102 



Allen and White — Polymorphic Forms of 



Curve II. — Melting of Pseudo-Wollastoiiite. 

 Melting Point, 1512°. 

 Eate of Heating, 2 • 7° per minute. 



17000 



1550 8 



15500 



10 20 



Time in Minutes. 



position. We have already shown that the liquid can be 

 undercooled to a glass ; it follows naturally that by a suita- 

 ble disturbance a sufficiently undercooled melt should crystal- 

 lize directly to wollastonite. In fact, we have in one or two 

 instances thus obtained well developed rosettes which the 

 microscope showed were optically negative, and in one case 

 the whole charge excepting a few surface grains yielded wollas- 

 tonite. That wollastonite rarely forms in this way is due, 

 first, to the difficulty of undercooling the melt sufficiently, and 

 second, to the release of the heat of fusion which tends to raise 

 the temperature again beyond the inversion point. 



The addition to calcium silicate of fluorides or borates in the 

 proper proportion (to which some investigators have resorted), 

 of course lowers the temperature of crystallization ; and it 

 is to this influence rather than to any mysterious " mineral- 

 izing " action that the synthesis of natural wollastonite is to be 

 ascribed. It seems not impossible also that for a similar 

 reason Gorgeu may have obtained it together with chloro- 

 silicates, as he claims to have done, by the addition of calcium 

 chloride. So in the glasses of commerce which contain much 

 sodium silicate, a slow cooling sometimes gives rise to crystals 

 of wollastonite below the inversion point. 



Although the temperature at which ivollastonite may crys- 

 tallize from a magma is conditioned by the composition of the 



