28 MICRO-ORGANISMS AND FERMENTATION. 



countings are very difficult in the case of the bacteria, and 

 often, indeed, quite impossible. In all cases the accuracy of 

 such calculations is very questionable. Thus, the question 

 remains to be solved : How are we to distinguish the flasks 

 which have only received one cell from those which, in spite 

 of the counting, have been infected with several cells f For 

 the bacteria, no means has as yet been found of solving this 

 difficulty. 



In the case of the yeast, this problem was solved by 

 Hansen, who developed the method to such a degree of 

 perfection as to create, in fact, an exact method (1881). He 

 employed dilution with water, in the following manner : 

 The yeast developed in the flask is diluted with an arbitrary 

 amount of sterilised water, and the number of cells in a 

 small drop of the vigorously-shaken liquid is found. The 

 counting, in this case, is effected in a very simple manner 

 by transferring a drop to a cover-glass, in the centre of 

 which some small squares are engraved, and this is then 

 connected with a moist chamber (Fig. 2) ; the drop must not 

 be allowed to extend beyond the limits of the squares. The 

 cells present in the drop are then counted. Suppose, for 

 instance, that 10 cells are found ; a drop of similar size is 

 transferred from the liquid, which must first be again 

 vigorously shaken, to a flask containing a known volume, 

 e.g., 20 ccm. of sterilised water. This flask, then, will in all 

 likelihood contain about 10 cells. If it is now vigorously 

 shaken for some length of time, and then 1 ccm. of the 

 liquid introduced into each of 20 flasks containing nutritive 

 liquid, it is probable that half of these 20 flasks have received 

 one cell each. But, here again, as in Lister's experiments, 

 it is entirely a calculation of probabilities. If the flasks are 

 left in repose for further development of micro-organisms, 

 there will be a chance of getting a pure culture in some of 

 them. But no certain inferences can be drawn. Hansen 

 succeeded, however, in adding a new link, which first gave 

 certainty to this experiment. If, namely, the freshly 



