512 Dr. Shand — Saturated & Unsaturated Igneous Rocks. 



that "all the visible alkaline rock of the world probably constitutes 

 less than 1 per cent of the total visible igneous rock ". If in 

 addition to the alkaline rocks we consider the olivine, melilite, 

 melanite, and coninduin-bearitig rocks, it seems likely tliat the total 

 mass of the undersaturated rocks will scarcely exceed 2 per cent of 

 the total visible igneous rock. 



2. Among major intrusions the vast majority consists of over- 

 saturated rocks. Saturated rocks have a much smaller representation, 

 and undersaturated rocks are relatively rare except as marginal facies 

 and differentiated bodies. Where nephelite-syenites occur they are 

 in surprisingly many cases associated with limestones, and Daly^ 

 makes out a good case for regarding some of them, at all events, as 

 due to absorption of limestone. Thoroughly unsaturated types 

 (e.g. dunite) never form trulj' major intrusions. 



3. Among lavas and minor intrusions the oversaturated types are 

 quite subordinate to the saturated and undersaturated. Thoroughly 

 unsaturated types, though rare, are recognized in dunites and some 

 monchiquites and alnoites. Undersaturated types, such as nephelite 

 and leucite tephrites, basanites, and basalts, olivine dolerites and 

 basalts, picrites, peridotites, and serpentines, are well represented. 



4. A curious point, whicli will, I think, survive the test of. 

 statistics, is the very frequent occurrence of undersaturated and 

 unsaturated rocks in the pipes of single-explosion volcanoes. The 

 olivine bombs of the Dreiser Weiher and other volcanoes of the Eifel 

 and Auvergne, and the ultrabasic agglomerate with content of olivine 

 and pyrope which fills the pipes of some of the volcanic necks of 

 Scotland, will serve to illustrate this point, but the most superb 

 examples are the kimberlite and alnoite pipes of South Africa. 

 Kimberlite, with its abundant olivine and pyi'ope, and its possible 

 melilite,^ is emphatically undersaturated; while some of the alnoites, 

 as described by Rogers,^ are entirely unsaturated. 



These observations can be reduced to the following form : Those 

 igneous rocks which have been brought up most rapidly from the 

 earth's interior', and have solidified most rapidly in or on the crust, 

 are to a marked extent undersaturated. Those others which have 

 slowly worked their way up into the crust (and have hence had 

 abundant opportunity for absorbing silica) are found to be pre- 

 dominantly oversaturated. We have here a strong suggestion that 

 undersaturation may be characteristic of the deeper zones of the 

 lithosj)here, as oversaturation is of the higher. I do not wish to 

 insist upon this point, but merely to show that the distinctions 

 employed here have their uses even in the discussion of the major 

 problems of geochemistry. 



The conceptions of saturation and undersaturation are capable of 

 application to the classification of igneous rocks, and provide just 

 those natural distinctions between different types, the absence of 

 which petrographers have been accustomed to deplore. If writers 



^ " Origin of the AlkaUne Eocks " : Bull. Geol. See. Amer., 1910. 

 ^ Carvill Lewis, 1897 ; Mennell, 1909. 



^ Trans. S.A. Phil. Soc, 1904; Ann. Eep. Geol. Comm. (Cape of Good 

 Hope), 1911. 



