I02 F. LOEWINSON-LESSING 



We can admit a third case (III), namely, a magma devoid of 

 magnetite, or we may suppose that the free sihcawill not combine with 

 the magnetite. This will give us SiOz 3 or 5 per cent of free silica. 



Therefore, even if we admit conditions most unnatural and 

 most unfavorable for Bowen's critics, a basaltic magma will either 

 give no free silica at all, or will give only such a small amount that 

 at best we can expect the formation of a quartz gabbro, poor in 

 quartz, but never of granite or quartz-diorite. 



Besides this it ought to be noted that Bowen^ considers clinoen- 

 statite as the only stable form under conditions of slow crystalliza- 

 tion. During slow cooling full equihbrium is established, and only 

 under conditions of artificial chilling can the minerals of the incom- 

 plete equihbrium stage (forsterite, enstatite, cristobaHte) be fixed. 

 The conditions of crystallization of the intrusive magma are those 

 bringing it to full equilibrium. 



And again, Bowen^ tells us that a mixture of 98 per cent MgSiO^ 

 and 2 per cent SiO^ will give forsterite and silica; but if there is 

 2 . 5 per cent or 3 per cent of silica, enstatite will be formed — a state- 

 ment invaHdating all presumptions on the formation of olivine and 

 free silica from a basaltic magma. 



Let us now make another calculation based on the specific 

 gravities. In Daly's Igneous Rocks (p. 202) we find the specific 

 gravities of gabbro, syenite, quartz, diorite, and granite at the 

 temperatures of 20° C, 1000° C, 1200° C, and 1300° C, in the solid 

 state and when molten. The specific gravity of crystalline labra- 

 dorite is 2.689 5 when molten to glass it shows an expansion of nearly 

 7 per cent — specific gravity 2.525. Heated to iooo°-i3oo°, if still 

 solid, the increase of volume if only equal to 3 per cent, will give a 

 specific gravity of 2.609. Now, comparing these figures with the 

 specific gravity of the liquid gabbro it is easy to see that the labra- 

 dorite will sink only in a gabbro with a specific gravity of 2.8a 

 (if molten 2.57-2.53); in all other cases the labradorite will float, 

 especially if we take into consideration that the average density 

 of gabbro is 2.933-2.975. I can easily conceive that during the 

 crystallization of a gabbroic magma there will be a tendency toward 



'N. Bowen, Amer. Jour. Sci., CLXXXVIII (1914), P- 237. 

 'Amer. Jour. Sci. (1914), p. 487. 



