VIABLE BACTERIA IN YOUNG CULTURES 411 



The method of dilution 



Before proceeding further it will be necessary to give details of 

 the methods of dilution which have been employed in both the 

 total and the viable counts. For this purpose dropping pipettes 

 have been preferred to volumetric pipettes, chiefly because in 

 following out the growth of a bacterial culture such a large 

 number of volumetric pipettes are required that the price of the 

 latter becomes prohibitive : moreover dropping pipettes are more 

 convenient to handle and more accurate in their delivery. 



For full details with regard to the preparation of these pipettes 

 and for an account of the precautions to be observed in handUng 

 them, the original article by Donald (1915) should be consulted. 

 Suffice it to say that his technique has been used throughout with 

 the exception of one or two modifications. Thus, for the purpose 

 of cahbration, an Imperial Standard wire gauge has been 

 employed and, in order to insure accuracy, the pipettes have 

 been subsequently tested by means of an ordinary screw microm- 

 eter measuring to 0.01 mm. Further, the drop values have 

 been estimated by gravimetric, instead of by volumetric methods. 

 The actual size selected was no. 22 on the wire gauge. After 

 working out the drop volumes of various fluids at the temperature 

 at which they were to be used, one factor remained to be deter- 

 mined, namely the effect of variations in density of a bacterial 

 emulsion on the size of the drop deUvered. On investigating 

 this point, it was found, contrarj^ to expectation, that there was no 

 appreciable difference in the drop values of emulsions ranging in 

 thickness from 500,000,000 to 10,000,000,000 organisms per cubic 

 centimeter. The explanation of tliis would appear to be that 

 as the emulsion increases in density, the weight of the organisms 

 in the drop just suffices to counterbalance the increasing viscosity 

 of the Hquid, so that the actual value of the drop remains ap- 

 proximately constant. 



Errors of the dropping pipettes 



In order to determine the errors of the dropping pipette a con- 

 siderable number of experiments were made, the details of which 



