54 Henry H. Dixon. 



volume, and so eluding the tensile condition by means of its mobilitj^ 

 The tensions developed are limited to the capillary forces of the 

 interstices of the soil which, except in very fine grained soils, will 

 be insignificant. To view the matter in another way, the soil, lacking 

 the compartmented structure of the water ways of plants, cannot put 

 an effective limit on the size of the bubbles, which are consequently 

 free to expand indefinitely and so render high tensions impossible. 



In dry weather, when there is but little water in the soil, it is 

 probable that a large percentage of water enters the root by con- 

 densation in the walls. For the water surface in the walls will 

 possess, owing to the concavity of its menisci, as explained, a lower 

 vapour tension than the water in the soil. 



This condensation ^) on the root surface appears to be a much 

 more satisfactory explanation of the absorption of water from com- 

 paratively dry soils, than the generally accepted one which was first 

 offered by Sachs.-) According to this physiologist the particles of 

 soil may be regarded as surrounded by concentric envelopes of water, 

 the outer envelopes being more mobile than those close to the particles. 

 When a root hair draws off a quantity of water from the mobile 

 envelopes which are adjacent to it, the loss is made good by a flow 

 from the outer envelopes of all the other particles in the neighbourhood 

 of the root hair. 



Sachs' diagram representing the surroundings of a root hair 

 in the soil shows the latter as consisting of grains of sand, water 

 and air bubbles. In such a configuration the most natural course of 

 events would be as follows: The root hair would drink in water 

 from the interstice in which it found itself, and so, hj diminishing 

 the amount of water, would enlarge the bubbles in its neighbourhood. 

 The result of this would be to diminish their surface tension forces 

 owing to the increased radius of curvature of their surfaces, and the 

 further drawing off' of water would lead to a continual thinning out 

 of the films of water covering the particles. 



Even when there are large quantities of water in the soil it 

 ma,j be shown that the aqueous envelope covering the clay particles 

 is very thin. Thus soils containing 30 per cent of their weight of 

 water may be regarded as wet soils, for few soils can retain as much 

 as 50 per cent. If we assume the mean diameter of the particles 

 to be about the same as the diameter of a root hair, which is about 

 the relation in Sachs' diagram referred to above, and if, for the 

 sake of estimating the thickness of the envelopes, we regard the 



^) H. H.Dixon, Phj'sics of the Transpiration Current. Notes from the Botanical 

 School, Trinity College, Dublin, No. 2, 1897, p. 22. 



^) J. von Sachs, Lectures on Plant Physiology. Trans, by H. Marshal 

 Ward. 



