PLANT FOOD IN SOILS. 207 * 



acid, he concludes that in the case of sandy loam and 

 peaty soils, the method affords a much better criterion of 

 their fertility than is afforded by extracting the soil with 

 strong acids ; but that in the case of clay soils, the ammonia- 

 soluble phosphoric acid is by no means a measure of the 

 soil's fertility. He gives a few complete analyses of the 

 mineral matter extracted by ammonia. All the essential 

 constituents of plant food are present. Ferric and aluminic 

 phosphate form about half of the ash, and silica is a large 

 constituent ; on the other hand, the bases — potash, lime and 

 magnesia — are present only in small quantity. 



Pitsch entirely disagrees with Grandeau's fundamental 

 proposition that the substances extracted by ammonia are 

 all combined with humus in the soil, and are taken up by 

 plants solely from such combinations. He believes that the 

 ferric and aluminic phosphate found in the ammonia solution 

 have been extracted from the soil in the course of the 

 analysis owing to their solubility in ammonium humate. 

 He prepared ammonium humate, and compared its solvent 

 action on various phosphates with that of alkaline ammonium 

 citrate, and found that the solvent power of the humate ex- 

 ceeded that of the citrate both in the case of tricalcic phos- 

 phate and ferric phosphate. The rationale of the method 

 is thus explained : if the soil contains a sufficient amount 

 of humic matter, the extraction with ammonia becomes a 

 similar proceeding to the extraction of a soil with ammonium 

 citrate. The facts brought to light by these investigations 

 appear to indicate that soluble humates have a special 

 solvent power for phosphates ; and, if this be the case, 

 soils w r ell supplied with humus, or manured with farm- 

 yard manure, will generally also be well supplied with 

 available phosphoric acid, and a part of Grandeau's con- 

 tention will be substantiated. It would be interesting to 

 know how a few typical Rothamsted soils behave when 

 treated by Grandeau's method. 



R. Warington. ^z> 



