452 S. HENRY AYERS AND COTJRTLAND S. MXTDGE 



in number as the growth seemed heavier. It will be observed 

 that the growth was equally good in the medium prepared from 

 the regular yeast and from the yeast extracted with the alcohol, 

 whereas only a slight growth was manifest after forty-eight hours 

 in the medium containing the alcohoUc extract of yeast. The 

 results would seem to indicate that water-soluble B is not the 

 growth-promoting substance of yeast at least for the streptococcus 

 used in this work. 



There is, of course, a possibihty that the water-soluble B was 

 not soluble in the 95 per cent alcohol and was not contained in 

 the extract, and for this reason another experiment was conducted 

 in which a method similar to that of Osborn and Wakeman 

 (1919) was used for the preparation of fractions containing dif- 

 ferent amounts of water-soluble B. Some of the same autolized 

 yeast was used, 10 grams being added to 500 cc. of boiUng water 

 containing 0.01 per cent acetic acid. This was stirred for some 

 time and filtered, the filtrate being concentrated to 300 cc. at a 

 temperature of about 80°C. To this 300 cc. of filtrate sufficient 

 95 per cent alcohol was added to get a concentration of about 

 52 per cent. The solution was allowed to stand over night at 

 about 5°C. and the precipitate which had appeared the next day 

 was filtered off. This precipitate was known as fraction I. 

 The filtrate was then concentrated to 200 cc. and sufficient 95 per 

 cent alcohol added to give a concentration of 79 per cent. This 

 solution also was allowed to stand over night at a temperature 

 of 5°C., and the precipitate filtered off. This was known as 

 fraction II. The filtrate was again concentrated, this time to 

 100 cc, and sufficient 95 per cent alcohol added to give a concen- 

 tration of about 90 per cent. After standing over night at 5°C. 

 the precipitate that formed was filtered off and kept as fraction. 

 III. The filtrate from fraction III, which we shall term filtrate A, 

 was evaporated to dryness, and dissolved in 1000 cc. of distilled 

 water, to which 1 per cent of Difco peptone was added, the reac- 

 tion adjusted to pH 7.2 and the medium placed in flasks and 

 sterilized in an autoclave. Fractions I and II were each dissolved 

 in 1000 cc. of distilled water and 1 per cent Difco peptone was 

 added to each lot. These media were adjusted, placed in flasks 



