1. Earlier attempts to trace the vector of tropical typhus in Kenya failed. The only references to the subject in the available literature consist of mere suggestions that a mite would most likely prove to be the transmitter.

2. An investigation made in an area whence most Nairobi cases of tropical typhus were reported, suggested that a tick (R. pulchellus) would be the most likely vector.

3. Transmission experiments made in the belief that one of the unclassed fevers of man was conveyed by R. pulchellus have so far yielded negative results. There is, however, sufficient circumstantial evidence available pointing to this tick as vector of a form of mild typhus to man—this demands further investigation.

4. At Mombasa and Nairobi, houses reported to be heavily infested with ticks, or houses investigated after the occurrence of the tropical typhus in them, have yielded only R. sanguineus.

5. R. sanguineus (3 ♀), taken from a dog in a house where the last typhus case had occurred 8 months previously, gave a typical typhus syndrome when emulsified and inoculated into a male guinea-pig. R. sanguineus (1 ♀, 12 ⊙), taken in a house where a child had recently contracted typhus, also gave a positive result with guinea-pigs and the virus was further transmitted by passage through other guinea-pigs.

6. The infestation of houses by R. sanguineus and the incidence of tropical typhus among human beings appear to be influenced by unfavourable weather conditions, causing the ticks to seek relatively dry and warm places for purposes of oviposition or metamorphosis, thus invading houses. In the absence of dogs, its usual hosts, the tick attacks man.

The characters of the Aronson Streptococcus from Prof. Neufeld's laboratory at the “Robert Koch” Institute in Berlin have been described. This coccus resembles the pneumococcus in many respects, viz. the appearance of the colonies on the surface of horse blood agar, its virulence and capsule production in mice and rabbits, the production of a specific precipitable substance in the peritoneal washings of infected mice, the formation of firm clumps and masses when mixed with homologous antiserum, the ease of production of active and passive immunity in mice and rabbits by intraperitoneal and intravenous inoculation, the alteration in the morphology of colonies, i.e. the appearance of R forms, associated with attenuation of virulence. It differs from the pneumococcus in the following features: the round shape of the cocci, bile-insolubility and the absence of autolysis in surface colonies, the beta haemolysis of deep colonies in horse blood agar, the production of a soluble haemolysin in broth cultures, the difficulty of producing active immunity in mice by the subcutaneous injection of heat-killed vaccines.

I have obtained the Aronson Streptococcus (Neufeld type), which Lance-field places in a group containing chiefly streptococci of bovine origin, from human throats, but there was no evidence in any instance that it was producing disease, and it seems probable that it is not pathogenic for man.

The results of my investigation of this strain are in agreement with those of Yoshioka (1923), Killian (1924) and Lancefield (1933, 1934).

There are in existence other laboratory strains designated Aronson Streptococcus. These have been found to exhibit specific characters identifying them with the Str. pyogenes. It is proposed that the name Streptococcus Aronson should be confined to strains possessing the characters of Aronson N above described.

1. The causes of failure of the brilliant green enrichment method for isolation of B. typhosus are discussed.

2. It has been shown that the negative results obtained depend partly on the number of resistant B. coli strains normally present in faeces, and partly on the use of McConkey's as a plating out medium.

3. The alleged selective action of telluric acid could not be confirmed.

4. The so-called antagonism between B. coli and B. typhosus has been investigated, and by the use of a selective bismuth medium it is shown that this is more apparent than real in most cases.

5. However, the existence of certain strains of B. coli or similar bacteria with a specific inhibitory effect on B. typhosus has been confirmed.

6. The value of Wilson's bismuth medium for the isolation of B. typhosus has been demonstrated.

The problem as to the frequency with which multiple cases of, or deaths from, a disease may be expected to occur in the same house has been attacked by Troup and Maynard (1911–12), Karl Pearson (1911–12, 1911–12 a, 1913), McKendrick (1911–12), Greenwood and Yule (1920), Greenwood (1910, 1931) and others. Troup and Maynard obtained a general expression for the frequency of occurrence, and the standard deviation of the frequency, of s casesin a house when n cases of a disease occur in a town with m houses, assumingeach house to contain the same number of persons, all equally liable to contract it. Karl Pearson (1911–12) considered the problem as analogous to throwing balls at random into equally accessible pigeon-holes, and showed that since the chance that any one case will fall into any one house is , and that it will not is , the numbers of houses having 0, 1, 2, 3, ... cases should be given by the successive terms of the binomial . In a later paper (1913) he showed that the correspondence between the observed andexpected frequencies could be measured by applying the ordinary X2 test to the frequencies, omitting the houses having no cases.

With increasing knowledge of the effects of inhaled dusts upon the lungs has come an increase in the number of cases of Workmen's Compensation in which pulmonary disease is attributed to exposure to this form of risk during employment. Where a post-mortem examination has been made, and the opinion of a pathologist is sought, no difficulty is encountered in those cases of silicosis in which the disease is well developed, since the lesions are characteristic and easily recognised. The position, however, is by no means so easy where lesser degrees of pulmonary fibrosis are present and, in particular, where it becomes necessary to try to assess the bearing which these may have upon the susceptibility to, and the progress of, intercurrent pulmonary disease—for example lobar pneumonia. In these difficult circumstances recourse is frequently had to the chemical estimation of the quantity of silica recoverable from samples of lung. Certain standards have been set up, and the presence of silica in quantities in excess of the “normal” forms valuable corroborative evidence that a pulmonary fibrosis is attributable to its influence, where the subject in question has been exposed to this risk in his employment, and on the other hand the absence of an excess of silica is evidence that any pathological condition in the lung is due to other causes.

Of the chemical structure of antibodies little or nothing is known. It is generally held that the specific properties of an antiserum are closely associated with the globulin fraction. An antiserum may be distinguished from a normal serum by the precipitation of some of its globulins in the presence of the homologous antigen. The formation of a precipitate in a mixture of antigen and antiserum may be explained in either of two ways: (1) The reaction which occurs when antigen is added to antiserum may produce such physical changes in the mixture that normal globulin is precipitated. The reaction takes place between two specific substances, the antigen and the antibody, and is rendered visible by the presence of a non-specific globulin which plays the part of an indicator. (2) Or we may suppose that in the process of immunisation the antigen produces a specific alteration in the structure of the globulin. That is to say, the antibody is a specifically altered globulin, which has undergone some change in its molecular structure, in virtue of which it can react and combine with the homologous antigen.

Dufton And Bedford (1933) have denied the claim made by me (1932) that bright sources antagonise the “nose-closing” rays given off by dull red and dark sources of heat. They are supported by Winslow, Greenburg and Herring-ton (1934). These five authors have used the same method, viz. the insertion of a well-fitting tube in one nostril of the subject, the connection of this tube with a recording instrument, and the leaving open the other nostril for breathing, the mouth being kept shut.

Adverse atmospheric conditions, such as occur in mines and various industries, have been shown by numerous workers to have profound effects on both the health and efficiency of the industrial worker.

1. Among a series of strains of haemolytic streptococci from thirty-five cases of scarlatina in the first week of illness, fourteen were found to correspond with one or other of Griffith's serological types I, II, III and IV.

2. Of these fourteen strains, twelve were selected for further examination and found to yield a true heat-labile exotoxin completely inactivated by heating for 30 min. at 100° C.

3. The concentration of exotoxin in 0·5 per cent, glucose broth cultures was at a maximum after 96 hours' incubation, and thereafter on further incubation progressively diminished.

4. No qualitative difference could be detected among the exotoxins from the different strains, the test criterion being the dermal reaction in Dick-positive persons.

5. Cultures of organisms of the same or different serological type isolated from the same source and thereafter similarly treated yielded approximately equivalent amounts of exotoxin.

6. Broth culture filtrates also contained an acid-insoluble toxic fraction, the concentration of which increased with the age of culture and which appeared to be identical with a similar acid-insoluble fraction derived from an alkaline extract of washed bacterial bodies.

7. This acid-insoluble fraction was extremely heat-resistant, 3 hours' boiling at 100° C. being required for inactivation. In this respect the acid-insoluble fraction corresponded to the bacterial endotoxins.

8. The acid-insoluble fractions from cultures of the same serological type produced equivalent skin reactions in susceptible persons.

9. The acid-insoluble fractions from cultures of different serological types differed qualitatively as determined by skin reactions.

10. The reaction to crude filtrate was found to be the sum of the reactions to the exotoxin fraction and to the acid-insoluble fraction present in the filtrate.

From an outbreak of very mild “typhus-like disease” in Pretoria, which strongly resembled the cases of “sporadic” or “mild” typhus occasionally occurring in that town, a virus was isolated which was studied in guinea-pigs and rabbits, and was found to belong to the typhus-group.

This virus in cross-immunity experiments with virus of South African louse-typhus and South African tick-bite fever and also in other respects behaved exactly like the typhus-like virus the authors on a previous occasion isolated from rats in a South African town where cases of “mild” or “ sporadic ” typhus have been known to occur regularly for many years.

This constitutes evidence that South African “mild” or “sporadic” typhus comes from rats and is communicated to man by rat-fleas.

In southern Africa, therefore, one has to reckon with three typhus-like diseases: tick-bite fever, flea-typhus from rats, and louse-typhus. Tick-bite fever has a primary sore, the tick-bite, as a pathognomic symptom, the other two can only be differentiated satisfactorily from one another by cross-immunity tests.

Agglutination reactions and cross-immunity tests show that South African louse-typhus and South African rat- or flea-typhus are not identical with similar diseases in other parts of the world.

Tick-bite fever is a mild typhus-like disease extending from the Cape to South Rhodesia.

The typhus-like disease of Kenya is not tick-bite fever, and seems to be identical with fièvre boutonneuse.

Mineral oils recovered from the peritoneal cavity of injected mice showed in all sixty-one instances, with the exception of liquid paraffin, a fall in the refractive index. This phenomenon is presumably due either to a chemical change (? oxidation or reduction) in the constituents of the oils brought about by the juices of the animal organism, or to a mechanical change resulting from the removal of selected constituents to other parts of the animal. This test will possibly prove to be superior to that of skin applications and estimation of the refractivity constants for measuring the amount of carcinogenic constituents in mineral lubricating oils and it has the advantage of necessitating the use of only one animal instead of 100. We are of opinion that our experiments may be of far-reaching importance. They carry us a definite step nearer to the understanding of the mechanism of cancer formation.

1. A mixture of anhydrous lanolin with about equal parts of olive oil was the most efficacious ointment tested for protecting our animals from mineral-oil dermatitis and cancer.

2. Some commercial products consisting of mixtures of the alcohols and esters contained in lanolin were, on the whole, less efficacious: possibly the apparent benefit was mostly due to the olive oil with which they were diluted.

3. Olive oil, glycerine, commercial soaps, etc., gave varying degrees of protection, the last being particularly useful under some circumstances.

4. In experiments with gas tars, lanolin does not appear to afford protection when relatively small quantities of it are mixed with the tar or when relatively large doses of tar are applied to the animal, before or after lanolin treatment.

5. Where the experimental conditions appear to conform more to those prevailing among most tar workers, lanolin has a definite protective action.

6. Separate application of the carcinogenic agent and the prophylactic agent as a rule results in a lower yield of tumours than applications of an admixture of the two. In this respect it is to be noted that, where the animals were treated with lanolin separately, the quantity applied was several times greater than that of the tar itself (similarly in mineral oil experiments). This, of course, was not the case in our experiments with the lanolin blends.

7. Relatively more lanolin is required to protect against gas-tar dermatitis and cancer than is required to protect against toxic oils or synthetic tars. This is probably due to the presence in gas tars of special inhibitory substances. Note the analogy in the action of small quantities of acid on mineral oils, the potency being raised, whereas larger quantities may lower it.

The infectious fevers have, during the last fifty years, been the subject of much epidemiological research, and have thus been made to yield up secrets which have materially affected the whole outlook of preventive medicine. But there yet remains for exploration a vast adjacent territory, that of minor coccal infections of the skin. However, whereas the former, being clearly defined clinically and in many cases subject to compulsory notification, lend themselves well to investigations of this sort, Impetigo, as a type of minor coccal dermatitis, surrounds itself with dificulties. For not only does it vary very much in its appearances by the time it is first seen but it is often of so little severity that it does not come under the observation of a physician at all, so that the influence of missed cases, always an important factor to be reckoned with in the study of epidemics, may assume such proportions as to give quite an erroneous impression of the total incidence of the disease. Consequently the epidemic aspects of impetigo have as yet received very little attention.

Pp. 474–485. Messrs W. Martindale inform us that the Lactic Acid Pessaries reported upon by Dr John R. Baker in the Journal of Hygiene, Vol. 34, pp. 474–485, are not recommended by them as a contraceptive. They are made to the order of a clinic. (Editor.)

P. 528, line 26, for “glucose media, and did not affect saccharose and maltose media” read “glucose and maltose media, and did not affect saccharose media”