A Microscopical study of the skins of mice painted with definite carcinogenic agents and with agents which appear to be incapable of exciting the production of tumours has, in the main, enabled us to confirm the observations of previous workers on the different histological changes which take place. A brief description of our findings will first be given.

During a recent investigation on the carrier problem as it occurs in diphtheria, which was carried out (under the auspices of the Scientific Advisory Committee of the Metropolitan Asylums Board) at the Northern Group Laboratory, North-Eastern Hospital, one of the first subjects considered was that of the bacterial flora of the carriers under examination, and of an adequate series of controls. It was suggested that possibly the flora of the carriers might differ (apart altogether from the presence of the Klebs-Loeffler bacillus) from that found in non-carrier persons of the same age. It was thought that some pathogenic organism might be found to be constantly present in the nose or throat flora of carriers and not in controls, and that this might be exerting an influence on the Klebs-Loeffler bacillus by allowing it to survive in the infected region for prolonged periods. If such were the case it was hoped that an elimination of the organism or organisms concerned would go a long way towards clearing up the carrier condition.

1. Bacteriological examination from the tonsils and post-nasal space in forty-six cases of acute and sub-acute rheumatism resulted in streptococci alone being isolated.

2. The streptococci isolated fell into three groups, S. mitis, S. salivarius, and eleven strains of S. cardioarthritidis.

3. Comparison of the strains of S. cardioarthritidis with strains isolated in America show them to be closely related.

4. No evidence was found of an exotoxin.

5. Evidence was found of the presence of an endotoxin or lysate which produces an antitoxin.

6. Intravenous inoculation of the lysate into rabbits appears definitely to affect the heart muscle and lungs, and to some extent the liver and kidneys.

1. The bacteriological examination of the throat or nose in 100 cases of scarlet fever on admission to hospital showed the following incidence of the different serological types of streptococci. Of 91 cases yielding positive results on the first examination, haemolytic streptococci of Type 1 were found in 8, Type 2 in 14, Type 3 in 26 and Type 4 in 12; the remaining 31 fell into the heterogeneous group. In the 9 patients who gave a negative swab on admission a later test showed Type 1 in 2 cases, Type 2 in 4 cases, Type 3 in 1 case and heterogeneous strains in 2 cases.

2. The heterogeneous group comprised individualistic strains, none of which occurred with sufficient frequency to form a fifth type.

3. In 50 out of the 100 cases examined at weekly intervals the type of streptococcus remained the same throughout the attack.

4. In 50 cases there was a change of type during the course of the disease; a second type appeared in 38, a third in 10 and a fourth in 2.

5. The change of type was considered to be a reinfection since it appeared most often in the youngest children and after the 3rd week of treatment; the source of reinfection was traced in two instances.

6. In 3 patients, in whom the Dick reaction remained positive, the appearance of the new type coincided with a second attack of clinical scarlatina.

7. Of the 50 patients in whom no change of type was demonstrated 20 still carried the haemolytic streptococci on discharge after periods in hospital ranging from 4 to 8 weeks.

8. Of the total 100 cases there were on discharge 49 throat carriers of haemolytic streptococci and 6 nose carriers. None of the 55, so far as was ascertained, gave rise to “return cases” of scarlet fever.

9. There was evidence of correlation between serological type of streptococcus on the one hand and severity of scarlatinal attack and occurrence of complications on the other. The importance of this in hospital practice is emphasised.

10. The Dick test and Schultz-Charlton reaction were carried out on all cases in the series; the results indicated that the four serological types have different immunological values.

11. The evidence in favour of qualitative as opposed to quantitative differences in toxin production by the different types is discussed.

For some years one of my principal interests has been to study, in collaboration with my colleagues, E. M. Newbold, W. W. C. Topley and J. Wilson, the epidemiological phenomena observable in communities of mice exposed to risk of infection. In this work we have often found it convenient to summarise part of our results under the form of what we call a Life Table. We treat the entrants to our communities as live-births and determine the rate of mortality prevalent in the community over a certain epoch for each day of life, in the familiar life-table form. The result has a formal resemblance to a life table, but, from the nature of the case, it can have little or no bearing upon the course of events in a normal community. Naturally, however, one desired to set up some normal standard of mortality for the animal species used and, having obtained some scanty data, one was led to speculate further upon the biology of the so-called “laws” which have been from time to time proposed to describe the course of mortality in man. In this paper, I have brought together the imperfect results of such study as I have been able to make. Their practical value, from the point of view of the description of human mortality, is, it need hardly be said, negligible, while as a contribution to the history of “laws” of mortality what is omitted is perhaps as important as what is discussed. I have, however, felt justified in printing this essay in the hope that it might be accepted as a tribute in piam memoriam of my friend and colleague John Brownlee. The title of one of Brownlee's papers—“The Biology of a Life Table”—was the inspiration of much of his life-work. He approached the problem with an erudition, both biological and mathematical, to which I have no pretensions and, had his power of exposition been equal to his natural sagacity and learning, there would have been small need of any other writer.

A given strain of Bact. aertrycke Mutton has been tested repeatedly for its virulence to mice, and on some of these occasions the virulence of 10 single colony cultures taken from this strain has likewise been tested. Between these single colony cultures such marked differences in virulence have been found as to constitute definite discontinuous variations. Side by side in the same culture there have been found virulent and avirulent organisms. Daily subculture in broth under certain atmospheric conditions resulted in the fall in virulence of the whole culture; this was accompanied by a replacement of the virulent organisms by organisms that were either completely avirulent or were only weakly virulent. The evidence suggests that the fall in virulence of the whole culture is not due to a simultaneous fall in the virulence of each of its constituent organisms, but to a replacement of the highly virulent organisms by organisms of a lower degree of virulence. During the process of replacement two or three different variants, showing discontinuous variations in virulence, may be demonstrated together in the same culture. The conclusions to be drawn from these findings, and their bearing on the interpretation of the results of experimental epidemiology, are discussed.

No information from official or non-official sources is available on the prevalence of enteric fevers in Palestine in pre-war days. Since the establishment of the British Administration (1919) the Annual Reports of the Government Department of Health invariably emphasise the high incidence of enteric fevers. Several papers dealing with the epidemiology of the disease appeared in the Hebrew periodicals Briuth Haam and Harefuah and in 1926 an elaborate Report was submitted to the Department of Health by a Committee of Enquiry appointed by the Department to consider and advise on the situation as regards enteric fever in the country. The data thus accumulated, although very valuable for the local health authorities, did not reveal any peculiar features of more than local interest. Their publication abroad seemed, therefore, superfluous.