158 NATURE AND PROPERTIES OP SOILS 



nitnde of the hygroscopic coefficient. With the Clyde and 

 Vergennes, howeTer, the organic colloidal matter is dominant, 

 since the Clyde with only 20 per cent, of clay has a higher 

 hygroscopic JSgure than the Vergennes which carries 74.5 per 

 cent, of that separate. The Clyde clay loam and the Dunkirk 

 subsoil have the same amount of clay, yet the former pos- 

 sesses a hygroscopic coeiScient over three times larger. 



Two external conditions seem to be important in determin- 

 ing the amount of hygroscopic water in soils — (1) humidity 

 and (2) temperature. It has been definitely established that 

 the higher the humidity the higher the content of hygro- 

 scopic moisture. An air-dry soil will, therefore, contain less 

 moisture in a dry atmosphere than in one carrying large 

 amounts of water-vapor. When the soil is in contact with a 

 saturated air it will take up hygroscopic water to its full 

 capacity and be at the point spoken of as the hygroscopic 

 coefficient. As the soil air is generally considered to be satu- 

 rated or almost saturated with water-vapor,^ except in the 

 surface layers or during periods of protracted drought, a soil 

 in normal condition may be considered, for all practical pur- 

 poses, to be at its maximum hygroscopicity. An increase of 

 the temperature of the saturated atmosphere seems to increase 

 hygroscopicity. With a partially saturated air the influence 

 seems to be in the opposite direction.^ This, however, is not 

 an important practical point. 



The hygroscopic coefficient, defined as the maximum hygro- 

 scopic water that a soil will hold, is controlled largely by the 

 texture and organic content of the soil. It may vary from a 

 very low figure in asandy soil to as high as 15 per cent, for 

 a clay high in organic matter. With a muck or peat, the per- 



^ Russell, E. J., and Applyard, A., The Atmosphere of the Soil: Its 

 ComposiUon and Causes of Variation; Jour. Agr. Sci., Vol. VII, Part 1, 

 p. 5, 1915. 



*For a fuU discussion of tMs point, see Lipman, C. B., and Sharp, 

 L. T., A Contribution to the Subject of the JEygrosco'piG Moisture of 

 Soils; Jour. Phys. Chem., Vol. 15, No. 8, pp. 709-722, Nov., 1911. 



