4 Strom FA'ER, The Formation of Mmerals in Granite. 



seems to have been vaguely felt that, as has been 

 explained above, quartz may possess a greater affinity for 

 steam than the other minerals, and that the effect of this 

 occluded steam may have been to lower the melting 

 temperature of quartz. Such a lowering is possible, but 

 only after the separation of minerals has taken place ; it 

 does not account for the separation, and it does not 

 account for the formation of those granites in which the 

 quartz contains little or no water. It also fails to account 

 for the diversity of miiieralogical composition of such 

 granites as are nearly alike in chemical composition. 



The whole difficulty, to my mind, may be explained if 

 such experiments on the changes of volumes while melting 

 as have already been carried out by "Mr. J. Joly* on basalt, 

 augite and orthoclasc, and Mr. Barusf on diabase, could 

 be extended to the individual minerals which go to make up 

 granite ; provided, however, that such experiments should 

 show that, on fusing, quartz expands less than the other 

 minerals, or, better still, that it contracts like ice, bismuth, 

 arsenic, and antimony. 



These few words will no doubt at once suggest the 

 explanation which it is intended to give. 



In 1849 Mr. J. Thomson,! and in 1850 Herr R. 

 Clausius,** showed theoretically that the melting tem- 

 perature of a solid could be altered by pressure; that it 

 would rise under pressure if the volume of the substance 

 increased on fusion, as in the case of sulphur, phosphorus, 

 etc. ; and that it would fall under pressure if the volume 

 decreased on fusion, as in the case of ice, bismuth, etc. 

 This theoretical deduction was speedily verified b}- Lord 



* Trans. Roy. Dublin Soc, Ser. 2, Vol. 6 (1S97), p. 283. 

 ^ Amer. Journ. Sii., Vol. 42(1891), p. 49S. 

 XTrans. A'oy. Soc. Edinb., Vol. 16 (1849), [). 575. 

 **Pogg. Aim., Bd. Si {1850), p. 168. 



