MERRILL] METEOR CRATER OF CANYON DIABLO, ARIZONA 475 



of molecular strain, and the cleavage lines are themselves at times 

 more or less wavy (fig. 5, pi. lxxi). The appearance indeed is 

 such as to suggest that the granules have been subjected to pressure 

 while in an almost putty-like or plastic condition. With a high 

 power and between crossed nicols the rock is seen to be not holo- 

 crystalline, but to contain comparatively small colorless interstitial 

 areas, showing by ordinary light a fibrous, scaly structure, but which 

 are for the most part completely isotropic between crossed nicols and 

 which the chemical analysis suggests may be opal. From this condi- 

 tion the rock passes through more or less vesicular (fig. 2, pi. lxxii) 

 to highly pumiceous forms(Cat. Nos. 76,839 and 76,840,11. S.N. M.), 

 showing to the unaided eye all the features of an obsidian pumice, but 

 of a white color (figs. 3 and 4, pi. lxxii) . This under the microscope 

 is resolved into a colorless vesicular glass, more or less muddied 

 through dust-like material (fig. 4, pi. lxxi), and showing here and 

 there residual particles of unaltered quartz. The glass does not, 

 however, resemble the glass of a pumice, nor is it like that obtained 

 by the artificial fusion of quartz in the geophysical laboratories of 

 the Carnegie Institution. So far as the writer's observations go, it 

 more closely resembles fulgurite glass, formed by the lightning 

 striking in siliceous sand. It is evident that the original molten 

 material was in a highly viscous, almost dough-like condition. The 

 cavity walls are stringy rather than smooth, as in ordinary pumice, 

 and rough fibers or strings of true quartz glass stretch from wall 

 to wall, as shown somewhat indistinctly in figure 3 of plate lxxi. 

 This form, it is well to note, is not abundant and is the material first 

 met with in what Mr. Barringer has designated as shaft No. 2 and 

 at a depth of 130 feet below the crater floor. A few small pieces 

 were found in digging the open cuts outside of the crater, and others 

 lying out on the surface. 



Chemical tests on (I), the unaltered sandstone; (II), what may 

 be called the crystalline variety, the finely laminated stone compared 

 to a decomposed chert, and (III), the pumice, gave Mr. Tassin re- 

 sults as below : 



J Si0 2 99.29 



{ Undet 0.71 



100.00 



Si0 2 98.63 



A1 2 3 0.18 



Fe 2 O s 0.10 



Ign 0.99 



Loss at 100° . 30 



100.20 



(I) Unaltered sandstone. 



(II) Altered sandstone -< 



