stirring of the air inside, and especially by use of a thermostat, the 

 temperature can be kept within one-half degree of that specified. 



The method of digestion in detail is as follows : 



Weigh and place in one of the digestion flasks the quantity which 

 will be specified in a supplementary table, this weight being based on 

 tin- moisture found to be in the sample and equivalent of 186 grains 

 of dry soil. Measure out the number of cubic centimeters of N-10 

 HC1, which will be given in the supplementary table, and dilute to 

 1 liter. This table is based upon the basicity of each soil sample. 

 Half fill the flask containing the soil ; shake with the diluted HC1 

 thoroughly, so as to displace the air held in the soil, and then fill up 

 to the neck with the dilute acid. The contents of the flask will be 

 the equivalent of 200 grams of soil in one litre of N-200 HC1. The 

 ilasks are stopped with rubber stoppers, which must be tied in. 

 Fasten the flasks in the shaking machine and digest for five hours 

 at 40 C., the apparatus making about 40 revolutions per minute. 

 This has the effect of keeping the soil constantly suspended in the 

 liquid, insuring a maximum and constant effect of the solvent. 



At the close of the digestion the flask with its contents is shaken 

 well and emptied into a 32 cm. fluted filter. Unless the sediment is 

 poured on with the liquid the filter will become clogged. The first 

 part of the filtrate is generally cloudy and is poured back as soon as 

 the liquid has begun to run clear. A small filter first placed in the 

 funnel will prevent the breaking of the larger paper when it is 

 1< >aded. These filtrates generally drain through in less than an hour, 

 and the volume is found to be so near 800 c. c. that it is not neces- 

 sary to measure. To the filtrate is added 2 or 3 c. c. of HNO 3 and 

 about 10 c. c. of HC1, and it is evaporated to dryness in a porcelain 

 dish G or 8 inches in diameter on the steam or water bath. The HNO :l 

 roys organic matter in solution and drives off any ammonium 

 chloride which may have resulted from the liberation of ammonia in 

 tin process of digestion, and which, if present, would subsequently 

 interfere with the determination of potash. To the residue in the 

 dish is now added strong HC1, at first slowly, to prevent loss by 

 spurting. It is then evaporated until the nitrates have become com- 

 pletely converted into chlorides, which can be told by the absence of 

 i eenish tinge and fumes upon the addition of HC1 to the dry 

 residue. 



It sometimes requires from 50 to 100 c. c. of HC1 before the con- 

 ion is completed. When complete, the solution is evaporated to 

 dryness and cooled; then a few drops of HC1 are added and about 

 r 30 c. c. of water, and the whole is warmed on the bath until 

 solution is secured. 



Filter to remove silica and wash until clean. To the filtrate, which 

 corresponds to IfiO grams of soil, in a :j or 4-inch porcelain dish, add 



