1895.] The Rubies of Burma and Associated Minerals. 393 



grannlites containing anorthite, and with various pyroxenites and 

 amphibolites. The lime felspars and Kme-soda felspars of these 

 rocks show the greatest tendency to undergo change passing into 

 scapolites by the process known as " werneritisation," and eventually 

 giving rise to the separation of calcium carbonate and hydrated 

 aluminium silicates. That from the last-mentioned salts the 

 hydrated oxides of aluminium (diaspore, gibbsite, bauxite, &c.) may 

 be separated has been shown by the studies of Liebrich and others, 

 while the conversion of these substances into the anhydrous alu- 

 minium oxide has been shown to take place by H. St. Claire Deville, 

 Stanislas Meunier, and others. 



Crystallised aluminium oxide (corundum) has now been formed by 

 chemists by no less than 20 different processes, and in some cases, 

 like those described by Senarmont, Weinschenk, Bruhns, and 

 Friedel, the formation and crystallisation of the substance has been 

 effected at very moderate temperatures under pressure. By one or 

 other of these or similar methods, it is probably that the formation 

 of the Burma corundum and spinel has been effected, the source of 

 the minerals being the decomposition products of basic and easily- 

 altered lime felspars in the pyroxene-gneisses. 



Of still greater interest than the question of the origin of the 

 corundums and spinels are the problems connected with the remark- 

 able changes that these minerals undergo in deep-seated rock masses. 

 The rubies of Burma, when found in situ in the limestones, are 

 usually seen to be enveloped in a mass of materials produced by the 

 alteration of their superficial portions. Nearest to the unaltered gem 

 is a zone of diaspore the hydrated aluminium oxide and this is found 

 to pass insensibly into various hydrous aluminous silicates mar- 

 garites and other clintonites, vermiculites, muscovites, kaolinites, &c. 

 While, in some instances, the corrosion of the rubies appears to have 

 gone on in a seemingly irregular manner, in the majority of cases 

 a very definite mode of metamorphosis may be detected by the study 

 of the various examples. There are evidently certain planes of " chemi- 

 cal weakness " (analogous to the cleavage planes, gliding planes, and 

 other directions of physical weakness) along which decomposition 

 goes on most readily. The principal of these solution planes is the 

 basal plane, and parallel to it we find the gems eaten away in a 

 series of step-like surfaces. Other less pronounced planes of chemical 

 weakness exist parallel to the prism faces. Unaltered corundum is, 

 like quartz, destitute of true cleavage, and breaks with a perfectly 

 conchoidal fracture. If, however, gliding planes and lamellar 

 twinning be developed in corundum (like those so easily produced in 

 the same way in calcite), parallel to the fundamental rhombohedron 

 of the crystals, then these gliding planes become " solution planes," 

 along which chemical action takes place most readily. Along the 



