The prominence recently given to the Rous sarcoma has increased the confusion of hypotheses about malignancy. With a view to clarification, one may say that there are three simple hypotheses of outstanding importance, viz.: (1) the living virus hypothesis, which regards living viruses as the actual and effective cause of both the avian and the mammalian disease; (2) the autogenous enzyme hypothesis, which ascribes both diseases to the development within living cells of an ens malignitatis resembling an enzyme rather than a virus; and (3) the “chronic irritation” hypothesis, which explains mammalian malignancy on this principle and considers fowl sarcoma to be of a different nature.

Charts, illustrating observations over several years, have been published by Hamer (1908–10) for London, by Niven (1910) for Birmingham and by Hope (1920–21) for Liverpool, which show that in the third quarter of the year the summer diarrhoea death curve usually reaches its maximum about a fortnight after the curve representing the numbers of flies caught in traps has reached its highest point.

Typhoid fever is one of the prevailing endemic diseases of Palestine. In 1925 and 1926 epidemic outbreaks of considerable magnitude occurred in various parts of the country (Kligler, 1927). As a part of a general programme of the study of the epidemiology of this disease, we have attempted to ascertain the incidence of typhoid carriers in a general population group. Owing to the large number of typhoid cases which occurred there in 1925 and 1926, Afule, a village of some 1200 inhabitants, was chosen as a desirable spot for such an investigation.

A series of spontaneous epidemics of a paratyphoid infection among guinea-pigs is described. It is shown that an epidemic may result from the introduction of a fresh susceptible population into an infected area and vice versa. Such epidemics developed only when the new population was exposed during certain months. The bringing in of a fresh population at the end of November was not followed at the time by an epidemic outbreak; the epidemic did not occur until the following July.

The observed seasonal occurrence was not due to a change in virulence. The strains isolated at the end of the 1926 epidemic did not differ in virulence from those cultivated at the height of the 1927 epidemic.

The guinea-pig may remain a carrier for a long time whilst eliminating bacteria in the faecal discharges only intermittently and in small numbers. In general, even in the case of infected animals, the discharge of pathogenic bacteria is relatively small, except a few days before death.

Attention is called to the parallelism between the seasonal prevalence of human typhoid epidemics in this country and the epidemics of paratyphoid observed in our guinea-pig stock.

It is pointed out that a sound theory of the epidemiology of certain diseases must take into account the seasonal variation in host susceptibility.

In the December issue of this Journal Peller (1928) presented a critical analysis of my paper (1927). His criticism is both general and specific. The general criticism requires no comment on my part. Polemics have no place in English scientific literature. Any contribution which contains an idea and stimulates thought and investigation is, in my opinion, of sufficient interest to warrant publication, provided the data presented are accurate.

Some comment, however brief, is required because the specific remarks create an impression that not only are the deductions wrong, but that the data on which they are based are inaccurate.

In the present paper an attempt has been made to devise a method of determining mortality rates at ages from causes of death associated with childbearing. The sources of possible fallacy in this procedure have been discussed, and it is probable that the results found will represent fairly accurately the true course of mortality in this country.

From all causes of death in childbearing, mortality follows this course: beginning with a high value in the age group 15–20, the rate drops to a minimum in the second quinquennium of reproductive life and thereafter gradually increases with age.

The main components of the death rate exhibit a similar age distribution, but other less important causes of death show an uninterrupted increase with age from the commencement of the reproductive period.

Eclampsia is much more common in young women, but this may be entirely due to the greater proportion of primiparae at these young ages. Estimated in terms of first births the death rate increases rapidly and steadily with age, subject to a small exception in the first quinquennium which may possibly be a consequence of unfavourable marital selection at that age.

The high puerperal mortality of young mothers cannot wholly be accounted for by the excessive dangers of primiparity. Evidence has, however, been submitted that these young women are, on the average, a less healthy sample of the total female population at that age and that, apart from primiparity and adverse selection, the death rate from childbed causes probably follows the same law of mortality as of women in general from all other causes of death.

There has been within recent years a decline of nearly 50 per cent, in the puerperal mortality in young mothers of 15–20 and a smaller decrease in the second quinquennium of reproductive life; but there is no evidence of any improvement at later ages. These phenomena have been brought into relation with the change which has taken place in the death rate from all other causes in married women at these ages. The youngest group of married women, relative to unmarried of the same age, were in the more recent period studied in a much more favourable position with regard to their mortality from causes other than those connected with childbearing than they had been formerly. No corresponding improvement has taken place in the relative mortalities from all other causes in older women within the childbearing ages.

These results suggest that the general health of the mother is a factor of great importance in determining the height of the mortality rate in pregnancy.

(1) The unusual organism previously described (1929) has since been recovered from two additional sources, in each case children suffering from dysenteric symptoms.

(2) It is suggested that this organism be named the “Newcastle dysentery bacillus.”

(3) A more extensive study of its biochemical reactions, using Lemco broth (Dudgeon and Pulvertaft, 1927), as a basis for carbohydrates, has shown that these reactions are notably free from variation and that they exhibit decided divergences from those characteristic of certain bacilli often associated with, but not specifically related to, dysentery. They are more or less closely paralleled in the case of the Glässer-Voldagsen sub-group of the Salmonella, and particularly the Dammann strain of Voldagsen's bacillus, but nevertheless show quite definite differences.

(4) Further examination of the serological peculiarities of the Newcastle bacillus has demonstrated that it is difficult to prepare from it a very active rabbit anti-serum, that its velocity of agglutination is low, and that the titre obtained with saline suspensions is higher than that obtained with broth cultures.

(5) Complete absence of agglutination has been the result in tests with B. typhosus, many members of the Salmonella group, including all most frequently met with and also the Glässer-Voldagsen sub-group, and with a number of the bacilli often associated with, but not specifically related to, dysentery. On the other hand, with one strain of B. dysenteriae Sonne (3) there was a suggestion of agglutination and with at least four out of five races (V, W, X, Y and Z) of B. dysenteriae Flexner this was well-marked.

(6) Saturation with a mixed suspension of these five races completely removes all trace of these secondary agglutinins without materially affecting the specific agglutinins for the Newcastle bacillus while, on the other hand, saturation with each of the four strains of the latter has, in each separate instance, practically eliminated the specific and completely removed the secondary agglutinins.

Finally we desire to acknowledge the helpful criticism we have received from Prof. H. J. Hutchens throughout our study of this organism, the facilities placed at our disposal by Dr Harold Kerr, Medical Officer of Health for Newcastle, and the great assistance rendered by the Deputy Medical Officer of Health, Dr J. A. Charles.

This paper presents a summary of the work on Plague carried out in the Bacteriological Section of the Kenya Medical Research Laboratory, during the period August, 1926, to December, 1928, in connection with the preparation of prophylactic plague vaccine. It is divisible under the following sections:

A. Preparation of a vaccine of Haffkine type at Nairobi.

B. Tests on rats of the immunising potency of the vaccine.

C. Notes on the application of plague vaccine for human prophylaxis:

I. Experience in Kenya.

II. Discussion of the relative values of agar-grown and Haffkine's plague vaccines.

D. Notes on bacteriology and serology of Bacillus pestis:

I. Growth in fluid cultures under various conditions, etc.

II. Agglutinating sera and absorption tests.

41·7 per cent. of the rats in the Oxford district were found to carry Leptospira icterohaemorrhagiae in their kidneys. The percentage infection increases until the animals are mature, reaching a maximum of 56 per cent. among one class of the population, but is less among the oldest rats. There is evidence that the organism is acquired comparatively early in life, and that immunity is obtained after a long period of infection.