CunnincHam—Crystallisation of Minerals in Igneous Rocks. 395 
and so it is more to the point to look for data regarding the 
behaviour of the oxides which enter into the composition of rook- 
forming minerals when mixed. Moissan' found that mixtures of 
CaO and Al,O; (also of CaO and Fe,Q,, and CaO and Cr,0;) 
were far more easily fusible than either of the oxides alone. It is 
familiar to the practical metallurgist that the addition of lime 
renders a siliceous slag much more fluid at the temperature of the 
furnace.” On the other hand, if a metallurgist®? wants a highly 
refractory material, as a rule he prefers a pure oxide, either 
CaO, MgO, or S10,, etc., but he avoids a mixture, the exceptions 
. being cases of similar oxides, and where, in consequence, the 
chemical union is weak, as, for instance, Fe,0; and Al,O; in 
Bauxite, and certain clays (Al,O; and 8i0,). We cannot, then, 
be surprised to find the melting points of the various silicates 
(which are certainly crystalline chemical compounds) distinctly 
below that of quartz or any of the constituent oxides. In fact, 
if we draw a few curves (fig. 7, p. 397) for such “alloys” they 
are quite similar to those previously shown for the metallic alloys 
(fig. 6, p. 896). ‘The melting points of CaO and A1,O; are according 
to the estimates of Moissan.* 
5. Finally, in any measurements of the melting points of quartz 
and silica, it is necessary to make sure that the specimens experi- 
mented with are pure 8i0,, and contain no alkalies in particular, 
even a trace of which would certainly lower the melting point very 
considerably. Prof. W. A. Shenstone and Mr. H. G. Lacell*® 
found that quartz very commonly contained sodium and lithium. 
It may be noted that all previous observers state that 
quartz does not melt at the temperature of a furnace. Moissan 
and Gaudin both found that the boiling point of silica was very 
little above its melting point. Mendeléeff says*:—‘ But silica 
fuses and volatilises (Moissan) in the heat of the electric furnace, 
about 8000°. iQ, is also partially volatile at the temperature 
1M. H. Moissan, ‘‘ Le Four Electrique,’’ pp. 32, 35, and 38. 
2 Cf. Mendeléeff, ‘‘ Principles of Chemistry,’’ chap. xvu1., p.;120, note 26. 
3 Cf. M. L. Gruner, ‘‘ Traité de Métallurgie,’’ vol. i., pp. 196 e¢ seg., Paris, 1875. 
4M. H. Moissan, ‘‘ Action d’une haute température sur les oxydes métalliques.”’ 
Comptes Rendus, 115 (1892), p. 1034. 
5 «6 Working Silica in the Oxy-gas Blowpipe Flame,’’ N ae 62 (1900), p. 20. 
8 Loc. cit., p. 100, n. 1 dis. 
