JOHN H. NORTHROP AND PAUL H. DE KRUTF 651 



fact that excess salt or acid confers a high potential upon the particles, 

 of opposite sign to that in low concentration. 



These various effects are all shown in the case of thorium chloride 

 (Fig. 2). In concentration below 5 X 10-« n no agglutination takes 

 place since the potential is greater than 15 millivolts (the organisms 

 being negative to the water) and the organisms are kept apart by the 

 repulsion due to this potential. In concentrations between 5 X 10~' 

 and 5 X 10"^ n there is agglutination, since in this range the potential 

 is less than 15 millivolts and the repulsion is therefore not sufficient 

 to overcome the cohesion. In concentrations of from 5 X 10~^ to 

 5 X 10"^ the potential is greater than 15 millivolts (though of the 

 opposite sign) and the suspension is again stable. At a concentration 

 above 0.05 n the potential drops below 15 millivolts but agglutination 

 does not occur since the cohesive force has also been reduced. A 

 smaller potential is therefore sufficient to prevent agglutination. At 

 a concentration of 0.10 n the potential is reduced practically to zero 

 and agglutination again occurs. In still higher concentration the 

 organisms are again stable due to a further decrease in the cohesive 

 force." The hydrochloric acid curve is interesting in that it shows a 

 zone of agglutination in concentrated solutions (> 0.3 n). This is 

 due to the sudden increase in the cohe&ive force at this point as is 

 shown in Fig. 9. This does not occur with the other chlorides and in 

 the latter solutions no agglutination occurs in this range. 



The stabilizing effect of sodium chloride in high concentration is 

 shown more strikingly in Fig. 10,^^ which gives the result of adding 

 increasing salt on the acid agglutination zone. The addition of 



"According to O. Porges {Centr. Batk., 1 te Abt., Orig., 1906, xl, 133) agglu- 

 tination occurs again in very strong salt solutions such as half saturated 

 (NH4)2S04. This is probably a salting out phenomenon, due to a decrease in the 

 forces between the surface of the particle and the liquid. For a review of the 

 effect of salts on agglutination see Buchanan, R. E., /. Bad., 1919, iv, 82. The ex- 

 periment itself shows that this is a different phenomenon since in saturated 

 (NH4)2S04 agglutination occurs immediately whereas the type of agglutination 

 studied in this paper requires considerable time. 



^^ It will be noted that in this experiment the isoelectric point was about pH 

 4.2 while in others with B. typhosus (Fig. 3) it is about 3.5. This difference 

 was noted several times and depends probably on the age and condition of the 

 suspension. 



