Immediately following on the International Plague Conference held in Mukden in April, 1911, the Chinese Government, anxious to carry out the recommendations of the Conference, instituted the North Manchurian Plague Prevention Service. The chief of these recommendations were briefly:

(a) Systematic investigations should be made as to whether epizootic plague occurs among Tarbagans and other rodents, and, if such exists, an accurate investigation should be made of the nature of the infection.

(b) A general improvement in the sanitary condition of cities and villages, especially with regard to overcrowding, is desirable.

(c) Education of the public by lectures and the issue of pamphlets and handbills, explaining preventive measures in simple language.

(d) The need for isolation of pneumonic-plague patients being urgent, permanent hospitals should be available.

(e) Hospital accommodation for suspected cases of plague should be provided.

(f) Contact quarantine stations should be constructed.

Of the many methods employed to render a serum inactive, that by means of mechanical agitation has been the subject of several recent works, but the real nature of this phenomenon, as it occurs in shaken sera, is still unknown. The reason is the uncertainty regarding the nature of complement itself, and even the most satisfying explanation of the inactivation of complement by shaking would not probably throw much light on the nature of the complement. I give in the following a brief summary of the literature dealing with the inactivation of complement by mechanical agitation.

In recent years immunity reactions have come to be regarded more particularly from the physico-chemical standpoint. Many of the phenomena occurring in immunity reactions are shown to be of the nature of colloidal chemical processes, and the laws which govern such processes have been found to hold, at least partially, for reactions occurring in sera, which belong to the class of colloidal protein solutions. Thus the application to immunity reactions, of such conceptions as viscosity and surface tension, has afforded a more reasonable explanation of these reactions. In the following paper special reference will be made to the surface tension, changes of which play an important part in all immunity reactions.

(1) Brilliant green exerts an inhibitory effect on the growth of bacilli of the coli group commonly occuring in faeces, which is in general more marked than its action on B.typhosus and paratyphiod bacilli.

(2) By taking advantage of this property of brilliant green a method has been devised is the inoculation B. typhosus from faeces. The procedure adopted is the inoculation of a series of tubes of peptone-water medium containing varying amounts of brilliant green, incubating for 20–24 hours, and then inoculation on a suitable solid medium from each tube.

(3) The reason for employing a series of concentrations of brilliant green is that the optium concentration for the growth and isolation of B. typhosus varies from case to case, depending probably both on the proportion of typhoid bacilli present and on the number and character of the accompnying bacteria as well as on the organie faecal material.

(4) The method is very easily and rapidly carried out.

Forty-one rats examined—internal organs and intestinal contents—for presence of Gaertner group bacilli. Bacilli identical with B. enteritidis isolated from five rats in each case from the spleen (in two from liver also). No members of this group isolated from the intestinal contents.

Several of the rat sera were capable of agglutinating Gaertner group bacilli in high dilution. These facts point to the view that old infection with Gaertner group bacilli had taken place.

The general result of the investigation is in favour of the view that while rats are lible to be infected with Gaertner group bailli and to be ill in consequence these bailli are not natural intestnal inhabitants. If this be accepted it may be stated that this group of bacilli are not natural intestinal inhabitants, of any known animal species.

Rats infected with Gaertner group bacilli may serve as a means of infecting meat with these bacilli and may possibly in this way be a cause of meat poisoning outbreaks.

(1) From a horse injected with sheep's red cells three months prior to the first bleeding and in a condition of constant haemolytic titre 122 litres of blood were taken in a period of 11 months. The horse's condition remained good throughout and has markedly improved during the year.

(2) The net result was that the haemolytic titre during that time fell only to about 66% of its original value, the leucocytes to about 66%. haemoglobin scarcely at all, while the specific gravity of the blood and total protein have increased, the former by 10% the latter by 5%;.

(3) There was no relationship between the total number of leucocytes and the amount of antibody. The differential count showed an increase of 12% in the polymorphonuclear and a decrease of 12% in the mononuclear cells, these figures being not very far outside the experimental error. The eosinophile and mast cells showed no marked alteration in number, size or staining reactions.