The original scope of this investigation was the study of the passage of hydrocyanic acid vapour through the skin. We have partly repeated and confirmed the results of Walton and Witherspoon (1926). In view of the extensive use of hydrocyanic acid vapour in the destruction of vermin in ships the subject is of considerable practical interest. In the course of our work certain other matters were considered and these are referred to in the text.

The control of the lead risk among shipbreakers presents great practical difficulty. Measures for the prevention of plumbism applied successfully in other industries are here impracticable, with the result that the breaking up of a heavily leaded ship is almost invariably responsible for a number of cases of lead poisoning and much ill-health stopping just short of incapacity for work.

As a result of some preliminary investigations upon the aetiology of an asthmatic type of disability among workers in carding-rooms of cotton mills, Maitland, Heap and Macdonald (1932) showed that it was possible to extract from some samples of cotton dust a substance which gave the biological reactions of histamine. The dust was that to which affected workers would be exposed. The amount of histamine recovered varied from 0·02 mg. to less than 0·004 mg. per g. of dust. Different bulk samples of dust contained unequal amounts, and the amount also varied in different fractions of one sample graded according to the size of dust particles. It was thought that the finer fractions contained the larger amounts of histamine and that some bulk samples were, relatively at any rate, histamine free.

1. A study of the suspension of dried faeces used in the Chick-Martin test, and of the process of disinfection in this test has shown that:

(a) The suspension of faeces is unstable, and rapidly forms a compact deposit.

(b) Surviving bacteria in critical dilutions of the disinfectant are contained exclusively in this deposit.

(c) Uncontrolled factors affect the degree of subdivision of the particles of faeces.

(d) The degree of subdivision of these particles materially affects the activity of a disinfectant in their presence.

2. It is suggested that these facts among others may account for inconstant results which the test is said to give.

3. A suspension of yeast is proposed as a possible alternative form of organic matter in such a test.

At the First Congress of the International Society for Microbiology held in Paris in July 1930, a Nomenclature Committee of the Society was authorised and constituted by resolution in plenary session. Under article 4 section b of the resolutions concerning the formation of the committee, authority and encouragement were given for the monographing of special groups or genera of bacteria. In furtherance of this object a subcommittee was called together early in 1933 to consider the question of the taxonomy and nomenclature of the Salmonella group, Dr H. Schütze, Lister Institute, London, being appointed Chairman; the other members of the subcommittee comprising: Dr E. O. Jordan, University of Chicago, U.S.A.; Dr F. Kauffmann, Statens Serum-institut, Copenhagen, Denmark; Dr W. M. Scott, Ministry of Health, London; Mr P. Bruce White, National Institute for Medical Research, London; with Dr R. St John-Brooks, National Collection of Type Cultures (Medical Research Council), Lister Institute, London, as Secretary. The late Dr J. Lignières, some time before his lamented death, consented to serve on the subcommittee. As, however, he died before he had an opportunity of considering the draft report, which he did not take part in drawing up, his name does not appear as a member of the subcommittee.

1. An account is given of sporadic Salmonella infections. In the period 1929–32, thirty outbreaks occurred, involving forty-six cases, whereas, in the year 1933, thirty outbreaks involving forty-four cases were encountered.

2. The investigation has shown that the most frequent cause of such infections are Salmonella types aertrycke, thompson, dublin, enteritidis and european suipestifer.

3. It has been found that the dublin and suipestifer types are definitely more tissue invasive than the other types. Septicaemia and meningitis have been associated with infections due to S. dublin, and septicaemia in cases infected with S. suipestifer.

4. The biochemical serological characters of a new Salmonella type—S. aberdeen—have been described.

5. An infection due to S. potsdam has been encountered. This type has not been previously described as occurring in this country.

1. Provided that due precautions are taken to ascertain the normal level of agglutinins in a given host, and that antigenically smooth strains are used, there is every reason to believe that the occurrence of agglutinins in a titre above the normal range of variation is due to infection—latent, active or past—with the specific organism in question, or in a few instances with an organism, usually of the same genus, sharing a similar antigen.

2. This conclusion may have to be modified for the occurrence of agglutinins to Proteus OX19 in human typhus sera, since the exact relationship of this organism to Rickettsia prowazeki is still doubtful. If, however, future work shows the truth of Felix's (1933) contention that serological types of Proteus X correspond to serological types of typhus virus (Rickettsia), even this apparent exception will fall within the general rule.

3. The examination of sixty-four horse and fifty-three cattle sera, and the performance of numerous cross-agglutination and cross-absorption experiments with rabbit and horse antisera, lend no support to the suggestion made by certain workers of the existence of an antigenic relationship between Brucella strains on the one hand, and strains of Pfeifferella, Pasteurella and Proteus X on the other.

Autopsies on all the guinea-pigs kept for as long as 20 months failed to reveal any trace of tuberculosis, although the grass grazed by the animals was proved by subcutaneous inoculation to bear virulent organisms, which might have been expected to reach the alimentary tract in considerable numbers, at least in the earlier experiments.

By repeated infection of a grazing area it has been possible to induce tuberculosis in guinea-pigs grazed in the open and in those fed indoors on cut grass. In the 1932 experiment a total of thirty guinea-pigs were allowed access to plots of grass which had been infected, one, four and eight days previously and all failed to become infected. The combined experiments (1932 and 1933) suggest that under some weather conditions, animals, although admitted to pasture within a very short time of its infection may escape infection with tuberculosis, but that repeated infection of pasture results in a high incidence of disease in animals grazing on it.

The importance of pulmonary tuberculosis as a cause of death in Wales has been the subject of much consideration. Many of the investigations, however, have been of a definitely local character, though this fact does not detract from their merit as is evinced by the work done by Wade, Bowen and the King Edward VII Welsh National Memorial Association. For a really comprehensive study we are indebted to the late Dr Brownlee. His contribution to the epidemiology of phthisis was unique in its suggestiveness. He postulated the theory that phthisis as a disease was, not an entity, but that it consisted of three types—“Young Adult,” “Middle Age,” and “ Old Age,” corresponding with separate strains of the infecting bacillus. According to his conception the great bulk of the phthisis deaths in Wales was due to early or late types of the disease. He reached this conclusion from a study of the pulmonary death-ratesof males between 1851 and 1870 in the registration districts of Wales.

1. This paper deals with the epidemiology of an outbreak of influenza which started in Leeds late in November 1932 and continued until the end of February 1933.

2. Returns of the actual cases which occurred each week in certain official departments and large firms made possible the investigation of a selected population. This consisted of 17,195 persons, and among its members there were 2772 cases of influenza with five deaths. The case-rate was therefore 16·1±0·2 per 100 of the selected population. It is shown that in this popu lation factory workers were less liable to contract influenza than were assistants in large shops, outdoor workers, clerks and office workers, and school teachers; and it is suggested that propaganda relating to respiratory infections may have had some effect in lessening the incidence of the disease among the workers in these factories. The mean duration of the outbreak in the selected establishments was 11·55±0·29 weeks.

3. The total cases which occurred each week in the selected population were plotted as a graph; the resulting curve was of the typical epidemic type. A Pearson's Type IV curve was fitted to these data. The available evidence points to the fact that the outbreak consisted of a single wave.

4. From the data for the city as a whole it is shown that the Leeds out break took approximately twice as long to reach its peak as did the Glasgow epidemic, though both outbreaks started practically simultaneously. Never theless, in actual epidemicity, the Leeds outbreak did not fall far short of that which occurred in Glasgow.

5. An investigation of the age distribution of deaths from influenza in Leeds since 1900 shows that, though the 1918 epidemic produced a sudden change in the age distribution, in that there was an unprecedented toll of young adult life, the return to the status quo was not sudden, as is often assumed, but was effected gradually over a period of about ten years.

6. It is shown from an investigation of influenza deaths in different parts of the city that there was no direct association between these deaths and the density of population in the wards in which they occurred. Neither was there any association between the number of influenza deaths and the average number of persons per room in each ward. From a study of the distribution of the deaths in time and space it appeared that, with certain exceptions discussed in the paper, the outbreak started in the centre of the city and extended outwards from the centre with decreasing potential.

7. An investigation of the meteorological conditions suggested that there was a definite relationship between a fall of temperature, especially if accompanied by fog, and an increase in the incidence of influenza and of pneumonia.

8. The clinical features of the outbreak are discussed fully in the text. The disease was essentially of the respiratory type. Two forms of this were in evidence, (a) a form resembling the classical picture of influenza, and (b) a form in which the predominant feature was tracheitis. The gastric type of influenza was not common. Complications were not very frequent, but the commonest were influenzal pneumonia and acute otitis media. The duration of incapacity for work varied according to the social status of the individuals. Brain workers and persons in the middle classes were generally off duty for about nine days. On the other hand, the duration of incapacity in artisans and factory workers was usually about a fortnight or three weeks.

9. From the evidence which is adduced in this paper as a whole, including a consideration of the tuberculosis statistics of the city, it would seem that, though the outbreak of influenza in Leeds during the winter of 1932–3 was of a mild and non-killing type, infection was widespread, and the total incapacity must have resulted in a considerable economic loss to the community.

I am indebted to Prof. J. Johnstone Jervis for suggesting this investigation. My thanks are also due to the heads of the various departments and commercial firms, and to the Staff of the Public Health Department, without whose co-operation the data for this study could not have been obtained.

It has been shown that S. schaudinnii will easily pass through Berkefeld V filters under a negative pressure of considerably less than one atmosphere, but B. pyocyaneus was always found in the filtrates in association with the spiro-chaetes. Attempts to pass S. schaudinnii through Berkefeld N filters were uniformly negative.

Inoculation experiments on native volunteers made with (a) filtrates containing S. schaudinnii and B. pyocyaneus, (b) a mixture of these organisms combined with B.fusiformis, and (c) an emulsion of B.fusiformis in broth, did not give rise to ulceration.