lo July, 1Q12.] Iiifiuciicc of Certain Soil Coiistiliicnts. &c. 



395 



The soil used was that described as "sandy" in the previous article, 

 and was taken from the same bulk. The chemical and mechanical analysis 

 of this soil has already been published. 



The method of experiment was essentially the same as that adopted 

 in the earlier investigations. Air-dry soil, equal to 300 grams of dry 

 soil, includ ng the added substance, if any, was wetted up to 60 per cent, 

 of its water-holding capacity, and 5 c.c. ammonium sulphate solution con- 

 taining 0.1 gram nitrogen was mixed with the wetted soil. This nitrogen 

 equalled SSS-^S P''irts per million of dry soil. After adding the ammonia 

 the soil was transferred to a bottle, and compacted by tapping to a uniform 

 volume in each case. With any substance, there were always two bottles, 

 one getting ammDnia, and one getting no ammonia (control). Where am- 

 monia was given, the water of solution was part of the water required to 



ASPIRATl?vIG THli LOTTLES. 



give the desired moisture. The bottles were corked during the incuba- 

 tion period, and were placed in a dark cupboard where the temperature 

 was read daily at 9 a.m. and 5 p.m. Twice weekly the bottles were 

 uncorked, and aspirated for five seconds with a suction pump in order 

 to maintain fresh air. Number 17, however, was sealed, and kept closed 

 throughout the whole course of the exper'ments. 



The eighteen tests, with their controls, required thirty-six bottles. An- 

 other set was filled similar to these, making seventy-two bottles. In addi- 

 tion, four extra bottles of Nos. i and 2 were placed aside in order to get 

 some idea when the tim-e had arrived to begin the full analyses. 



The methods of analys's adopted have already been described. The 

 work was performed by Mr. W. C. Robertson, chief deputy chemist, who 

 also attended to the filling of bottles and keeping of records. 



o 2 



