1076 THE MIO - ?E IN GEOLOGICAL INVESTIGATION 



salt slowly evapoi-ating. Restoration of the broken angles first takes 

 place : then deposition goes on over the whole exposed sin face, in 

 per:- . crystalline continuity, so as To change a broken 



fragment into a definite crystal. A similar process frequently takes 

 place in limestones which are not absolutely }>nre. x times this 

 adary deposit is carried so far on the gi-ains of a clean sandstone 

 that the interstici - ' :y filled up and the rock is converted 



into a quartzite. 



By the mi. - mination of volcanic- dust o 1 . - - .t is 



possible to determine the constitution ol the igneous mass whose 

 eruption gave rise to such material. Thus the a.shes and dust which 

 fell at various places after th- _ I Krakatoa eruption in ISS.'iwere 

 found to belong to an acid lava, a pyroxene andesite.'-' 



Further, glacial boulders can be satisfactorily identified with rocks 



- : '.y a microscopical examination of their thin s g Thus 



X - _ian rocks have been shown to occur as boulders in the 



Eastern (Amnties. while Swedish and Finnish 

 rocks are common in the drift of Xorth 

 (.Termany and Saxony. 



"We now come to the discussion of the 

 metamorphism to which all rock--. - ..re 

 liable. The metamorphism caused by atmo- 

 spheric agencies results in decomposition and 

 disintegration. The constituents are. of 

 course, very differently affected, but rapidity 

 of disintegration demands the decomposition 

 of one of the principal constituents. Such a 



with much quan nstituent is felspar, which decomposes under 



posited on the surface the influence of water charged with carbonic 

 (After Dr. Sorby.) :lc id i llto kaolin ; while the products of the 



decomposition of non-aluminous minerals are 



carbonates, ferric oxide, and quartz. The minute accessory con- 

 stituents, such as the titanium oxides, are not affected by these 

 agencies, and hence are to be found in all clays and sands. 3 At 

 greater depths from the surface disintegration is replaced by the 

 formation of new, especially hydrous, nrinerals. Thus serpentine 

 is formed from olivine, and sometimes from suitable varieties of 

 augite or hornblende ; chlorite from biotite : epidote from suitable 

 ruinenils. and so on. 



Thermal waters charged with various substances are common in 

 all volcanic districts and play their part in the metamorphosis of 

 rocks. In this way a volcanic rock may become silicified through 

 the percolation of such solutions ; and microscopical examination ha.s 



J E. Wethered, Qitart. Joitrn. Geol. Soc. xlviii. (1892), p. 877. 



1 See J. Murray and A. Renard on 'Volcanic Ashes and Cosmic Dust' in Xature, 

 vol. xxix. p. 585 : also J. W. Judd, Krakatoa Report, published bv the Roval 



M. Hutching, however, is of opinion that rutile is produced as a 



condary mineral in certain slates, though he would not dispute its occurrence as 



ated <-}e 1. Mag. 1890, p. 264). A series of papers bearing on the subject 



which - published since that date in the same periodical are all worthv of 



iidv. 



SOS. Sand-grain (A) 





