CunnincHoam—Crystallization of Minerals in Igneous Rocks. 405 
supplied us by the difference of the densities of the crystallized 
mineral and of its glass, z.c. on the assumption that the coefficient. 
of expansion of the substance is the same in the two states. This 
assumption is of course not really a safe one to make; but the 
coefficients of expansion are not, in all probability, very different. 
They are, moreover, comparately simple, physical measure- 
ments to make; and, knowing them, together with the volume 
change on melting of the mineral glass as supplied us by such ex- 
periments as those of Dr. Joly,’ our knowledge of the volume 
change of the crystalline mineral on fusion would be complete. 
Thus, to take a specific example, I have drawn a diagram (Plate 
' XXII.) to show the expansion of quartz on melting. I have 
got from available experimental data the coefficient of expan- 
sion of quartz, taken as 0:00004,? (slope of DB) the melting 
temperature, and the densities of quartz and amorphous silica at 
0°C. Now, if I also knew the coefficient of expansion of fused 
silica (slope of HF) and its behaviour on fusion, I should know 
all that is necessary. For reasons given above (p. 389), I have 
assumed that there is little or no sudden volume change on melting 
amorphous silica. The slope of EF in the diagram (Plate XXII.) 
is a pure assumption. However, it seems likely that the volume 
change on melting is not very different from the difference of 
specific volumes of the two bodies at ordinary temperatures. I 
have rounded the corners at B and F in accordance with the ob- 
servations of Mr. Ralph Cusack.? I have inserted a line to repre- 
sent provisionally the behaviour of the form of silica obtained 
by Moissan‘ by distillation in his electric furnace. The density of 
this substance is stated to be 2:4, and the little spherules exhibited 
dimples owing to sudden contraction on solidification. I had 
hoped to have obtained specimens of these spherules, and to have 
measured the volume change (DF) on consolidation, assuming that 
the dimple was caused by the molten spherule forming a solid skin 
which was adapted to contain its melted volume, but was too large 
for the contracted solid volume. This contraction, together with 
1J. Joly, “On the Volume Change of Rocks and Minerals attending Fusion,’’ 
Trans. Roy. Dublin Soc., Ser. 11., vol. vi. (1897), p. 283. 
2 Kopp, ‘‘ Ueber die Ausdehnung einiger fester Korper durch die Warme,’’ Ann. 
Chem. Pharm., 81 (1852), 1. Pogg. Ann., 86, 156. 
3 Loc. cit., p. 407. 
£M. H. Moissan, ‘‘ Le Four Electrique,”’ p. 50. 
