EUTECTICS IN ROCK MAGMAS 87 



We have remarked that the eutectic ratio continues in a line 

 from micropegmatite to labradorite (Ab^ Ang). This feldspar 

 has a silica percentage of 53 and an alkali-silica ratio of i : 12. 

 This is so close to the ratios of the average rock that they are 

 within the limits of the probable error; in fact, we may say that 

 the average rock has the same silica percentage and alkali ratio 

 as labradorite. Now, we will notice upon the chart that once 

 the silica falls below this ratio there is marked change in the 

 behavior of the analyses, and they seem to stream off toward the 

 lower right-hand corner, the alkali ratio dropping with the silica 

 ratio. We might infer, therefore, that for rocks less silicious the 

 eutectic ratio above given did not hold ; or, rather, it may hold, 

 so far as the alkalis are concerned, that they still find the most 

 fusible compound Na^O • AlgOg • 6 SiOg, and 6 ROSiOg, but 

 that in the presence of an excess of bases there is some other 

 equally or more fusible compound ROSiOg. 



This, remember, is a purely theoretic inference from laws of 

 chemistry and Iddings's diagram. A moment's thought shows 

 how amply it is confirmed by petrographic research. This more 

 fusible mineral is augite. As the percentage of SiOg in augite 

 is about the same as in labradorite (between 55 and 44), and 

 cannot anywhere in the pyroxene group get above that of enstatite 

 (60), we see why the distribution of analyses turns a square cor- 

 ner, and we find them quite frequently from 0.58 SiOg down with 

 all kinds of alkali ratios. We may go on to ask if there is any 

 eutectic balance between labradorite and augite. As the ratios 

 of lime to silica and percentage of silica are practically the same 

 in the augite and in the labradorite series, the question as to the 

 predominance of the one or the other tendency will probably be 

 a question of balance mainly between the femic and alkaline 

 constituents. The femic constituents in a magma mainly of fel- 

 sitic eutectic are only slightly soluble and tend to crystallize out 

 early as magnetite, biotite, hornblende, etc., even though present 

 in very small quantities; and even though they increase markedly 

 in abundance, their solubility or fusibility is still small, except as 

 they can be taken into the augite molecule. An excess of mag- 

 netite or olivine crystallizes readily." The metallurgists tell us 



'I am neglecting a lot of minor matters. 



