1897] SOME NEW BOOKS 419 



the original rock was soluble in hydrochloric acid and sodium car- 



1 1< mate, the proportion is now nearly 70 per cent. The instance is 

 that of a gneiss from Virginia. 



100-08 30-17 99-9S 69-18 44-67 



i Ga'n. 



This table makes at once obvious the great loss in silica, lime, and 

 alkalies, and the gain in water ; while it further indicates from the 

 increase in the soluble portion that the state of combination in many 

 of the compounds has changed. Mechanical separation and micro- 

 scopic examination of the decomposed material throws further light 

 on the character of the decomposition and nature of the surviving 

 minerals. 



In basic rocks the loss of silica is rather less, and that of iron and 

 magnesia more conspicuous, while in a French basalt the felspars 

 were the first to yield, and the augite and olivine the last. 



Under the heading " Physical Manifestations " are treated : Dis- 

 integration without decomposition, influences of crystalline and rock- 

 structure, mineral composition (roughening and crusting), indura- 

 tion and colour changes. Original characters are sometimes lost 

 entirely through weathering, basalts pass into apparent argil- 

 laceous deposits, and granites and gneisses with their veins and 

 every structural detail well preserved, may become so soft that a 

 stick can be thrust deeply into them. Chemical compounds become 

 on the whole simplified by weathering, but oxidation and all chemical 

 change, except hydration, seems to cease below the permanent water- 

 level. More refractory and dense residues, like xenotime, monazite, 

 tourmalines, rutiles and precious gems, have tended to accumulate in 

 favoured spots, on account of the weathering and destruction of the 

 rocks in which they were originally contained. 



The rate of weathering is influenced by composition, texture, and 

 position of rocks, and by humidity of climate. Weathering in humid 

 climates differs in kind, as well as degree from that effected in dry 

 climates as shown in California by the author and by Judd in the 

 Xile delta. Prehistoric implements and old surfaces covered uncon- 

 formably by newer deposits may sometimes enable us to guess at the 

 amount of work done in a given time, but the ascertaining of the actual 

 rate of work is difficult, and the results not very satisfactory. Geikie's 

 results, on this branch of the subject, not quoted here by the author, 



