IDENTIFICATION OF BACTERIAL SPECIES 7 
the agglutinin is used up during its action on the bacilli, this being one reason for always 
employing a constant quantity of culture. Consequently, it appears possible that a rapid increase 
in the number of micro-organisms, which may easily occur if they light on some fresh suitable 
soil, would use up the available agglutinin and reduce the amount temporarily below what it had 
been the day before, and may reduce it below the level of normal blood. And it explains why 
the most severe cases of typhoid may have comparatively little agglutinin in tlieir blood. 
By the aid, therefore, of a known culture of typhoid bacilli, and by suitable dilution of the 
serum to be tested to avoid any error due to the presence of normal agglutinin, it is possible to 
diagnose enterica by the agglutinin contained in the patient's blood. It is obvious that the 
principles of serum diagnosis apply to other diseases as well as enterica. The method has indeed 
been used for the diagnosis of cholera, Malta fever, and glanders. In diphtheria and tubercle it 
has not yet been successful. It might very well be applied also to the diagnosis of pneumonia, 
erysipelas, appendicitis, and influenza, if required, since it is not necessary that the specific organism 
should be motile. From a case of scarlet fever I isolated a coccus which agglutinates with scarlet 
fever serum {'jb). 
One general precaution must be mentioned regarding the serum diagnosis of any disease. 
Although absurdly obvious, neglect of it has already led into error, namely, that the test is not 
reliable if injections of ' antitoxic ' serum have been made and agglutinins thus artificially 
introduced. 
Not only may the agglutinin reaction be applied to the diagnosis of disease already present, 
but is also likely to be of use in the discovery of specific micro-organisms as yet unknown, e.g.^ 
small-pox, typhus, measles, etc. For this purpose bacteria must be isolated from the various tissues, 
and their reaction with the serum of patients convalescent from the disease in question ascertained. 
Naturally, any positive result must be controlled with other sera, and by suitable dilution of the 
specific serum. 
There is one source of fallacy against which it is necessary to guard. Some infectious 
diseases, e.g., scarlet fever, and small-pox in its later stages, are nearly certainly cases of mixed 
infection, more particularly with the organisms of sepsis. Under such conditions more than 
micro-organism might give a reaction with the serum, and other criteria would have to decide 
which is the specific one. Partly for this reason I do not consider the coccus above-mentioned to 
be the specific organism of scarlet fever. 
Without Durham and Gruber's exact and complete investigation and extension of the 
older isolated observations, a valuable addition to both practical bacteriology and to practical 
medicine would have been, if nothing else, for some time delayed. The application of the serum 
test to disease in Man is but the natural outcome of their work. Being now part of routine 
diagnosis it is past the stage when the name of any individual should be attached to it, but 
in the history of this new departure in diagnosis their names have the first claim to be 
mentioned. 
October, 1897. 
