INTRODUCTION. 



Definition of Soils. In the most general meaning- of the 

 term, a soil is the more or less loose and friable material in 

 which, by means of their roots, plants may or do find a foot- 

 hold and nourishment, as well as other conditions of growth. 

 Soils form the uppermost layer of the earth's crust ; but the 

 term does not indicate any such definite average texture as is 

 sometimes implied by its popular use to designate certain loose, 

 loamy materials found in older geological formations. We do 

 find in these, not unfrequently, layers that in the past have 

 served to support vegetation, as evidenced by remains of plants 

 found therein. But as a rule, such ancient soils are much 

 compacted and otherwise changed, and would not now be 

 capable of performing the office of plant nutrition without 

 previous, long-continued exposure to the same agencies by 

 which all soils were originally formed from pre-existing rocks. 

 Within the latter category must be included, in scientific par- 

 lance, not only the hard rocks known as such in daily life, but 

 also such soft materials as clay, sand, marls, etc., which often 

 compose, partially or wholly, the bodies of wide-spread geo- 

 logical formations. 



Elements Constituting the Earth's Crust. More than 

 seventy elementary substances have been found within the 

 portion of the earth accessible to man ; most of these are present 

 only in very minute proportions ; of those occurring in relatively 

 considerable quantities, a list showing their approximate pro- 

 portions is given below. 



Average quantitative composition of the Earth's Crust. 

 The total thickness of the outer shell of the earth, thus far 

 known to us, does not exceed about 95,000 feet, as observed 

 in the accessible rock deposits. Estimates of the proportions 

 in which the more abundant elements contribute to the com- 

 position of these constituent rocks, have repeatedly been made. 

 The latest and most widely accepted of these, by F. W. Clarke, 

 of the U. S. Geological Survey, is given herewith. It in- 



xxiii 



