362 THE EEGOLITH 



The invocation of atmospheric electricity to account for any 

 part of the nitrates of the soils, they regard as quite unneces- 

 sary, the same being of indirect influence only, furnishing first 

 nitrogen for growing plants which in their turn serve as food 

 for animals. These same authorities give the figures shown in 

 table at bottom of page 361 relative to South American soils. 



(2) The Mineral Composition of Soils. — This is essentially 

 the same as that of the regolith of which the soil forms a part. 

 Fragmental quartzes and feldspars form the larger proportion 

 of most soils. These are intermingled with shreds of mica, 

 amphibole, pyroxene, caleite or aragonite, iron and manganese 

 oxides, and in variable proportions, kaolin and other silicates, 

 carbonates and oxides. The presence of these constituents is 

 usually somewhat obscured by iron oxides and carbonaceous 

 matter; but when these are removed by acids or by ignition, 

 and the residue submitted to microscopic analyses, the true 

 mineral nature can be, as a rule, made out with approximate 

 accuracy.^ 



From what has gone before, it must be evident that the con- 

 stituents of any soil are almost universally in a finely fragmen- 

 tal condition, and, with the exception of quartz and the rarer 

 minerals, as tourmaline, garnet, zircon, etc., in varying and 

 often advanced stages of hydration and decay.^ Silica in the 

 form of free quartz and various silicates, alumina as hydrous 

 silicates, and iron as hydrated oxides, make up from 80% to 

 90% of the superficial portions of most deposits of this nature. 

 It is possible that under favorable conditions new minerals 

 may be temporarily formed. Alumina in the form of hydrated 

 oxides — diaspore, beauxite, gibbsite, hydrargillite, etc. — un- 

 doubtedly exists under certain circumstances. Max Bauer^ has 

 apparently shown the presence of hydrargillite in the laterite 

 of the Seychellian Islands, and van Bemmelen* evidently regards 

 the mineral as a normal final product of the weathering of alumi- 

 nous rocks. Liebich^ on the other hand, states, as a result of his 

 studies on beauxite, that alumina is not liberated from silicates 



* See Anleitung znr Mineralogischen Bodenanalyse, etc., by Pranz Stein- 

 riecle, Inaug. Bis. Friedrichs-UniYersitat Halle-Wittenberg. Plalle, 1889. 



^ See papers by MM. Delage and Legatn, in Comptes Rendu, "Vol. 139, 

 1904, p. 1043, and Vol. 140, 1905, p. 1555. 



^ ISTeues Jahrb. f iir Min. u. Petrog., 1898, Yol. II, p. 163. 



*Zeit. Anorgan. Chemie, YoL 42, 1904, p. 265. 



"Zeit. Prakt. Geol., 1897 (as quoted by Cameron and Bell). 



