Diopside — Forsterite — Silica. 



219 



The fact that equilibrium is readily obtained between solid 

 and liquid in these melts, and that the high temperature of the 

 points found is not due merely to a sluggishness in the solution 

 of cristobalite, was shown in a manner which is illustrated by 

 the following example. The completely crystallized mixture 

 diopside 75 per cent, Si0 2 25 per cent was held at 1575° for 

 periods varying from 15 minutes up to 5 hrs. In each case 

 the resultant product was glass and a very little cristobalite. 

 On the other hand the same mixture, again completely crys- 

 talline, was held at 1585° for only 15 minutes and the product 



Fig. 7 



1560 



1480 



1400 



Fig. 7. Vertical sections along boundary curves. AB forsterite-pyroxene, 

 CD pyroxene-silica. 



is pure glass without cristobalite. This same method of check- 

 ing the temperatures of the silica surface was applied to all 

 the points except those above 1600°. It is not advisable to 

 hold the furnace at these higher temperatures for a period of 

 5 hrs., but since it has been shown above that a period of 15 

 minutes is sufficient to obtain equilibrium, holding the mixture 

 for one hour, as was done for example with the mixture 

 MgSi0 3 75 per cent, Si0 2 25 per cent at 1625°, provides an 

 ample margin of safety to insure equilibrium. 



It may therefore be repeated that the results point con- 

 sistently to a melting-point of cristobalite much higher than 

 1625°. Cristobalite formed by other methods and consisting 

 of various molecules in metastable equilibrium in the single 

 solid phase may, however, show a beginning of melting at the 

 lower temperatures found by Fenner and by Endell and 

 Rieke. 



