I40 SOIL CONDITIONS AND PLANT GROWTH 



58*2 per cent, carbon and 5*0 per cent, hydrogen. In 

 view, however, of Hoppe-Seyler's discovery, confirmed by 

 Od6n,^ that soil humus is partly soluble and partly in- 

 soluble in hot alcohol, it is obvious that at least two sub- 

 stances must have been present in the material analysed in the 

 above cases and the validity of any formula becomes doubtful.^ 



The purification of humus for investigation in the labora- 

 tory is rendered extraordinarily difficult by its colloidal nature, 

 which was demonstrated by van Bemmelen in a remarkable 

 paper in 1888 (19). Baumann's researches (10) have carried 

 the subject a good deal farther, and it is now known that 

 "humus," freshly precipitated by acids from an alkaline ex- 

 tract of soil, compost, etc., possesses the following colloidal 

 properties : — 



(i) Very high capacity for retaining water. 



(2) Extraordinary shrinkage on drying. 



(3) Is coagulated by acids and salts, the electric current 

 and frost. 



(4) Reversibility, i.e. the freshly precipitated material re- 

 dissolves when the precipitant is washed away. 



(5) Decomposes salts — calcium carbonate, calcium phos- 

 phate, etc. 



(6) Forms difficultly soluble and easily decomposable col- 

 loidal mixtures with other colloids. 



(7) Masks certain ion reactions {e.g. Fe cannot be detected 

 by potassium ferrocyanide, etc.). 



(8) Forms absorption compounds. 



The colloidal nature of the humus accounts for much of 

 the failure of the earlier chemical work, and it adds to the 

 difficulties of those modern chemists who have had the courage 

 to tackle the problem. In spite of these difficulties several 

 good investigations have been made. 



1 Ber., igi2, 45, 651. 



^ Miklauz (Zeit. f. Moorkultur u. Torfverwertung, 1908, 285) and Mayer 

 both showed that variations in the time or the method of extraction cause differ- 

 ences in the results even from the same soil, Sostegni (Landw. Versuchs-Stat., 

 1886, 32, 9) had shown that humus is readily fractionated. 



