IGNEOUS ROCK-MAGMAS AS SOLUTIONS 253 



that two or more of them have an ion in common, such as 

 potassium or magnesium, or the electro-negative group Si0 4 

 (in orthosilicates) or Si0 3 (in metasilicates). It is therefore 

 probable that the law of the common ion, first deduced by 

 Nernst, is an important one in application to rock-magmas. 

 Vogt invokes it more especially for the case of the minor 

 accessory minerals, which almost always crystallise very early, 

 even when they are present in very small amount. Some of 

 these minerals (spinel, chromite, zircon, etc.) have very high 

 melting-points, and it is probable that the eutectic mixture of, 

 say, spinel and a felspar contains a small proportion of spinel 

 to a large amount of felspar. This, however, would not be 

 sufficient to account for the constantly early separation 

 of spinel. But almost any natural magma will contain 

 in addition one or more constituents (olivine, pyroxenes, 

 biotite) which have one ion (magnesium) in common with the 

 spinel ; and the presence of these may displace the eutectic 

 point so far away from spinel as to cause that mineral to 

 crystallise first, even when the proportion of it in the mixed 

 magma is very small. Similarly apatite, perofskite, and sphene 

 have one ion (calcium) in common with lime-felspar ; zircon has 

 the group Si0 4 in common with any orthosilicate compound ; 

 and so for other cases. Nor is it only in respect of these minor 

 accessory constituents that Nernst's law may be supposed to 

 have importance. It may afford an explanation of the strong 

 tendency of olivine to crystallise at an early stage from magmas 

 which contain pyroxenes. It has already been remarked, as an 

 illustration of the simple eutectic law, that in the olivine-felspar 

 rocks of Rum one or other mineral has c^stallised first, 

 according to the relative amounts of the two. Associated 

 with these rocks are others containing augite and enstatite 

 in addition to the above constituents, and here it is seen that 

 the olivine has crystallised first, even when the felspar was 

 largely in excess. 



Where positive knowledge is so scanty, dogmatism would 

 be out of place ; but there seems to be good reason for believing 

 that this interaction dependent on ionisation is actually operative 

 in rock-magmas, and accounts for facts which would otherwise 

 appear anomalous. It tends to make the order of crystallisation 

 depend less upon the relative proportions of the several con- 

 stituents and more upon their specific properties. In other 



