of midges which are not always the same species and might not in some cases be repelled. It should be realized that for civilian use there is still much that is experimental. For those with a liking for experiment it might be well worth while to try out that very pleasant smelling substance hydroxycitronellal, a note on which will be found in § III.

Mention should also be made of certain proprietary products issued by certain American firms which might at any time be obtainable in this country, such as ‘Skat’ or ‘Eveready repellent’. Skat (Skol Company Inc., New York) is issued in two forms, viz. as containing dimethyl phthalate and as containing indalone (in both cases practically these substances), and Eveready Insecticide (National Carbon Co. Inc., New York) as ‘formula 612’, in this case Rutger's 612. Very likely in time good repellents will eventually be put on the market by various firms in this country.

9. As regards the future no limit can probably be placed upon what may result from organized research which is likely to be taken up more seriously now than was previously thought worth while. Research in this field may be said to have reached a point where methods of testing will enable precise results to be obtained and if work by the synthetic chemist is linked with that of the insect biologist further progress in the discovery of still more effective repellents is extremely likely. Research should not, however, stop here for much still remains to be done in the direction of ascertaining methods of using such repellents to the best advantage to meet different requirements and even of ascertaining exactly what such requirements are.

In this investigation the hitherto unexplored field of tuberculous Lanarkshire has been surveyed.

The relative incidence of human and bovine strains of the tubercle bacillus in 80 cases of pulmonary tuberculosis and in 311 cases of extrapulmonary tuberculosis has been examined. Results have shown that bovine infections were present in 2·5% of the pulmonary cases, a figure that is in keeping with the expected incidence for an area as highly industrialized as Lanarkshire.

The extrapulmonary strains have been further subdivided into 165 strains isolated from patients suffering from tuberculous meningitis, 72 strains isolated from patients with cervical adenitis and 74 strains from patients with tuberculous lesions affecting bone or joint. Each group has been analysed separately.

In the 165 cases of tuberculous meningitis bovine infections accounted for 28·5 % of the total. This figure was slightly higher than the 22% given by Blacklock & Griffen (1935) for the Glasgow area and the 24% of Macgregor & Green (1937) for the Edinburgh area, but it was lower than the 36% found by Munro & Scott (1936) in the east of Scotland and the 40·5% given for Scotland by Griffith (1934) in his general survey. The greater preponderance of bovine infections in rural areas supported the observations of these workers.

The cervical adenitis group gave the percentage of bovine infections as 30·4, a result much lower than any previously published for Scotland, but there can be no doubt that a family history of pulmonary tuberculosis obtained from 50 % of the patients in the industrial group who supplied human strains must have been a most important factor in influencing the relative incidence of the two types of tubercle bacillus. There were 72 cases of cervical adenitis investigated and 52 came from the industrial area.

Seventy-four patients suffering from tuberculous lesions affecting bone or joint gave a bovine incidence of 18·9%, a percentage that is lower than that given by Munro & Cumming (1926) of 36·4% but approximating to the 20·5% given by Wang (1917) for the Edinburgh area. Munro & Cumming were dealing with a less industrialized area than Lanarkshire which probably accounts for the variation as also does the fact that, of 7 human cases occurring in the rural part of Lanarkshire, 5 gave a family history of pulmonary tuberculosis.

These figures only serve to substantiate observations of regional variation and endorse the findings of other workers who have found a lower proportion of bovine infections in industrial districts. As Lanarkshire is the most highly industrialized county in Scotland it was to be expected that the results of this investigation might have given an even greater predominance of human infection than actually was found.

An investigation into the probable influence of environmental factors on the incidence of extrapulmonary tuberculous infections and into the possible explanation of their effect on regional differences in the proportional frequency of the type of infecting strain of the tubercle bacillus has been described.

The importance of substandard housing and of family history in infection with the human type of the tubercle bacillus, particularly in children, has been noted.

Circumstantial evidence of both low economic resources and milk supply influencing infection with the bovine type of the tubercle bacillus pointed to these factors being important in childhood.

I wish to acknowledge the co-operation I received from the Medical Officers of Health for the several Local Authorities within the county of Lanark.