210 



SCIENCE 



[N. S. Vol. XXIX. No. 736 



feldspar and hornblende in the first-men- 

 tioned rock. 



That hornblendes are less stable com- 

 pounds in igneous magmas than pyroxenes 

 and numerous other minerals, is shown by 

 the frequent occurrence of paramorphs of 

 other minerals after hornblende, commonly 

 seen in so-called black borders, and the ab- 

 sence of correspondingly changed crystals 

 of other minerals. 



Another chemical principle involved in 

 the production of pyrogenetie minerals is 

 that affecting the formation of compounds 

 that possess common ions when in solu- 

 tion. It is known that Avhen there are 

 in a solution ions capable of entering two 

 or more compounds, the concentration of 

 the least soluble compound may be in- 

 creased by the entrance of ions derived 

 from other compounds into its molecules. 

 And this may proceed to the complete in- 

 corporation of the common ions within 

 one compound upon its separation in the 

 solid phase. This has sometimes been 

 called erroneously "mass action." That 

 compound forms at the expense of another 

 in any particular instance which is the 

 more stable under attendant conditions. 

 Illustrations of this action are found: in 

 the case of the complex amphibole in the 

 hornblendite of Gran already mentioned; 

 in aluminous pyroxenes (augites), which 

 contain components capable of forming 

 lime-soda-feldspars, as pointed out by 

 Pirsson, and in numerous other rock min- 

 erals. This principle is probably con- 

 cerned in the production of the lime-soda- 

 feldspars with notable amounts of albite 

 molecules, as in andesine and labradorite, 

 in maginas so low in silica as to necessitate 

 the production of leucite from the potas- 

 sium present, when the more active potas- 

 sium should have combined with the 

 silicon in a polysilicate (orthoclase), leav- 

 ing the less active sodium to enter ortho- 

 silicate (nephelite). 



Following out the discussion of all the 

 probable compounds likely to form under 

 known chemical laws from molten rock 

 magmas upon cooling, and taking into 

 consideration the relative chemical ac- 

 tivities of the several constituent elements 

 in igneous rocks, it is possible to deduce a 

 probable mineral composition for any 

 given magma, under given conditions of 

 cooling. The mineral composition of igne- 

 ous rocks then becomes a necessary conse- 

 quence of the chemical reactions likely to 

 obtain in molten rock magTuas, . and de- 

 pends not only on the kinds and amounts 

 of the elements present in each case, but 

 also on the conditions of temperature and 

 pressure modifying the chemical activities 

 of the elements and the stability of the 

 compounds. As these conditions are 

 known to vary with the experience of dif- 

 ferent magmas during eruption and solidi- 

 fication, the minerals produced in chemi- 

 cally similar magmas are not to be expected 

 to be always alike, and the variations in 

 composition are in this way understood. 



Having considered the possible chemical 

 reactions that may give rise to mineral 

 compounds in rock magmas, the next step 

 in the treatment of the subject is a dis- 

 cussion of the process and results of sepa- 

 ration of various compounds or substances 

 from magma solutions upon change of 

 physical conditions attending the eruption 

 of magmas. These may separate as gases, 

 liquids or solids, chiefly as solids. But 

 gases escape in large volumes upon the 

 eruption of lavas, mostly as water vapor. 

 There are other kinds in smaller, though 

 often in considerable, amounts. The ef- 

 fects of this loss of gases are in the chemi- 

 cal composition of the rock magma, in the 

 concentration of the remaining substances, 

 and in the viscosity of the magma, which 

 may increase notably upon loss of gas. 



Liquids, probably, do not separate as 

 such from molten magmas to any consider- 



