THIO AMMONIA OK I'M K SOIL. 



249 



The fact that caustic potasli, a more energetic decom- 

 posing agent than lime, disengaged more ammonia than 

 the latter from the yellow clay, strengthens the view that 

 ammonia is produced and not merely driven oif under the 

 conditions of these experiments, and that accordingly the 

 figures^ are too high. Other chemists employing the same 

 method have obtained similar results. 



Boussingault {A(/ronomie, T. Ill, p. 206) was the first 

 to substitute magnesia for potash and lime in the estima- 

 tion of ammonia, having first demonstrated that this sub- 

 stance, so feebly alkaline, does not perceptibly decompose 

 gelatine, albumin, or asparagine, all of which bodies, espe- 

 cially the latter, give anmionia when boiled with milk of 

 lime or solutions of potash. The results of Boussingault 

 here follow. 



Localities. Quantity of Ammonia per cent. 



Liebfrauenberg, Alsatia 0.0022 



Bischwillor, " 0.0020 



Merckwiller, " 0.0011 



Bechelbronn, " 0.0009 



Mittelhausbei-gen, " 0.0007 



He Napoleon, Mulhouse, 0.0006 



Argentan, Onie, 0.0060 



Quesnoy-snr-Deule, Nord, 0.0012 



Eio Madeira, America, 0.0090 



Eio Trombetto, " 0.0030 



RioNegro, " 0.0038 



Santarem, " 0.0083 



He du Salut, " 0.0080 



Martinique, " 0.0085 



Rio Cupari, (leaf mold,) " 0.0525 



Peat, Paris, 0.0180 



The above results on French soils correspond with those 

 obtained more recently on soils of Saxony by Knop and 

 Wolif, who have devised an ingenious method of estimat- 

 ing ammonia, Avhich is founded on altogether a different 

 l)rinciple. Knop and "Wolff measure the nitrogen gas 

 which is set free by the action of chloride of soda (Ja- 

 velle water*) in a specially constructed apparatus, the 



* More properly hypochlorite of soda, which is used in mixture with broimina 

 Kud caustic soda. 

 11* 



