202 
ME. ROBERT MALLET ON VOLCANIC ENERGY. 
not much differ from that of glass itself ; but we cannot attach much value to experi- 
ments thus made. 
164. Admitting them for what they are worth, however, the curve (Plate X.) would 
seem to represent something like the increments of the coefficients of expansion of these 
slags at various temperatures, through the range from 0° to 3680°. 
165. We now pass from the basic to the acid slags or silicates, of which British plate- 
glass may he taken as a type, as in chemical constitution not widely differing from the 
acid silicates in natural rocks, as a comparison of the following analyses shows : — 
Plate-glass (Dumas, ‘ Chim. appl. aux Arts ’). 
Silica 
73-85 
68-6 
Alumina . 
• • • • 
3-50 
1-2 
Magnesia . 
. 
55 
2-1 
Lime . . 
5-60 
11-0 
Oxide iron 
• • . • 
55 
0-2 
Oxide manganese . . 
55 
0-1 
Potash . . 
5-50 
6-9 
Soda . . . 
12-05 
8-1 
Gneiss (extremes 
of 4 analyses). 
Silica 
75-91 
66-46 
Alumina 
14-11 
16-20 
Oxide iron . 
2-03 
5-81 
Lime . . 
1-14 
2-82 
Magnesia . 
0-40 
2-17 
Potash . 
4-16 
3-98 
Soda 
1-77 
3-20 
Water . 
1-16 
1-59 
Granites and Syenites. 
A. 
B. 
c. 
D. 
Silica 
74-25 
68-56 
61-72 
56*78 
Alumina .... 
11-58 
14-44 
13-57 
16-64 
Oxide iron . . . 
2-41 
5-04 
7-16 
9-58 
Lime 
1-08 
3-85 
5-88 
5-12 
Magnesia . . . 
jio-oi 
J2-78 
3-33 
2-63 
Soda 
l „ 
3-12 
5-30 
Potash . . . . 
55 
3-36 
3-37 
2-58 
are the extremes of five analyses 
of granites, 
C and D the extremes 
analyses of syenites. 
