52 
in which he showed that their microscopical characters often serve to 
throw much light on the origin of rocks. Mr. Butler therefore placed the 
whole of his collection in Mr. Sorby's hands for careful examination, and 
it was decided that a paper should be written by the two conjointly; and 
since Mr. Sorby had previously made many experiments in connexion with 
the expansion of liquids, as already described in a paper published in the 
Philosophical Magazine*, he took advantage of the opportunity to investi- 
gate the law of the expansion of the very interesting fluid met with in the 
cavities of sapphire. 
In describing the various facts, it will be well to consider them in rela- 
tion to the following general principles : — 
(1) The structure of the various minerals as mere microscopical objects. 
(2) The physical characters of the fluid- cavities, as throwing light on 
the origin of the minerals. 
(3) The influence of some included crystals on the structure of the 
surrounding mineral. 
Sapphires. 
By far the most interesting objects contained in sapphires are the fluid- 
cavities. Their occasional presence has been already noticed by Brewsterf, 
who met with one no less than about i inch long, two-thirds full of a liquid 
which expanded so as to fill the whole cavity when heated to 82° F. (28° C). 
He thought the liquid was less mobile than that described by him in topaz, 
and could not see a second liquid in the cavity. Though many thousand 
sapphires have been examined by the authors, no such large cavity has 
been found ; but several have been met with about inch in diameter; the 
greater number are far less, and some are very minute ; and they seem to 
contain only the liquid which expands so much when warmed. The size 
of the included bubble varies much, according to the temperature. At 
the ordinary heat of a room it is sometimes equal to one -half of the 
capacity of the cavity, whereas in other cases the cavity is quite full. 
This is especially the case with the very small cavities, and is to some 
extent due to the forced dilatation of the liquid. But if we only take into 
consideration the larger cavities, the temperature required to expand the 
fluid so as to fill them certainly varies from 20° to 32° C. (6S Q to 90 e F.), 
* "On the Expansion of Water and Saline Solutions at High 
Temperatures," August, 1859, vol. xviii. p. 81. 
+ Sochting's Einschliisse von Mineralien in krystalllsirten Mmeralien, 
p. 121, who refers to Edin. Journ. of Sc.. vol. vi. p. 115. 
