1909.] Ahsorption" of Specific Agglutinins hy Bacteria, etc. 175 



Somewhat similar results are found in Experiment 2 (Table IV) carried 

 out at 2° C, where, out of 162 units, there were left after 0 minutes 

 12'5 units, after 4 minutes 97, and after 20 minutes 6'06. 



It is thus clearly seen that at temperatures but little higher than 0° C, 

 equilibrium is by no means established even in times of more than 

 30 minutes. This is in no sort of agreement with the statement of 

 Eisenberg and Volk which has been generally accepted by Arrhenius, Craw, 

 and others. 



As it now was of interest to see whether equilibrium would be established 

 in a short time at higher temperatures, such as 15° to 17° C. and 37° C, 

 Experiments 3 and 4 (Tables V and VI) were carried out. In Experiment 3, 

 carried out at 37° C, the serum diluted 12'5 times contained 29-2 units. 

 After 0 minutes were left 3-49 units, after 9 minutes 2-27, after 30 minutes 

 2 units, but that equilibrium had not been reached even after half an hour is 

 seen from the fact that after 2, 4, and 8 hours there has been left res^jectively 

 1"85, 1'78, and 1'59 units. Exactly the same is found in Experiment 4, 

 carried out at 16° C, as will be seen from Table VI, save that the velocity of 

 reaction is slower than in Experiment 3. 



To show with how great accuracy experiments of this kind can be carried 

 out, and to prove beyond doubt the correctness of the statements, in each of 

 the Experiments 3 and 4 there were put up three tubes for the time denoted 

 as 0 minutes, and two for the time denoted as 9 minutes. It is clear that 

 in Experiment 3, where the greatest individual differences between the 

 observed values are found, the mean variation is, nevertheless, not greater 

 than 3"1 per cent, from the mean value ; and for 9 minutes the two 

 determinations are identical. In Experiment 4, for the 0 minute point the 

 mean variation is less than 0'9 per cent, of the mean value, and at 9 minutes, 

 just as in the previous experiment, the figures are identical. As one would 

 expect, the greatest error is found where the time reaction is the shortest, 

 iind is due to the inevitable small differences in time which can hardly be 

 avoided. Similar control tubes were used in Experiments 7, 8, 9 

 (Tables IX, X, XI). 



These experiments clearly show that one of the fundamental points stated 

 by Arrhenius to be in good agreement with his theory that the absorption 

 of agglutinin by bacteria should be considered as following a partition law, 

 C = A;B", similar to that which holds good for the distribution of benzoic 

 acid between two different solvents, water and benzene, is unsound, since the 

 reaction, instead of being a very rapid one (finished in less than five minutes 

 at 0° C, Eisenberg and Volk), is a slow one and is still incomplete at room 

 temperature even after several hours have elapsed. 



