82 Ferguson and Merwin — The Ternary System. 



s Previous Investigations. 



A somewhat brief review of the results of the earlier 

 workers in this field will be given. The component 

 oxides will be considered first. 



Lime, CaO. 



The melting point of lime has been determined by 

 Kanolt 2 as 2570°. Two crystalline modifications, both 

 isometric, appear to exist; 3 the one found at ordinary 

 temperatures has perfect cubic cleavage and a refractive 

 index of 183. 4 In the ternary melts it has always 

 appeared in rounded grains. 4 



Magnesia (periclase), MgO. 



Kanolt 5 places the melting point of magnesia at 2800° C. 

 Only one crystalline form is known, which is isometric, 

 with perfect cubic cleavage and refractive index 6 of about 

 1-737. It has been observed in melts as rounded grains, 

 sharp octahedrons or cuboctahedrons, and skeletal octa- 

 hedrons. 7 



Silica, Si0 2 . 



The several crystalline forms of silica have been thor- 

 oughly investigated by C. N. Fenner. s Of these only 

 tridymite and cristobalite occur as primary phases in 

 this ternary system. Tridymite appears as thin plates 9 

 or platy aggregates 10 having refractive indices 11 of a = 

 1469, y = 1473. 12 Cristobalite appears as aggregates, 10 

 octahedra and cubes : 9 a =1484, y = 1-487. 8 It melts at 

 1710 ± 10 °C. 12 The sluggish transition between cristo- 



2 J. Wash. Acad. Sci., 3, 315, 1915. 



3 For summary of evidence see J. Wash. Acad. Sci., 5, 567, 1915. 



4 Kankiii and Wright, this Journal (4), 39, 1, 1915; Sosman, Hostetter, 

 and Merwin, J. Wash. Acad. Sci., 5, 566, 1915. 



5 J. Wash. Acad. Sci., 3, 315, '1915. 



Slight differences in refractive index indicate solid solution under some 

 conditions. Mallard, Bull. Soc. Min. Fr., found 1-7364; Wright, this 

 Journal (4), 28, 325, found 1-734 ±-002; Sommerfeldt, Centralbl. Min. 

 Geol. Pal., 1907, 213, found 1-7350. We have found 1-7375 for some fused 

 magnesia of optical quality furnished by the Alundum Co. The dispersion 

 of this sample follows: C = 1-7335, F = 1-7475. 



7 Definite crystals have been observed during this study. See also Kan- 

 kin and Merwin, J. Am. Chem. Soc, 38, 570, 1916. 



s This Journal (4), 36, 331, 1913. 



°N. L. Bowen, this Journal (4), 38, 245, 1914. 



10 G. A. Eankin and F. E. Wright, this Journal (4), 39, 1, 1915. 



11 Fenner 's values were confirmed by Schaller, when account is taken of the 

 higher values — probably caused by solid solution — observed for this natural 

 material. See note 12 below. 



12 J. B. Ferguson and H. E. Merwin, this Journal, 46, 417, 1918. 



