Bowen and Andersen — Binary System MgO-Si0 2 . 499 



We have therefore made no observations which throw further 

 light on the relation between enstatite and clino-enstatite. The 

 two are very similar in physical properties but clino-enstatite is 

 monoclinic. 



Clino-enstatite crystals have been measured and the figures 

 given in a former publication of this laboratory. It has the 

 characteristic, perfect prismatic cleavage of the pyroxenes. 

 Usually rough cracks //(001) show as well. Polysynthetic twin- 

 ning after (100) is exceedingly characteristic. The plane of the 

 optic axes is normal to (010) as would be the case in enstatite if 

 it were oriented after the manner of the monoclinic pyroxenes. 

 Sections //( 010) do not, therefore, show the highest interference 

 colors as in most pyroxenes. The angle CAc = 22°. The 

 optic axial angle is difficult to measure on account of the poly- 

 synthetic twinning. Wright and Larsen give the value 53'5°* 

 The refractive indices were determined on material crystallized 

 from a clear glass and therefore free from inclusions of silica, 

 which tend to give a value slightly too low. The results were 

 a = 1'651 ± *001, 7 = 1*660 d= "001 determined by immersion 

 in liquids of known refractive index. 



Bearing of the Results on Petrologic Problems. 



In the chemically pure products described it has been shown 

 that, over a wide range of compositions, the olivine, forsterite, 

 crystallizes out and is later either partly or completely 

 redissolved, reacting with the liquid to give the pyroxene, 

 clino-enstatite. This resorption takes place as a necessary 

 result of equilibrium and during the normal course of crystal- 

 lization. In natural rocks the resorption of olivine and the 

 formation about the olivine crystals of reaction-rims of 

 enstatite has often been noted.f Such reaction between the 

 crystals in a magma and its still liquid portion has sometimes 

 been assumed to imply some drastic change of conditions such 

 as the sudden relief of pressure. It has also been considered 

 possible that such crystals may have sunk into magma of such 

 a composition that they became unstable. In the case of olivine 

 it is probable that such assumptions are unnecessary. Olivine 

 crystals may perhaps be partly or wholly resorbed during the 

 normal course of crystallization as a simple result of cooling, 

 as in the case of the more simple mixtures described. 



It has been shown also that certain crystalline products, 

 when quickly formed, may contain forsterite and free silica 

 (cristobalite) together and this condition may persist indefinitely 

 although it does not represent equilibrium. Dr. Cross:}: has 

 found cristobalite in cavities of divine-bearing lavas from 



*This Journal, (4), xxvii, 30, 1909. 



f Harker, A., Natural History of Igneous Eocks, p. 269. 



% Personal communication. 



