530 Scientific Proceeding fi. Royal. Dublin Society. 



of the growing season. In such districts the soil containing tlie greater 

 amount of moisture requires a much longer time to be dried and heated up 

 to the point of injury to the crops than in the case of soils of low liygro- 

 scopicity, whose moisture is exhausted by such winds in a few hours, and the 

 surface heats up rapidly to the point of injury. That such occurs in sandy 

 soils much sooner than in clay soils is a matter of common knowledge. 



The accurate estimation of the hygroscopic moisture in soils thus becomes 

 important, since it is the amount in excess of this that is directly available 

 for the plant to draw on for its nourishment. Tiie ordinary method of 

 determining the liygroscopic moisture, by heating in the water-oven — that is, 

 at a temperature approximately equal to 100° C, for an arbitrary period of 

 from twelve to twenty-four hours, apart from the lengtli of time the 

 operation takes — does not give very satisfactory results. A soil which 

 contains much organic matter will slowly lose weight for weeks, and it is 

 questionable how much of this loss is due to water and to volatile matter 

 other than water. Even at as low a temperature as 100° C, under the 

 prolonged heating, some of the combined water of the hydrated silicates in 

 the soil may be lost. 



The power of calcium carbide to act as a desiccating agent has of recent 

 years been put to practical use. It was first employed by H. A. Danne, who, 

 in a paper read before the Society of Chemical Industry of Victoria 

 (Australia) in 1900, described its use in determiuiug moisture in organic 

 substances. In 1906, P. V. Dupres (Analyst, 31, 213) described a method 

 for estimating by means of carbide the moisture in cordite and other 

 substances which give off volatile matter other than water-vapour when 

 heated. A modification of Danne's method is now used in Australia to 

 determine the moisture in wool. J. Masson has used it to determine the 

 water of crystallization in salts (Jour. Chem. Soc. Trans. 97, 851). It has also 

 been found of value in determining the water in petroleum. 



Consequently it was thought that calciilm carbide might be successfully 

 employed to determine the hygroscopic moisture in soils, and a series of 

 experiments were carried out, the results of which are given below. 



The apparatus set up was similar, with slight modifications, to that used 

 by Masson. 



It consisted of a thick-walled glass tube about one and a half centimetres 

 in diameter, with a bulb on one end, the tube being bent to an obtuse angle 

 just above the bulb. Tliis tube was connected by a short flexible joint to a 

 nitrometer containing mercury ; a small sample tube or test-tube which will 

 justfl.t inside the larger tube, but which will not pass the bend in the latter 

 when the tube is tilted, is slipped inside the larger tube. 



