106 BACTERIOLOGY. 



for each sample deviate but very slightly or not at all 

 the one from the other, the mean of these amounts is 

 taken as the amount of alkali necessary to neutralize 

 the quantity of bouillon employed. If 10 c.c. of bouillon 

 were employed, then for the whole amount of 1 litre 

 just 100 times as much, minus that for the two samples 

 used in titration, will be needed. For example : to 

 neutralize 10 c.c. of bouillon 2 c.c. of the diluted (0.4 

 per cent.) caustic soda solution were employed. For 

 the remaining 980 c.c. of the litre of bouillon, then, 

 196 c.c. (200 c.c. less 4 c.c., the amount employed for the 

 two samples of 10 c.c. each of bouillon) are needed of 

 the 0.4 per cent, solution, or one-tenth of this amount 

 of the 4 per cent, caustic soda solution. 



For the neutralization of the whole bulk of the 

 bouillon it is better to employ the stronger alkaline 

 solution, as by its use the volume is not increased 

 to so great an extent as when the dilute solution is 

 used. 



It is evident that this method is much more exact 

 than that ordinarily employed ; but at the same time it 

 must be remembered that for its success exactness in 

 the measurement of the volumes and in the preparation 

 of the dilutions is required. To obviate error, it is 

 better to employ this method when the solutions are all 

 cool and of nearly the same temperature, so that rapid 

 fluctuations in temperature, and consequent alterations 

 in volume, will not materially interfere with the accu- 

 racy of the results. 



This method of neutralization, as suggested by 

 Schultz, should be adopted for experiments in which it 

 is desirable to have the reaction of the medium accurate 

 and constantly of the same degree. 



