SOIL PHYSICS, CHEMISTRY AND MICROBIOLOGY 937 



of energ}' to the denitrifying bacteria of the soil, and for this reason coni- 

 ferous forest soils contain less nitrates than others. 



Infltience of Tannin. For 2 months the nitrification of soil samples 

 to which 5 % and 10 % respectively of tannin + 2 grms of ammonium 

 sulphate had been added was studied. It was found that the soil treated 

 with 10 % of tannin contained less nitrates than that treated with 5 % of 

 tannin, just as in the observ^ations relating to resin. The same experi- 

 ments al.so showed that the organisms which converted the nitrates of 

 the soil utilised tannin as a source of energy. It becomes clearly evident 

 that a soil rich in tannin must be poor in nitrates. The studies undertaken 

 in order to ascertain the soil organisms which utilise both tannin and nitrates 

 disclosed the presence of a fungus, Aspergillus niger, sometimes accom- 

 panied by other fungi. Contrary to the general belief, this Aspergillus abounds 

 in the soil, but only works under given conditions, namely when the soil 

 is rich in tannin. It decomposes the latter and reduces the nitrates to ni- 

 trites. Sugar is the component of tannin which it uses, but there are pro- 

 bably others also. The presence of calcium carbonate increase its decom- 

 posing power. Like bacteria, Aspergillus niger uses nitrates as a source 

 of nitrogen, and also salts of ammonia when the soil contains tannin. The 

 latter therefore is injurious. 



As Aspergillus converts sugar into oxalic acid, the writers investigat- 

 ed the influence of this acid on nitrification. Their conclusion is negative. 



726 - Adsorption of Potassium by the Soil. — mc call a. g. ; hildebrandt f. ji. aud 



Johnston E. S. (J. Phys. Chem., XX, 51-63, 1916). Journal of the Society of Chemical 

 Industry, VoL XXXV, No. 4, p. 267. London, Feb. 29,1916. 



To ascertain the effect of contact for a short time between a soil and 

 a salt solution, and the influence of surface area upon the amount and rate 

 of adsorption, an approximately five hundreth normal solution of potassium 

 chloride was caused to percolate through a sandy loam soil, the potas,sium 

 in the percolate being estimated colorimetrically. This soil was used in two 

 different states : a) dried and passed through a 2 mm. sieve, and b) very 

 finel}^ ground in a porcelain-lined ball mill for 4 da^'s. The apparatus consisted 

 of a Pasteur-Chamberland filter tube surrounded by a brass jacket to 

 hold the soil, the whole being enclosed in a porcelain-lined filter chamber, 

 into which the solution was poured, and which was fitted with an air-tight 

 cap in connection with an automobile tyre pump. The amount of potas- 

 sium leached out of the soil samples by pure water was first ascertained : 

 much more was dissolved from [b) than from [a); 20 gr. of (a) was treated 

 with 250 cc. of the potassium chloride .solution containing 62 parts per. mil- 

 lion cf potassium, the flow was maintained at a uniform rate, and the per- 

 colate was collected in fractions cf 50 cc. The amount of adsorbed potassium 

 was then immediately found by leaching with pure water, at the same rate 

 of flow. The amounts of potassium in the first five percolates were, 40, 36, 

 40, 44, and 59 p. p. m. respectively, and the corresponding amounts retained 

 p. p. m. of dry soil were 58,124,181, 226 and 233. The leaching action of 

 water gave 7, ii, 9, 9, 8 p. p. m. of potassium in the fractions. With soil 

 sample [b), it was found very unexpectedly, that the amount of potassium 



