10 THEORY OF IGNEOUS ROCKS AND VOLCANOES. I.] 



of eozoic ages which in many regions are decomposed to great depths, 

 the feldspar being converted into kaolin, while the hornblende has 

 lost its protoxide bases, the peroxidized iron and the silica remaining 

 behind. This change has affected the crystalline rocks of the southern 

 United States and of Brazil to depths of a hundred feet or more, and 

 doubtless at one time extended to all such rocks as were above the 

 surface of the ocean. The absence of this decayed material from certain 

 regions of crystalline rocks is to be attributed to its subsequent removal 

 by denudation, a process which in the northern parts of Europe and 

 America terminated at the close of the pliocene period, when the remain- 

 ing softened material was swept away by the action of water and ice, 

 and the hard and unchanged rocks beneath were exposed and glaciated, 

 since which time the chemical decomposition of the surface has been 

 insignificant. It is this process which was called by Dolomieu the 

 maladie du granit, and ascribed by him to the influence of carbonic-acid 

 gas from subterranean sources. It was, however, in my opinion a uni- 

 versal phenomenon, and dependent upon the peculiar composition of the 

 atmosphere in early times. These decomposed strata furnished the great 

 deposits of clay and sand of the paleozoic and later periods ; and from 

 them was dissolved the iron which in various forms is found at different 

 horizons in the uncrystalline rocks ; while the silica and the alkaline and 

 earthy carbonates, removed in a soluble form from these decaying eozoic 

 rocks, have generated the limestones, dolomites, and various silicious de- 

 posits. (See Proceedings Boston Society of Natural History for October 

 15, 1873.) 



In the Proceedings of the same Society for February 18, 1874, I have 

 called attention to the fact that the clay resulting from this decay of 

 rocks remains for many days suspended in pure water, though not in 

 waters even slightly saline, and is therefore readily precipitated in a few 

 hours when the turbid fresh waters mingle with those of the sea, thus 

 forming fine argillaceous sediments. The geological significance of this 

 fact was, it is believed, first pointed out in 1861 by Mr. Sidell in Hum- 

 phreys and Abbot's Report on the Physics and Hydraulics of the ft 

 sippi River (Appendix A, page xi.), where he applied it to explain 

 the accumulations of -mud at this river's mouth. Many chemical pre- 

 cipitates, in like manner, which may be washed on a filter with arid 

 or saline solutions, readily pass through its pores if suspended in pun- 

 water. I have sought to explain these phenomena by the principle that 

 saline matters reduce the cohesion between water and the suspended 

 particles, thus allowing gravity and their own cohesion to come into 

 play. Guthrie (Proceedings Royal Society, XIV.) has shown that tin- 

 addition of small quantities of saline matters to water diminishes the size 

 of its drops, evidently for the same reason.] 



