NITRIFYING BACTERIA. 393 



remain. These are to be seen as an almost transparent 

 film attached to the clumps and granules of magnesium 

 carbonate on the bottom of the flask. 



For their cultivation upon a solid medium he em- 

 ploys a mineral gelatin, the gelatinizing principle of 

 which is silicic acid. A solution of from 3 to 4 per 

 cent, of silicic acid in distilled water, and having a 

 specific gravity of 1.02, remains fluid and can be pre- 

 served in flasks in this condition (Kuhue). By the 

 addition of certain salts to such a solution gelatin ization 

 occurs and will be more or less complete, according to 

 the proportion of salts added. The salts that have given 

 the best results and the method of mixing them are as 

 follows : 



f Ammonium sulphate . . . .' .0.4 gramme. 



a -j Magnesium sulphate . . ... . 0.05 " 



I Calcium chloride . . . . trace. 



f Potassium phosphate . . . . . .0.1 gramme. 



b< Sodium carbonate 0.6 to 0.9 " 



(. Distilled water 100 c.c. 



The sulphates and chloride (a) are mixed in 50 c.c. of 

 the distilled water, and the phosphate and carbonate (6) 

 in the remaining 50 c.c., in separate flasks. 



Each flask is then sterilized with its contents, which 

 after cooling are mixed together. This represents the 

 solution of mineral salts that is to be added to the silicic 

 acid, little by little, until the proper degree of consist- 

 ency is obtained (that of ordinary nutrient gelatin). This 

 part of the process is best conducted in the culture dish. 

 If it is desired to separate the colonies, as in an ordinary 

 plate, the inoculation and mixing of the material intro- 

 duced must be done before gelatin ization is complete; if 

 the material is to be distributed only over the surface 



