COUNTING YEAST CELLS. 57 



The sample taken should always be a fair average. In 

 most cases it must be diluted and thoroughly agitated for 

 a long time, in order to obtain an equal distribution of the 

 cells ; the specific gravity of the liquid must also be such 

 that it will allow the cells to remain suspended in it for a 

 short time. A small drop is then withdrawn in a capillary 

 tube, transferred to the counting apparatus, and covered with 

 the cover-glass. The apparatus is allowed to remain at rest 

 for some time, in order that the cells may settle to the 

 bottom of the enclosed space, and on this account the specific 

 gravity of the liquid must not be greater than will allow this 

 to take place in a convenient time. Both these require- 

 ments are generally satisfied by the wort employed in 

 breweries. 



If it is found that the determinate volume contains too 

 many cells to be counted with certainty, the liquid must be 

 diluted. This may be advisable for other reasons, partly to 

 prevent the formation of froth, which may otherwise form 

 abundantly from the violent agitation, and partly to isolate 

 the single cells which frequently cluster as colonies in the wort, 

 and are not always separated by shaking. Finally, it is 

 necessary, whilst the counting is going on, to arrest the de- 

 velopment of the yeast cells in the sample. 



Hansen found that dilute sulphuric acid (1 to 10) on the 

 whole answers these requirements ; hydrochloric acid, ammonia 

 and caustic soda may also be used, but they are not so good. 

 If very great dilution is required, distilled water may be added, 

 after the addition of one to two volumes of dilute sulphuric 

 acid. 



When the different volumes of liquid are measured with 

 accuracy, and particular care taken that the cells are thoroughly 

 distributed by vigorous and prolonged shaking, the determin- 

 ation can be made with great accuracy. Two similar dilutions 

 must always be made, and samples taken from each for count- 

 ing. As a matter of course, experiments must also be made to 

 determine the number of the small squares, the cell contents 

 of which must be counted to arrive at a true average. Such 

 counting and determination of the average numbers is con- 

 tinued until the number finally obtained is found to have no 



