William Weir 247 



more water than the other. Beans were phinted in both : they developed 

 normally on the untreated soil but failed entirely on the extracted soil. 



During the past two seasons this experiment has been repeated on 

 a larger scale, with results sufficiently interesting to justify their being 

 placed on record. The removal of the soluble humus was effected by 

 washing the soil with dilute hydrochloric acid to remove bases, and then 

 repeatedly extracting with dilute soda. Some 40 per cent, of the 

 nitrogen in the soils was thereby removed. 



Vegetation experiments were then conducted with large quantities 

 of mitreated and extracted soils, and approximately equal total yields 

 both of dry matter and of nitrogen were obtained over four successive 

 crops. Thus it appears that the removal of the soluble humus had no 

 effect in diminishing the productiveness of the soil in spite of the fact 

 that the soil used was known to respond to nitrogenous fertihsers. 



Laboratory experiments were then started to ascertain the effect of 

 removing soluble humus on the rates of production of nitrate and 

 ammonia in the soils. Here the indications were that the removal 

 of the soluble humus increased the amount of ammonia but diminished 

 that of nitrates in the soil, and the sum of ammonia and nitrate was 

 usually less than in the untreated soil. The numbers of bacteria 

 however were considerably increased. No marked difference was pro- 

 duced where 0-5 per cent, untreated soil was added to replace the 

 bacterial flora that might have been destroyed by the acid and alkali 

 treatment. These results are not necessarily inconsistent with the 

 vegetation experiments. It has been shown in these laboratories that 

 the accumulation of ammonia and nitrate in an uncropped soil will 

 only proceed to a certain stage when it stops-*-. In vegetation experi- 

 ments on the other hand the nitrate is jjerpetually being removed by 

 the growing plant. There is thus a considerable difference in conditions. 



Method of Extraction. 250 grams of air-dried soil, which has passed 

 through a 3 mm. sieve, are shaken up with 2500 c.c. N/5 HCl. After 

 standing in contact with the acid for one hour, the soil is thrown on to 

 a filtering funnel and washed free of acid ; 500 grams of soil treated in 

 this manner are shaken with 2000 c.c. 2 per cent. NaOH and allowed to 

 settle overnight. The dark-coloured humus extract is siphoned oft'. 

 This process is repeated five times in all, and at the fifth extraction the 

 liquid is of a lightish brown tint. The soil is then shaken with water, 

 allowed to settle, the supernatant liquid is removed, and finally the 



1 "The Effect of Partial Sterilisation of Soil on the Produetion of Plant Food," by 

 E. J. Russell, and H. B. Hutchinson. Journal of Agricultural Science, Y. p. 193. 



Journ. of Agrio. Sci. vii 17 



