RESEARCH METHODS IN STUDY OF FOREST EtfVIKONMENT. 75 



for the well soil, can never be considered as an exacl measure of the 

 moisture outside. The well samples will be principally useful to 

 showing changes, and without doubt should occasionally be compared 

 with native samples taken near by. It is believed, however tin, for 

 practical purposes a certain constant relation between ,1., two soils 

 may be assumed. So far, because of the great difficulty of actual 

 contact tests between two soils, the moisture ratio al equilibrium 

 must be established on theoretical consideration^ 



From what is known of capillary movement in soil. (116) ,, w, lU |,l 

 seem that, when the moisture content of two soils is near the satura- 

 tion point, they will be in equilibrium at moisture values measurable 

 by the amount which either soil can hold against the force of gravity. 



Similarly, at much lower moisture contents, the amounts which 

 the two soils hold against a force one hundred or one thousand tin 

 as great as gravity, would appear to establish abasis for equilibrium. 

 But, in view of the fact that at a low-moisture content actual capil- 

 lary movement becomes negligible while transfer from one to t 

 other by the vapor-transfer method can be readily accomplished, it 

 seems more logical that we should consider an equilibrium existing 

 which would mean equal osmotic pressures in the two soils. 

 points can be determined for each soil by freezing-] ><>mt depressioi 

 or by assuming equal osmotic pressures at the wilting coefficients. 



Diagram 3 shows a method for working out a scale of relations fo 

 the soil of any well and soil from three depths, obtained when 

 well was dug. The curve for the well soil is a straight line whos 



