viii AGGLUTINATION OF B. PESTIS 207 



slightly turbid. As in other similar experiments on 

 agglutination, too great turbidity, i.e. too thick an 

 emulsion, is to be avoided. While in most instances an 

 emulsion is obtainable in this manner, in others it is not 

 possible to get rid of small aggregations of the bacilli. 

 No amount of shaking can dissociate these, and such an 

 emulsion is for obvious reasons not of sufficient reliability 

 for application of the agglutination test. .When such is 

 the case I adopt the following plan for making a workable 

 emulsion : I make (a) a thick emulsion and filter this 

 through a double filter-paper, by which means all the 

 larger aggregations are kept back and only the isolated or 

 fairly isolated or very minute groups are let through ; or 

 (b) I take a particle of growth from the gelatine culture 

 tube and rub it over the slanting surface of a fresh gelatine 

 tube, after which a few cc. of salt solution are poured over 

 this new surface and the tube slightly shaken till the fluid 

 has worked off the matter from the surface. 



I have made also an extensive series of observations 

 with regard to the mere sedimentation of emulsions (broth, 

 salt solution, water) of B. pestis in tubes, in capillary 

 pipettes, etc. As a result I have found that any conclu- 

 sion as to positive sedimentation and agglutination result- 

 ing from addition to them of blood has no value whatever, 

 and for the simple reason that some (in fact many) 

 emulsions of plague bacilli sediment in such tubes and 

 capillary pipettes spontaneously without the addition of 

 anything. Similarly, they sediment after the addition of 

 various indifferent fluids, e.g. normal blood serum, aqueous 

 humour, strong salt solution, peptone solution. All these 

 cause the bacilli to settle down to the bottom of the fluid, 

 which itself becomes quite limpid. I would go so far as 



