156 CONTRIBUTIONS TO OUR KNOWLEDGE OF SOIL FERTILITY, xvi., 



same manner that chalk maintains a faint aciditv in feimenting 

 solutions of sugar. The only substance that promised to achieve 

 this purpose was humic acid, and its use appeared advisable, as 

 it is ever present in soils. 



A quantity of humic acid was accoi-dingly prepared from 

 rotted bamboo-mould by treatment with sodium h\'drate, followed 

 by acidification, washing with water by decantation, filtration, 

 and finally by dialysis. The acids were dissolved in normal soda 

 and sterilised. 



Tests were made with strung hay-infusion with and without 

 dextrose by the cotton-wool method, but, although several 

 organisms were used, no results of any importance were obtained: 

 the solutions were far too nutritive. A converse test with tap- 

 water and no nutrient showed that the humic acid acted as a 

 poison. Although an opalescent suspension of bacteria had been 

 added, no living bacteria were to be found by the third day. In 

 one case, a pink yeast had obtained access to the fiask, and, on 

 the thirteenth day, the extract showed a rise upon boiling; and, 

 upon dilution with dilute hay-infusion, the same sterility, 

 towards the introduced bacteria, was noted after a few days. 

 The humic acid was undoul)tedly the cause of the plienomenon. 

 It had been dissolved in normal soda, and the solution was 

 added to the flask of medium, and was followed by the quantity 

 of normal hydrochloric acid necessary to neutralise the soda. 

 The humic acid remained largely dissolved, and acted as a soluble 

 acid. In one test, that of the pink yeast on the thirteentii day, 

 the acidity of the extract to phenolphthalein was = -t-0"5' by 

 Fuller's scale, that is the extract contained Go c.c. of normal 

 acid per litre. The experiments suggest that, to beefiective, the 

 humic acid should be insoluble in water. 



Experiments with Humus. 



As the toxin-formation in soil must be related to the changes 

 in the organic matter, an attempt was made to see in how far 

 humus would be useful in elucidating the problem. Rotted 

 bamboo-grass was sifted, partly dried, mixed and sifted again to 

 obtain a uniform mould. Portions weighing 10 grams were put 

 into deep Petri-dishes, and sterilised by heating for two hours at 



