SYSTEMATICALLY SELECTED YEASTS. 7 



centimeters of the mixture ; but this result is not always 

 attained, even with the accurate method of counting which 

 I adopted. The dilute yeast was introduced into flasks 

 containing sterilised wort, the degree of dilution beir^ such 

 that only a small proportion of the flasks became infected. 

 The new point about my method was, in the main, that I 

 discovered a means by which I was enabled to distinguish 

 between the flasks which received only one cell and those 

 which received more than one, this result being dependent 

 upon the observation that after the cells had been well 

 distributed by agitation in the nutrient liquid, each settled 

 to the bottom and gave rise to a separate yeast speck. Only 

 those cultures which contained a single yeast speck were 

 assumed to be pure cultivations. By means of the divided 

 cover-glass it is evidently also possible to inoculate directly 

 from a single cell. An exact dilution method was thus for 

 the first time achieved. By its aid I prepared pure cultures 

 of the six Saccharomycetes, the description of which I 

 published in 1883, and also of several brewery yeasts. I 

 also made use of this method in my investigations on the 

 conditions affecting spore formation, on the temperature 

 curves for the development of the spores of the six species 

 alluded to, and on the diseases in beer caused by Saccharo- 

 mycetes. A brief account of these researches was published in 

 1882 (I.e., pp. 206 and 216), and formed the foundation of my 

 yeast studies. All my experiments at that time were carried 

 out with liquids, for the important advance plate culture by 

 means of nutrient gelatine which we owe to Koch, had not 

 then been made. 



This historical account is given mainly as a reply to 

 attempts which were made to show that my studies had their 

 origin in Koch's method of plate cultivation ; we have seen, 

 however, that this was impossible. 



The basis for pure cultivation on a solid substratum was 

 given by Schroter in 1872, and, as was pointed out above, the 



