THE GENESIS OF IGNEOUS ROCKS 309 



If, to select a simple example, we suppose that a magma 

 which represents a combination of olivine and diopside in the 

 proportion of 40 to 60 is slowly cooled from a state of complete 

 fusion, olivine will at a certain temperature begin to crystallise 

 out, and as it does so the percentage it forms of the magma will 

 diminish until it is as low as 32 per cent. If now there is any 

 further loss of heat, the two minerals will crystallise out to- 

 gether in the ratio of 32 to 68 until the whole rock is solid. 



If there had been originally more than 68 per cent, of 

 diopside, this would have crystallised out alone until the same 

 ratio of 32 to 68 was reached. 



A magma therefore containing olivine and diopside in this 

 proportion remains liquid at a lower temperature than any 

 other combination of those minerals and is said to be a eutectic 

 mixture. The theory of eutectic mixtures is of considerable 

 importance in the interpretation of the characters of igneous 

 rocks, and as was pointed out by Mr. Teall as early as 1888, 

 the association of quartz and felspar known as graphic granite 

 may be regarded as an example of such a eutectic combination. 



In the case of three or more substances there is in like 

 manner a eutectic mixture which consolidates at a lower tem- 

 perature than a magma containing them in other proportions. 

 As the cooling of such a magma progresses first one and then 

 two minerals will crystallise out, and so on till the eutectic 

 mixture is reached, when all will crystallise out together. The 

 order in which any two of these minerals crystallise is, however, 

 not necessarily the same as if they alone were present, but is 

 dependent on the nature and amount of the other minerals. If 

 two minerals have an " ion " in common, such as magnesium in 

 the case of spinel MgO . A1 2 3 and olivine (MgO) 3 Si0 2 (with a 

 portion of the magnesium replaced by iron), their solubility 

 will be markedly diminished, and each will be liable to crystal- 

 lise out earlier than if the other were not present. 



In an ordinary igneous magma containing the elements of 

 a number of minerals most of the basic minerals have ions in 

 common, which is not the case with the more acid minerals. 

 The former are therefore more sparingly soluble in the magma. 

 In this way the empyrical rule enunciated by Rosenbusch that 

 the basic minerals crystallise out before those which are more 

 acid in composition is in part at least explained. 



These principles are, however, complicated by a number of 



