246 
locked up in the cavities of minerals forming fundamental 
rocks; of these the most noteworthy and essential to this 
inquiry, are the lacune or enclosures in the quartz crystals of 
granites and granulites, &c. These enclosures in quartz are 
sometimes filled with liquids, such as pure water or aqueous 
solutions of chloride of sodium or other salts, and sometimes 
they contain water holding carbon dioxide, commonly called 
carbonic acid in solution, sometimes liquid carbon dioxide itself ; 
the shape of the cavities is irregular, or otherwise, varying in 
size; some may be seen with the naked eye, but in general their 
largest diameter would be 0:06 millimetre. The smallest are 
not visible without the aid of a microscope, with an augmentation 
of 700-800 diameters, in which case more than 120 occlusions 
have been counted within the space of the 100th part of a 
square millimetre. In respect of the temperature and pressure 
at which the quartz containing them was consolidated, Mr H. 
C. Sorby estimated in the course of his valuable researches upon 
the subject, that in the quartz of a trachyte from the islands 
of Ponza, the temperature was probably 356°. Some important 
observations have been made by Bischof. on the contraction of 
the igneous rocks, as they pass from a fluid or pasty state to a 
consolidated condition. It has been ascertained experimentally 
that rocks expand on being heated, and contract on cooling, 
this contraction would affect the number of atmospheres of 
pressure upon the contents of the lacunz. Bischof got in his 
experiments, the following results for granites :— 
Volume in the state of glass. In crystalline state 
Granite ... ax eae i Nia “re x .. 0.8420 
In the fluid state In crystalline state 
Granite ... Ss ase ie ge ee a+ -. 0.7481 
From this it would appear that granite contracts 25 per 
cent., or a quarter of its volume in passing from a fluid to a 
crystalline state, and 16 per cent. in passing from a glassy to a 
crystalline state. M. Deville and M. Delesse (Bul. Soc. Geol. 
France, 2nd ser. iv. p. 1312) arrive at results rather different 
from Bischof’s, and M. Delesse gives the following as comprising 
