1846.] Analysis of Soils. 233 



by evaporating the salts after acidulation with chlorohydric acid, 

 to entire dryness, and redissolving in acidulated water, which 

 will take up every thing but the silicic acid of the insoluble sili- 

 cates. The proportion in which these salts exist in soils is so 

 small that we are often obliged to make a solution of several 

 pounds of the soil before the saline matters can all be quantative- 

 ly determined. 



To discover the gaseous contents of a soil we have only to 

 place a quantity of the soil in a flask and connect the flask by 

 means of a syphon tube with a flask containing a solution of ba- 

 ryta water, and from that flask a tube connects with the bell glass 

 of a pneumatic cistern. On boiling the soil in water, carbonic 

 acid is copiously extricated, and precipitates the baryta from its 

 solution in the state of insoluble carbonate of baryta, which being 

 collected, dried and weighed, will indicate the proportion of car- 

 bonic acid gas in the soil. The gas collected in the bell glass is 

 probably atmospheric air, and may be analyzed to determine how 

 much oxygen and nitrogen it contains. It will be found to con- 

 tain less oxygen than ordinary atmospheric air. I have never 

 discovered any free hydrogen or sulphydric acid gas in cultivated 

 soils, but the latter gas is often abundantly present in swamp 

 muck and peat soils, where per-sulphate of iron has been decom- 

 posed by the organic acids. The proportions of oxygen may be 

 learned by abstracting it from a given measure of air, either by 

 protQsulphate of iron, or by explosion with hydrogen gas. Sulphy- 

 dric acid gas may be determined by passing the gas through a 

 solution of acetate of lead the latter being placed beyond the ba- 

 ryta solution. From the proportion of sulphate of lead precipita- 

 ted, we can calculate the sulphydric acid gas. 



4. Analysis of the Organic matters in Soils. 



This is the most difficult department of agricultural chemistry, 

 and it has been unfortunately neglected by most writers on the sub- 

 ject. The processes are more easily described than executed, and 

 much time is required for their performance. I refer not to those 

 imperfect organic analyses which consist in buining the organic 



