WRIGHT! LITHOPHYSAE IN A SPECIMEN OF OBSIDIAN 369 



the spherulites; new crystal compounds, such as tridymite and 

 fayalite, were formed which are absent in the solid spherulites 

 and in ordinary rhyolites, and indicate physico-chemical condi- 

 tions of formation different from those which ordinarily obtain 

 during the crystallization of a silicate magma. The pressure 

 of the liberated volatile components aided effectively in the origi- 

 nal formation and subsequent enlargement of the lithophysal 

 cavities; this is evident both from the shape of the outer walls of 

 the cavity whose radius of curvature is not constant and whose 

 thickness varies inversely with the radius of curvature, and also 

 from the fact that segments of the radial spherulite were forced 

 apart as crystallization proceeded. All these phenomena were 

 observed in the Icelandic lithophysae; and the conclusions there 

 drawn are corroborated in detail by the present specimen, although 

 here the lithophysae are less symmetrical and the mechanism 

 of the enlargement of the cavities is less difficult to understand. 

 Additional evidence in support of the gas pressure hypothesis 

 is presented by three parallel bands or planes which traverse the 

 specimen and probably represent contact planes between por- 

 tions of the thick viscous lava which flowed together; flow lines 

 occur in the obsidian parallel to these planes, which are now 

 marked by the crystallization of minute, chiefly solid, radial 

 spherulites. The fact that relatively few of the lithophysae are 

 elongated parallel to these flow planes proves that the cavities 

 are not original vesicles, from which crystallization subsequently 

 spread in the later stages of the flow; the directions of expansion 

 of the cavities bear, moveover, no relation to the structural planes 

 in the obsidian as they should do, were the lithophysal cavities 

 the result solely of contraction of the magma on cooling. In 

 this occurrence, therefore, the evidence is in favor of the hy- 

 pothesis that in the formation of the lithophysal cavities volatile 

 gases set free during the crystallization of the spherulites were 

 the active factor, and not a secondary phenomenon accompany- 

 ing the opening of the cavities by hydrostatic tension. 



