EXTERNAL CONDITIONS OF ABSORPTION, ETC. 509 



food. For agriculture these circumstances may be divided into those 

 over which man exercises entire or great control, and those over which 

 he has no influence, and which he must take as they come, or which at 

 most he may press into his service by a knowledge of their perfect regu- 

 larity. These last the child-like man beautifully, and in a certain sense 

 truly, calls "the blessing of Heaven." But for the scientific consi- 

 deration of this subject we need another division, in order to arrange the 

 few facts which are at present known to us. I shall consider, first, the 

 soil; and, secondly, the imponderables, in their relation to the nutri- 

 tionary process in plants. 



1. The Soil. By soil I understand here the earth in the narrowest 

 sense ; all that is necessary has been said above about water and air. We 

 must regard the soil, in relation to the plants growing in it, in a three- 

 fold sense : 



a. According to its chemical constitution, as it contains the inorganic 

 food of plants. 



b. According to its mechanical properties, through which it is fitted for 

 the penetration of the roots, and for holding them firmly. 



c. According to its physical peculiarities. 



The first point has been already dwelt upon ; for the second we have 

 neither facts nor laws. Climate serves the wild vegetation ; this deter- 

 mines the distribution of plants. In agriculture we change the me- 

 chanical constitution of the soil through the plough, the harrow, and 

 manure. It is to the last point, then, we shall address ourselves here. 



Water, as the universal solvent of the nutritive matters, is indis- 

 pensable, and much unnecessary trouble has been taken to calculate the 

 quantity of water, as rain or snow, that falls upon the surface of the 

 earth. The free water in the soil is seldom beneficial to plants, and it is 

 a well-known fact, that when a soil is saturated with water it becomes an 

 injurious locality for the great majority of plants, and that only bog 

 plants, or those which grow in water, will exist in it. In those portions 

 of the earth's surface which produce the most plants, water is only occa- 

 sionally present (as after rain, &c.) as a coherent fluid. The normal 

 condition of water in the soil is as hygroscopic water or absorbed vapour.* 

 The complete independence of vegetation of the atmospheric precipitation 

 of rain in a liquid form, is seen in the vegetation of the Oasis, and of the 

 cloudless coasts of Chili and Peru (see Darwin and Loudon), and in 

 a small way in the experiments of Ward.f The sand of the Sahara pro- 

 duces no vegetation, not because no rain falls upon it, but because it has 

 not the power of condensing aqueous vapour.^ Of the water which falls 

 as rain, very little is used directly by the plant : the greatest part runs 



* How essentially this condition of the water in the soil is connected with chemical 

 processes, and thence with the preparation of the food for plants, is shown by Boussin- 

 gault (Econ. Rur. vol. ii. p. 199.) in a striking manner, in an explanation of the value 

 of gypsum (sulphate of lime) as a manure. Whilst in the presence of fluid water 

 gypsum and carbonate of ammonia are mutually decomposed, in ordinary soil exactly 

 the contrary takes place, and carbonate of lime is decomposed by sulphate of ammonia. 



f On the Growth of Plants in closely glazed Cases: London, 1842. Ward's plan of 

 growing plants in closed cases, where the moisture exhaled by the plant is constantly 

 again absorbed from the soil, is now very generally come into use in Europe, for the 

 purpose of cultivating tropical plants, and with the best possible results. Ward relates 

 cases in which he has kept plants, especially Ferns, in a state of luxurious vegetation for 

 upwards of nine years in a sealed flask. 



\ Perhaps also on account of the absence of aqueous vapours in the air. I know of 

 no hygrometric experiments on the Sahara and other deserts. 



