82 THE MECHANISM OF AGGLUTINATION, 



to an equal volume of a suspension of bacteria in water. This 

 appears to be due to a dehydration rather than to a flocculation 

 in which the loose water molecules are withdrawn. But such as 

 it is, these strong reagents produce the neai'est approach to true 

 flocculation that can be obtained with bacteria. Another of his 

 agglutinating agents is dilute mercuric chloride. This salt 

 undoubtedly acts like calcium chloride in producing a precipitate 

 of the culture salts that is flocculated together with the bacteria. 

 The case of a dilute solution of saffranin promised to be different. 

 This stain when in dilute solution (1-1000) and added to an equal 

 volume of bacterial suspension produced an apparent agglutina- 

 tion of the bacteria. A test of the stain, however, with sterile 

 bouillon showed the formation of an immediate precipitate which 

 was found microscopically to resemble clumps of bacteria and 

 cocci. This shows that the case of saffranin is no exception to 

 the rule that in agglutination a precipitate is first formed in the 

 fluid. On investigating the constituent of the bouillon that is pre- 

 cipitated by saffranin, it was found to be among those that are 

 precipitated by lime, since no agglutination was obtained with 

 bacteria that had been grown upon or in media that had been 

 treated with lime to remove phosphoric acid. It does not appear 

 to be a phosphate, because neither ammonium nor potassium 

 phosphate forms a precipitate with the dilute stain. Malvoz 

 ascribes the coagulating effect of dilute alkalies to the formation 

 of calcium carbonate. I have already shown that is due to the 

 formation of calcium phosphate. 



In conclusion, it appears that agglutination is caused by the 

 formation of a delicate precipitate on the outer surfaces of the 

 bacteria and in the fluid in which the bacteria are suspended. 

 This precipitate is flocculated or coagulated by the saline con- 

 stituents of the medium and of the serum. Since the precipitate 

 is invisible to ordinary microscopical observation and the bacteria 

 are visible, an apparent agglutination of the latter only is seen 

 to take place by the action of active sera. 



