While investigating variations in agglutinability, occurring among certain strains of paratyphoid bacilli, it became desirable to obtain cultures derived from single bacterial cells. The technique at first adopted was that devised by Mutch (1919), but the results obtained were not satisfactory. Certain modifications of this method were attempted, but were no more successful.

The main difficulty, and one which appears to us to have been far too little emphasised in all the methods hitherto described, has been encountered in obtaining satisfactory conditions for the microscopical observation of the bacillary suspensions employed. The difficulty of accurately observing hanging-drop preparations, and still more of controlling manipulation of them by microscopical observation, would seem to us to have been insufficiently realised; and the ease with which faulty interpretation of microscopical appearances may introduce fatal errors into work of this kind has not, perhaps, been fully taken into account.

In a series of investigations on the epidemic spread of bacterial infection among mice (Topley, 1919, 1921), it has repeatedly been noted that, if an epidemic be started among a mouse-population by feeding certain animals on cultures of B. gaertner and subsequently introducing other susceptible mice into the cage, a varying proportion of these latter may succumb to infection with an organism morphologically and culturally identical with B. gaertner, but differing sharply from it in regard to its agglutination reactions. This organism has been referred to in previous communications as belonging to the B. suipestifer group, using that term to include B. aertrycke and other closely related types.

There seems to be considerable evidence in favour of the view that the immediate cause of this acute infective fever is an organism in the form of a streptococcus although there does not yet appear to be unanimity as to the specific characteristics of the latter. With regard to the secondary or indirect or predisposing causes of the complaint, the agencies which influence favourably the development and activity of the organism and consequently the prevalence of the disease, there has been a definite tendency in recent years not only to abandon the hitherto prevailing opinion on the influence of certain meteorological factors, namely, that increase in the amount of rainfall and decrease in the temperature are generally associated with increased incidence of the disease but even to assert that the action of these factors is the reverse of this.

1. The organism of Trench Fever is present in the “whole blood.”

2. The evidence seems to be in favour of the organism being an extra-corpuscular rather than an intra-corpuscular parasite.

3. The smallest quantity of whole blood which has given rise to the disease was 0.5 c.c.

4. The blood is infective from the first day of the disease, and in one case was found to be still infective after 443 days.

5. The microscopical examination of the blood has not up to the present revealed the organism or at least has not differentiated it from other granules present in blood.

6. There is some evidence that the organism may leave the body in the sputum and urine, but the chief and only way which has any practical significance is by means of a blood-sucking insect, the louse.

7. There is no evidence that infection takes place through food, drink or air, but only by inoculation of the organism by means of this insect.

8. The chief method of infection is not, as might be expected, by the bite of the louse, but by the infected excreta of the louse being brought in contact with an abraded surface of the skin.

9. After a louse has fed on a case of Trench Fever, five to nine days elapse before the excreta become infective.

10. If lice are fed on a case for some time a high percentage of the lice become infective.

11. There is no transmission of the organism of Trench Fever from infected lice through the egg to their offspring.

12. The tenth of a milligram of excreta has been found sufficient to set up the disease.

13. The organism of Trench Fever retains its virulence in louse excreta for at least four months.

14. The virulence of lice excreta is lost by exposure to dry heat of 100°C. for 20 minutes; to moist heat of 60°C. for a similar time.

15. It is doubtful if any of the laboratory animals are susceptible to Trench Fever.

16. It would appear that most men are susceptible to the disease, only a very few individuals having been found apparently naturally immune.

17. One attack seems to produce only a partial and limited immunity.

18. In regard to the etiology of Trench Fever there is strong evidence that the Rickettsia-like bodies found in infected louse excreta are the micro organisms of the disease, but as they have never been cultivated outside the body the final proof is still wanting.

19. The organism cannot be said to be a filter-passer; the evidence would seem to be more in favour of its being a small body such as the so-called Rickettsia.

1. That there is reason to believe that sporadic cases of cerebro-spinal fever occurred in Hong Kong previous to the outbreak of the epidemic in 1918.

2. That the epidemic followed a widespread infection of influenza colds.

3. That the mortality out of a total of 1040 cases was 85–48 per cent.

4. That the most susceptible age was found to be the age period under five years.

The younger the individual, the greater the susceptibility, with the added proviso that the extremes of life suffered most.

5. That once the epidemic was established, a drop in the temperature, steadily maintained, accompanied by an increase in the saturation of the air as regards moisture, associated with a lack of sunshine, was found to be followed after a lag of a few days, by a rise in the number of cases. This lag showed a tendency to increase as the epidemic progressed, from three to four days at the beginning to six to seven days at the end of the epidemic.

6. That the number of cases in any given area varied directly with the overcrowding in that area, and this bears out the contention that overcrowding is one of the most important features in the epidemiology of cerebro-spinal fever.

7. That the housing conditions in Hong Kong are such as to necessitate immediate action on the part of the Government, to safeguard the public health of the colony.

8. That new entries to Hong Kong formed a large proportion of the cases, and that this appears to be an important factor in the epidemiology of the disease.

9. That the poorest, most hard worked and badly housed portion of the community suffered most in the epidemic.

10. That the recognition and isolation of carriers is impracticable in an epidemic of any considerable dimensions.

11. That in the present condition of the knowledge of the means by which infection is spread, the wearing of a screen composed of some impervious material such as celluloid would seem to afford complete protection to the wearer against droplet infection.

12. That known carriers should be compelled to wear this screen until such time as they are proved free from the meningococcus. Thus protected, the carrier could pursue his usual avocation and the community be safeguarded from infection at a minimum of expense.