438 H. S. Summers: 



What little evidence we have, therefore, strongly supports 

 this idea of provincial distribution, and if on further work this 

 is upheld, the cosmic origin of obsidianites is practically deter- 

 mined, because such distribution is impossible by means of any 

 of the agencies suggested by the advocates of a volcanic or 

 artificial origin. The agencies which have been suggested are 

 water, ice, aborigines, birds, winds, volcanic explosions and 

 hypothetical bubbles. It is quite inconceivable that chemical 

 distribution could be effected by any of these means. 



The two princijial arguments against the meteoritic hypo- 

 thesis are — the form and the composition of the obsidianites. 

 This paper is only concerned with the latter. It has been 

 argued that obsidianites cannot be meteoritic in origin, because 

 they differ so completely in composition from all known mete- 

 orites. The stonj;- meteorites are all extremely basic in com- 

 position. This argument cuts both ways, however, for we may 

 with quite as much justification say that as the obsidianites^ do 

 not agree in composition with terrestrial rocks, they are there- 

 fore extra-terrestrial. 



As the artificial origin is impossible, and us none of the sug- 

 gested volcanic sources have produced lavas at all agreeing in 

 composition with that of obsidianites, these two hypotheses 

 appear untenable. This leaves us the meteoritic hypothesis, and 

 the almost unique composition of the obsidianites, together with 

 their apparent provincial distribution, makes it practically 

 certain that this is the correct explanation of the origin of these 

 interesting substances. 



BiLLrrONITKS .\ND .VIOLDAVITES. 



Dr. Suess^ quotes tliree analyses of billitonites, but only two are 

 sufficiently complete for purposes of classification. The analyses, 

 with their molecular proportions, are as follows : — 



1 Die Herkuiift der ^^olflavite mid verwaiidter Glaser. Jalirb. d. k. k. jreol. Ufidisiinst 

 Vienna, 1900, vol. .10. 



