ARTICLES 575 



available water in different types of soil, of the manner of its 

 distribution over the minute soil grains, and so on. Early in 

 the nineteenth century, Davy, in the course of a series of 

 lectures under the auspices of the newly formed Board of Agri- 

 culture, laid stress on the importance of the physical properties 

 of soil, while Schiibler in Germany, and Boussingault in France, 

 laid the foundations of the subject in a series of classical 

 experiments. 



They regarded the soil as a framework of mineral particles 

 of all shapes and sizes, something like a heap of sand and 

 considered that the soil moisture was spread over these particles 

 in the form of a thin film. The movement of the water in 

 the soil was controlled by the force of gravity, capillary attrac- 

 tion in the minute pore spaces between the grains, evaporation 

 from the soil surface, imbibition by plant roots, and so on. 

 The well-established principles of pure physics were directly 

 applicable to a soil of this type, and it was rightly concluded, 

 from the methods and evidence then available, that a fairly 

 complete survey of soil physics had been made. This was the 

 first historical period, which closed about the year 1850, 



At this time the famous researches of Lawes and Gilbert, 

 and Warington at Rothamsted, and Liebig in Germany, directed 

 world-wide attention to agricultural chemistry, and soil physics 

 fell into the background. A small amount of work was still 

 done, however, by agricultural chemists and others, who 

 found — as we have already mentioned — that the solution of 

 a particular problem demanded some knowledge of the physical 

 environment in the soil. 



The opening of the second historical period may be dated 

 from 1900, when Warington published a monograph on the 

 physical properties of soil. He discussed in detail the know- 

 ledge obtained from the old water-film and sand-grain hypo- 

 thesis, and indicated certain soil characteristics whose explana- 

 tion demanded at least an extension of this hypothesis. This 

 extension was first given, apparently, by J. Dumont in 

 France, who assumed the soil particles to be coated, more or 

 less completely, with a film of colloidal materials derived, 

 probably, both from the mineral material of the soil — zeolites, 

 etc. — and from the organic residues of plants. On this view, 

 soil should show more or less marked colloidal properties such 

 as those demonstrated earlier, in the famous researches of 

 Van Bemmelen at Leiden, and hence, considerable advances 

 should be possible in our knowledge of its physical properties, 

 by applying to it the powerful technique and theories of the 

 colloidal state of matter. We are hardly more than at the 

 beginning of this second historical period, but sufficient results 

 have already been obtained — first of all in Germany, later in 



