180 



JV. L. Bowen — Crystallization — 



A mixture of 92 per cent of the pyroxene, diopside 56 per 

 cent, MgSi0 3 44 per cent, with 8 per cent Si0 2 was placed in one 

 of these crucibles and after raising the temperature to about 

 1500° it was lowered slowly to 1420° and held at that tempera- 

 ture for varying periods. At this temperature the mixture 

 should consist of about 95 per cent liquid and 5 per cent of a 

 pyro-xene much richer in magnesia than is the liquid. Quite 



Fig. 2. 



3 







&* 



y.«- 









:> rt 





Fig. 2. Forsterite crystals in glass, showing rapid passage from lower 

 portion rich in forsterite to upper portion free from forsterite. x 25. 



definite sinking of pyroxene crystals was observed. After 

 3/4 hr. about l/2 cm from the surface of the liquid was free from 

 crystals as a result of their sinking towards the bottom. In 

 2 1/2 hrs. all the crystals had collected on the bottom, about 

 l*7 cm of the liquid above them being entirely free from crystals. 

 Tlie pyroxene crystals so produced were stout prisms 0'5 mm 

 long, remarkably large for crystals obtained in small crucibles 

 without a special flux. 



