ON THE FORMATION OF CRYSTALLINE ROCKS. 257 



By melting silica with difterent bases in definite proportions, he 

 obtained in 1823 crystalline combinations identical with those of na- 

 ture; especially pyroxine.* Afterwards, Ebelmen succeeded, by a 

 very ingenious process due to him, in obtaining infusible combinations. 

 This process consists in employing solvents in a state of fusion, and 

 which can be vaporized slowly at very high temperatures, such as 

 boracic acid, and the alkaline phosphates, or carbonates. It is thus 

 that he produced corundum, the different kinds of spinel, cymophane, 

 peridot, perowskite, and other species. t 



The mutual reaction at high temperatures of the volatile metallic 

 fluorides and oxygen compounds constitutes a process which has 

 lately furnished to its authors, Henri Deville and Caron, very beau- 

 tiful reproductions of infusible minerals, such as corundum, colored 

 in different ways, and staurotide.:!: The same chemists have invented 

 a different process for reproducing apatite. § It is likewise by a par- 

 tial volatilization that M. Gaudin has obtained artificial ruby by melt- 

 ing at a very high temperature a mixture of alum and sulphate of 

 potash. II Despretz has announced that he has obtained the dia- 

 mond by different processes based upon the transposition and slow 

 deposition of carbon by the electric current. IT By melting certain 

 mixtures of salts and treating the residue with water, M. Manross 

 has imitated sulphate of baryta, apatite, wolfram, and other minerals.''^* 

 Rock salt alone employed as a flux has sufficed Forchhammer for 

 producing crystallized apatite, operating, as he did, even on rocks 

 which contained only traces of phosphates. ft If we reflect on the 

 enormous amount of chloride of sodium contained in the liquid en- 

 velope of the globe, we can hardly doubt that this salt has co-operated 

 in the crystallization of certain species, especially at the time when 

 the water was not yet entirely condensed. Charles Deville has 

 recently made experiments in this direction, by heating clay or a 

 quartz sandstone previously moistened with chloride of sodium. :|:| 



Sec. 4. Synthetical experiments by the aid of vapors reacting mutually 



or on fixed bodies. 



Some minerals may be imitated by simple sublimation; such are 

 arsenic, galena, and senarmontite.§§ But it is especially by causing 

 certain vapors to react among themselves, as in metallurgical manu- 



* Annates de Chimie et de Physique, vol. xxiv, p. 365, 1823. 



^ Annates de Chimie et de Physique, vol. xxil, p. 221, and vol. xxv, p. 279. — Annates des 

 Mines, 5th geiie.s, vol. ii, p. 349. 



J (Xmptes Rendus de V Academie des Sciences, vol. xlvi, 1858, p. 765. 



§ Apatite and wagnerite have been obtained by a kind of distillation of the phosphates 

 in the chlorides of the same metals. — {Comptes Rendus, vol. Ixvii, p. 985, 1858 ) 



II Comptes Rendus, vol. xlvi, p. 765, 1857. — The melted alumina in the form of ruby, pre- 

 viously obtained by this author, was amorphous, vol v, p. 803, 1837. 



^ Comptes Rendus, vol. xxxvii, p. 369, 1853. 



••'■•* Annaten der Chemie und Phdrmicie, vol. Ixxxii, p. 348, 1852. 



j\ Poggendorff' s Annalen, vol. Ixxxi, p. 568, 1854. Forchhammer has even proposed 

 this method for detecting phosphates and certain metals in rocks, when by the ordinary 

 processes only slight traces are found. 



XX Comptes Rendus de V Academie, vol. xlvii, p. 80, 1838. 



§§ Annates des Mines, 5th series, vol. i. 



17 



