258 LECTURE XV. 



evident not merely in the immediate neighbourhood of the root-hair, but at the same 

 time affect the more distant parts, since by the absorption through the root-hair at 

 n or r, a continuous flow of the adherent water of S takes place towards y, ß, and a. 

 This assumption is supported by the observation that the earth of large glass vessels 

 in which plants are grown, dries up not merely in the immediate neighbourhood of the 

 absorbing roots, but, so far as the colour of the earth allows us to recognise it, 

 the desiccation increases equally in all parts even far from the roots. This move- 

 ment of water on the surfaces of the particles of soil, easily deduced from molecular 

 forces, is confined therefore not simply to microscopic distances. Every root-hair is 

 itself the centre of a current directed from all sides towards it ; and at the surface 

 of a small piece of root covered with thousands of root-hairs there results a similar 

 movement, which carries the aqueous particles in the soil from all sides, but especially 

 radially, towards the axis of the root. 



If we suppose the aqueous envelope of a particle of soil to consist of several 

 very thin layers, a, d, c . . . . n, according to its thickness, so that n is the outermost, 

 and a the one immediately in contact with the particle of soil ; then the water 

 molecules of the elementary layer a will be attracted with a maximum force, and 

 in the layers 5 c, &c., further removed, this attraction diminishes progressively ; 

 and if n represents the outermost layer, when the soil is just saturated with water, 

 the molecular attraction in it is only just great enough to prevent the water 

 from dropping off. In the case last assumed, when an absorption of water takes 

 place at a or r, the outermost layer of the aqueous spheres moves first, in order to 

 restore the disturbed equilibrium of the whole system, and flows towards a and t; 

 because this outermost elementary layer is most loosely held, and is therefore most 

 easily put in motion. The more water the root-hair has already taken up, the thinner 

 are the aqueous spheres of the entire system, and the greater the force with which the 

 elementary layer now outermost {c, for instance) is held fast ; and the greater must the 

 forces be which draw the water into the wall of the root-hair, and the more difficult 

 and slow the transmission of a disturbance from a to ß, y, and 8. A condition of 

 the aqueous envelopes may finally ensue, where the elementary layers still remaining 

 are held so fast by the particles of soil that no more water enters the wall of the 

 root-hair. In this case, their surfaces may possibly be still clothed with a very thin 

 layer of fluid, which indeed is wanting to no saturated body. If now the root is 

 in connection with a leafy stem above ground, transpiration from this organ will go 

 on removing water from the plant; this loss, however, under the circumstances 

 given, can no longer be compensated by absorption on the part of the root, and the 

 interior of the plant becomes deficient in water, the cells, no longer sufficiently turgid, 

 become flaccid, and the leaves droop. Conversely, it is also possible under 

 certain circumstances to conclude from the drooping of the leaves, and from 

 the quantity of water known to be contained in the soil, as to the condition which 

 denotes equilibrium between the suction of the root and the forces of adhesion 

 in the soil. 



I have in a few cases sought to determine the percentage of water contained in the 

 soil when Tobacco plants rooted in it were no longer in a condition to withdraw 

 the minimum of water from it. This takes place when the leaves droop in a moist 

 atmosphere, even at night, and therefore when the loss by transpiration is very 



