58 GEOLOGY OF AGRICULTURE. 



qnence, of all soils, inasmuch as soils are formed from 

 rocks. If we consider that quartz, bj far the most 

 abundant mineral in nature, is nearly pure silica, and 

 that the other leading minerals are more than half 

 silica, we need not be surprised to learn that soils con- 

 tain all the way from 60 to 90 per cent, of this ingre- 

 dient. Sandy soils contain a higher per cent, still. 

 Peats and bogs may be excepted, as not being strictly 

 soils, but rather collections of organic matter — par- 

 tially decayed vegetables. The average of silica in soils 

 cannot be less than from 75 to 80 per cent. 



92. From an inspection of analyses of rocks by dis- 

 tinguished chemists, it appears that the older rocks 

 contain rather more silica, and a little less magnesia, 

 alumina and lime, than the newer. If this is really 

 so, then we might infer that there would be found a 

 characteristic difference of soils in the neighborhoods 

 of different rocks; were it not for the f\xct, before 

 stated, that all soils have been so transported and mix- 

 ed, as to preclude the expectation of finding any now 

 remaining unmixed in the region of their formation. 

 When we take this fact into view, I think we may 

 safely conclude that rocks afford but a poor criterion 

 forjudging of the character of a soil, and poorer still 

 for deciding upon the treatment best suited to it. 



AMENDING SOILS. 



93. Most, if not all soils, produce well, when first 

 brought under cultivation. Few, if any, continue to 

 pi'oduce -vyell long, unless well managed. These facts 



