8 HARRY WARREN ANDERSON 



medium may prevent solidification. Ten minutes at 5 pounds pressure is usually 

 sufficient. This is better than discontinuous sterilization, since reheating the 

 agar seems to destroy its power of solidifying under these conditions. The high 

 acidity of the medium prevents the bacteria of the feces from developing rapidly. 



Collecting and Plating Samples. Since yeasts are present in fairly large 

 numbers in the air and on unsterilized objects, it was necessary to collect the 

 samples with as little contamination as possible. For this purpose small vials, 

 20 X 70 mm., with wooden spatulas firmly fixed in the cork and extending 

 almost to the bottom of the vial were prepared. These could be sterilized in 

 the hot air sterilizer with the cork and spatula in place. By means of the 

 spatula a small portion of the sample to be studied was introduced into the 

 vial, without contact with external sources of contamination. In this condition 

 the vial could be shipped some distance without danger of contamination and 

 with little change in the water content of the sample. 



The regular dilution and poured plate method was found unsatisfactory 

 for the primary cultivation of the yeasts and their separation from the bacteria. 

 This was probably due to the fact that the yeasts develop rapidly only on the 

 surface of the plate and few are left on the surface under these conditions. 

 Another method was therefore employed in most of the isolations. This is 

 a modification of the method used by Ashford. 



Three plates of the acid glucose agar were poured and allowed to harden. 

 A stiff platinum needle was then sterilized and thrust into the sample. The 

 agar was touched lightly at points about 4 mm. apart in a line across the plate. 

 This process was repeated, each series of contacts forming a row across the 

 plate, 10-20 contacts being the usual number in each row. There were usually 

 about 20 rows on each plate, thus giving 200-400 contacts. This procedure was 

 modified somewhat when a large or very small number of yeasts was expected, 

 or where the consistence of the feces made a change necessary. 



While no exact quantitative results were expected by the use of this method 

 some interesting comparative results were obtained. A sample which contained 

 a relatively large number of yeasts would develop 20-200 yeast colonies on 

 each plate, while in those in which only the normal number was present, only 

 2 or 3 colonies appeared on each plate, or 5 or 6 colonies would appear on 

 1 plate and none on the other two. This striking difference was especially clear 

 in the feeding experiments where the number would jump from less than 

 1% of the contacts before feeding to 80 or 95% immediately after feeding, 

 and then back again to 1% or less a few days later. The dilution method was 

 used in a number of cases as a check on this contact method with the result 

 that the latter was found to be even more accurate than was expected. There 

 is no question but that it gives accurate comparative results when carefully 

 carried out. 



The yeast colonies which were developed on the medium were usually 

 pearl-white or pink in color and could be distinguished from the usually 

 clear bacterial colonies. However, every distinctive colony on a plate was 

 examined in order not to overlook a possible unusual yeast growth. An 

 objection that might be urged against this method is the possibility of yeasts 

 existing in the feces as sensitive to acid conditions as are the bacteria. In 

 order to test this, a large number of yeasts were obtained from various sources 

 and each was plated on agar directly, or mixed with the feces and then plated. 

 In no case did the yeast fail to develop. All the common species of Saccharo- 

 myces, Mycoderma, Torula, Oidium, and Cryptococcus were tried. Also unde- 

 termined yeast-like fungi from fruits, fresh yeast-cake, air, soil, stomach con- 

 tents, etc., were used with like results. 



