132 ON FALLOW. 



contain the same chemical constituents; thus sheep's dung is 

 close and heavy, while the dung of cows and of horses, especially 

 when mixed with straw, is light and pcrous. 



In hot summers, accompanied by light and partial showers of 

 rain, porous soils of no great fertility yield often better crops 

 than richer stiff soils. The rain falling on the porous soil is im- 

 mediately absorbed and reaches the roots, whilst that falling on 

 the heavy soils is evaporated before it is enabled to penetrate 

 them. 



A soil destitute of cohesion, like quick-sand, is not fitted for 

 the cultivation of plants in general. Finally, there are certain 

 kinds of soils which ought, from their chemical composition, to 

 be very fertile, but which, on the contrary, are sterile for many 

 kinds of plants : such soils are those that consist of clay mixed 

 with a large quantity of very fine sand. Such a soil converts 

 itself into a kind of thick mud after a heavy fall of rain, and thus 

 prevents all access of air, and it dries without much contraction. 



If we were to apply, in all their extent, to porous, sandy, or 

 calcareous soils, or to a soil of the nature mentioned above, the 

 principles upon which depend the improvement of land by fallow, 

 we could not hope to obtain favorable results. A soil of great 

 porosity, through which water penetrates with great ease, and 

 which does not yield sufficient hold to the roots of plants, and also 

 a stiff soil, with its particles too finely divided, and of small fer- 

 tility on account of its physical properties, cannot be benefited by 

 the mechanical operations of the field ; for these are intended to 

 effect a still further reduction of the particles. 



The physical conditions essential to the fertility of a soil are 

 usually neglected in the calculations of the chemist, and thus 

 render a mere chemical analysis of very subordinate value ; for 

 the existence of all the mineral means of nourishment in a soil 

 does not necessarily indicate its value. Hut when the chemical 

 is combined with the mechanical analysis* (for the latter of which 

 Mr. Rham has described an equally simple and convenient instru- 

 ment), then we are furnished with data upon which to form 

 accurate conclusions. 



* The estimation of the unequal quantities of mixed ingredients, such 

 as of the coarse and fine sand, and of the clay and vegetable matter? 



