1886.] Alteration by Heat in certain Vitreous Rocks. 435 



Microscopic examination of a section of this obsidian in its normal 

 condition shows the presence of numerous trichites, resembling small 

 eyelashes, and often occurring in radial or stellate groups ; globulites 

 are also plentiful, but, with the exception of these and some minute 

 gas-pores, the obsidian is remarkably free from enclosures, although 

 here and there a few porphyritic felspar crystals and a spherule or 

 two occur. This was, therefore, regarded as a very favourable speci- 

 men to experiment upon. 



The appearance of a section of this rock in its normal condition is 

 represented in fig. 5, Plate 4, as seen under a magnifying power of 

 570 diameters. In this drawing rather faint indications of banding 

 are shown. Part of the same section is also represented in fig. 7, 

 magnified 570 linear, in which some of the trichites are visible. The 

 specimen which was heated for 216 hours has developed an exceed- 

 ingly vesicular structure, but apart from this it appears to have 

 undergone but little change. The vesicles are large, fig. 6, Plate 4. 

 Their sections are nearly all circular or approximately so. The 

 trichites seem to have disappeared entirely, but some small opaque 

 granules are now visible, and in some instances they have distinctly 

 triangular sections. These by reflected light appear black and are no 

 doubt magnetite. 



In order to ascertain what change would take place by further 

 heating, another fragment, taken from the same specimen as the pre- 

 ceding, was heated for a period of 701 hours at a temperature of from 

 850° to about 1100° C. 



The rock is still vitreous, but a marked change has occurred. The 

 specimen had no sand adhering to its surface, and it perfectly pre- 

 served its original external form, the conchoidal fractures and sharp 

 cutting edges remaining quite distinct, but the outer surface has 

 merely a dull sub-resinous or flint-like lustre, although on freshly 

 fractured surfaces the lustre is quite vitreous. Here and there upon 

 the surface very slight elevations occur, and these are mostly per- 

 forated by a diminutive hole, as if made with a common sewing- 

 needle. When a point was inserted in one of these apertures and the 

 crust was prized off, a remarkably cavernous interior was exposed, 

 the cavities appearing to have been formed by the coalescence of more 

 or less spherical vesicles, averaging from a quarter to half an inch in 

 diameter. 



In these cavities white crystalline pellets were found, for the most 

 part about a third the size of the cavities in which they respectively 

 occurred. 



Three of these pellets are represented in the middle line of figures on 

 Plate 4. Some of them are approximately spherical, and they are usually 

 crystalline crusts, either empty or enveloping a smaller pellet of the 

 same kiud. The walls of the cavities in which they occur are also at 



2 g 2 



