424 H. S. Stimmers : 



a number presented by Mr. Kemp, of Peake Station, Lake Eyre 

 District, to the members of Professor Gregory's party dviring 

 their trip to that area. Besides making accurate determinations 

 of the main constituents of the obsidianites, Mr. Ampt has made 

 an exhaustive determination of the rarer constituents, and 

 must be congratulated on the high standard of his analytical 

 work. 



Several writers have referred to the chemical composition of 

 obsidianites. Mr. Walcott,! while pointing out that the analyses 

 he quotes show that the glasses belong rather to the trachytic 

 than to the rhyolitic series, adds: — "It is to be regretted that 

 so little has been done in their chemical examination, because 

 it is cpiite possible that each occurrence may present features in 

 common, while differing from those of others. We should also 

 be able to ascertain whether any divergence from ordinary 

 obsidian can be established." Mr. Simpson, 2 after quoting an 

 analysis, states: — ^" In chemical composition this specimen 

 agrees with those found in Victoria, New South Wales, and Cen- 

 tral South x\ustralia, as well as from Billiton and other islands 

 of the East Indies. It is identical with that of ordinary obsidian 

 of undoubted volcanic origin." Referring^ to the analyses of 

 obsidianites from the Upper Weld and Pieman, Dr. Hillebrand 

 says: — ''The analyses revealed compositions which, while not 

 absolutely unique in petrographic literature, are seemingly 

 approached but once or twice. Vei-y unusual is the molecular 

 preponderance of potash over soda in a rock of this character 

 so high in lime." 



In dealing with glasses such as the obsidianites, the micro- 

 scope gives us no help in determining the composition, so that 

 Ave must depend entirely on chemical analyses. It should, 

 however, be noted that, having obtained a good series' of 

 analyses, the careful determination of the specific gravity of a 

 specimen should give an approximate idea of its composition. 

 It is impossible to propei'ly compare analyses, simply stated as 

 percentages of oxides, and the use of some such scheme of 

 classification as that worked out by Messrs. Cross, Iddings, 



1 Proc. Roy. Soo. \'ictoriii, vol. xi., n.s., i>t. I., 180S, jt. ."U. 



2 Bull. No. (). West Aiist. Geol. Surve.v, p. 7!). 



;! Animal Itpj). of the Sec. for Mines, Tiisinuniu, I'.H)'.. |i. 21. 



