Diopside — Forsterite — Silica. 255 



According to this view, then, quick cooling, such as occurs in 

 meteorites and artificial melts, gives coarse twinning which is 

 visible under the microscope, whereas relatively slow cooling, 

 as in terrestrial rocks, gives line, sub-microscopic twinning. 

 During the course of the present work it has been found that 

 quick and slow cooling have precisely the opposite effects, at 

 least, for these pyroxenes. When the crystals form instan- 

 taneously from an under-cooled liquid the twinning is very fine ; 

 when they are grown by holding for a couple of hours in con- 

 tact with liquid the twinning bands are broad. It seems 

 extremely unlikely, then, that the crystals of enstatite of ter- 

 restrial rocks, which undoubtedly grew in contact with liquid 

 for a considerable period, would have twinning bands so fine 

 as to escape observation. If twinning were present in crystals 

 formed under these conditions the twinning bands would be 

 broad and their presence unmistakable. The fact that no 

 twinning bands are discernible leads the writer to believe that 

 natural enstatite crystals cannot be identical with clino-enstatite 

 and that the relation is not merely one of polysymmetry. 



A further indication that enstatite and clino-enstatite should 

 be considered different allotropic forms is, of course, the fact 

 that clino-enstatite and diopside form a complete series of solid 

 solutions, whereas natural minerals point to limited solid 

 solutions between diopside and enstatite. 



It should be noted, moreover, that no orthorhombic pyroxene 

 corresponding with enstatite ever crystallizes from any of the 

 mixtures here studied, when they are crystallized by cooling 

 from the liquid state. The writer believes that the so-called 

 enstatite, often described as occurring in products formed by 

 treating like mixtures in this manner, is really forsterite, for 

 forsterite should be present and its presence is not noted. 



Bearing of the Results on Petrologic Problems. 

 The Series of Lime-Magnesia Pyroxenes. 



The experimental work described throws some light on the 

 physico-chemical meaning of zoned pyroxenes. In the mix- 

 tures dealt with, such zoning sometimes occurs and when 

 formed during the normal course of cooling the inner zones are 

 more magnesian, the outer more calcic. To be sure we are 

 dealing with purely lime-magnesian pyroxenes, but it might be 

 well to recall here that, in spite of the fact that we are prone 

 to speak of the pyroxenes as ferro-magnesian minerals, the 

 common pyroxene of igneous rocks usually contains lime, 

 magnesia and iron in proportions in the order named. 



Reference has already been made to the enstatite-augites of 

 Wahl and attention called to their similarity in optical proper- 



