New York Agricultural Experiment Statio 



:•;. 



207 



Table II. — Tests Comparing Different Batches of Soil-Extract Gelatin. 



Soil type. 



Dunkirk silty clay loam. 

 Dunkirk silty clay loam . 

 Dunkirk silty clay loam . 

 Dunkirk silty clay loam. 

 Dunkirk silty clay loam. 

 Dunkirk silty clay loam. 

 Dunkirk silty clay loam. 

 Dunkirk silty clay loam . 

 Ontario fine sandy loam . 

 Ontario fine sandy loam . 



Bacteria Per Gram Dry Soil, as 

 Determined with — 



Batch I - 



tnontl 



old. 



24,000,000 

 14,000,000 

 12,000,000 

 35,000,000 

 24,000,00!) 

 21,500,000 

 39,000,000 

 20,000,000 



Batch II — 



fresh, but 



decomposed. 



10,000,000 



9,000,000 



8,500,000 



23,000,000 



13,000,000 



13,500,000 



25,000,000 



14,500,000 



9,500,000 



9,500,000 



Batch III 

 fresh ; 

 good. 



19,000,000 

 *20,000,000 

 32,000,000 

 16,500,000 

 *9, 500, 000 

 11,500,000 



* These counts are inexact because of rapid liquefaction. 



SIMPLIFICATION OF THE FORMULA OF THE SOIL-EXTRACT 



GELATIN. 



The opportunity for such variations in composition seems, a priori, 

 to be greater in the case of this gelatin than with any of the agar 

 media discussed in this paper. Soil-extract is unquestionably of 

 variable composition. Gelatin itself also may be the cause of con- 

 siderable irregularity. It is more complex in chemical composition 

 than agar and presumably more variable. It may perhaps contain 

 fewer impurities; but it is used in ten times as large quantities as 

 agar, which must result in the introduction of large amounts of 

 whatever impurities it does contain. Lastly, gelatin is a food for 

 many bacteria, and for that reason variations in its composition 

 must have more influence upon bacterial growth than those in 

 agar, which is not ordinarily of nutrient value for bacteria. 



In the hope of eliminating some of these causes of variation, an 

 attempt was made, toward the close of the present investigation, to 

 simplify the formula of the gelatin. If any way of purifying the 

 gelatin itself had been known, that would have been undertaken. 

 In the lack of such knowledge, attention was turned to the soil- 

 extract. Eliminating the soil-extract could not prevent the sort of 

 variations shown in Table II, but it might prevent others equally 

 great. 



The soil-extract was first replaced by tap-water. The results 

 were so surprisingly successful that both the tap-water and the 

 dextrose were finally eliminated, leaving only a solution of gelatin 

 in distilled water, clarified with white of egg. The results are given 

 in Table III. It will be seen, first, that the tap-water gelatin with 



