86 Ferguson and Menvin — The Ternary System. 



ature relations lare shown in fi.g. 3. This is a reproduc- 

 tion of the diagram given by Bowen and Andersen 25 upon 

 which the melting point of cristobalite has been cor- 

 rected. The following descriptions of these compounds 

 are given by the same investigators. 



Clino-enstatite, MgO.Si0 2 , Monoclinic; polysynthetic 

 twinning after (100) is exceedingly characteristic. 26 

 The plane of the optic axis is normal to (010). The 

 angle y A c = 22°. Refractive indices: a = 1-651, y = 

 1-660. 



Fig. 5. 



1700 



1650 



1600 



1550 



1500 



1450 



WOO 



/350 



Fig. 5.— The binary system 2MgO.Si0 2 -CaO.Mg0.2SiOo. wt. per cent. 



Forsterite, 2MgO.Si0 2 . <* = 1-635 1? fi — 1-651 , y = 



1-670 ; 2 V = 85 ° 16'. Optically positive. 27 



Partial Studies of Ternary System CaO-3IgO-Si0 2 . 

 Besides these binary systems dealing with the oxides, 

 several systems have been studied which form part of the 

 ternary system itself. (See fig. 5.) The first of these 

 systems is the system CaO.Si0 2 -MgO.Si0 2 . This system 

 may be divided into two parts, the ternary compound 

 CaO.Mg0.2Si0 2 , called diopside, representing the point 

 of division. Of the two resultant systems, the system 

 CaO.Mg0.2Si0 2 -MgO.Si0 2 has been shown by Bowen to 

 be not a true binary system and will be discussed when 

 our later work is considered. The system CaO.MgO. 

 2Si0 2 -CaO.Si0 2 was first studied by Allen, White, 

 Wright and Larsen 28 and their results were interpreted 



25 N. L. Bowen and Olaf Andersen, this Journal (4), 37, 487, 1914. 

 2,J In ternary melts with magnesia, twinning may be rare, this Journal, 

 45, 302, 1918. 



27 Allen, Wright, and Clement, this Journal, 22, 391, 1906. 

 25 This Journal (4), 27, 1, 1909. 



1700 



/ i 

 s 



1650 



/ 

 s 



1600 





1550 



-CccO-JX 9 o-z^o e / ef&>->S*i -*& 



1500 



- +J*Celt-y / 



1450 



// 



1400 



- / / 



1350 



CaOJ^O 2S^O £ -t-ZJVgO-SjuOz 



1 1 1 1 1 1 1 1 1 1 1 1 1 1— — i 1 1 1 1 



