ARTICLES 585 



shape of the experimental curves. But ignition of the soil, 

 which destroys the colloidal properties of the clay, does affect 

 the evaporation relationship ; the curve given by remoistened 

 ignited soil is identical with that given by sands, and the in- 

 ference is that the colloidal portion of the clay fraction in the 

 soil is predominantly concerned in the moisture relationships. 

 It will readily be seen that the conception of a gel-like coating 

 over the particle surface would be necessary under these 

 conditions. Space does not permit of further elaboration of 

 these views : those who wish to follow up the matter will find 

 a discussion by the writer in the Journal of Agricultural Science, 

 vol. X (1920), p. 44. 



(3) The Soil Temperature 



Temperature readings taken in the top layers of soil show 

 that, normally, it experiences a daily rise and fall of tempera- 

 ture, due to the heat received from the sun in the day-time, and 

 radiation from the soil surface into space during the night. This 

 daily temperature wave is propagated downwards into the 

 soil ; the temperature fluctuations due to it decrease rapidly 

 with increasing depth, becoming inappreciable at the 3-foot 

 depth. This diminution of temperature amplitude is accom- 

 panied by a " time-lag " : it takes from 4 to 6 hours before a 

 temperature change at the soil surface iDecomes apparent at 

 a depth of 6 inches. 



The general problems of heat conduction when the heat 

 supply varies in a periodic manner have been worked out by 

 Kelvin, Forbes, and Angstrom. The agricultural significance 

 of these fluctuations in the soil has not yet been fully investi- 

 gated. Ignoring their effect on biological conditions, and 

 confining attention to the physical problems, the important 

 factors are : {a) the soil material is not continuous, but consists 

 of irregular particles touching one another at one or two points ; 

 {b) there is a film containing water surrounding these parti- 

 cles. Patten, in an extensive laboratory investigation on heat 

 conduction in soils, has shown that, although soil material in 

 the continuous, or rock condition, has a higher conductivity 

 than water, the conducting power of actual soil is improved 

 and not decreased by the presence of water, up to a certain 

 point. This is ascribed to the improved thermal contact 

 produced between the particles owing to the presence of the 

 water film. An increase in moisture content beyond the 

 critical point results in a decrease of heat conductivity, which 

 slowly falls towards the value for water. 



Temperature changes have another important effect on 

 the soil moisture. Its viscosity and surface tension decrease 



