154 A. L. DAY MINERAL RELATIONS FROM LABORATORY VIEWPOINT 



magnitude, and to do this successfully requires that each variable shall be 

 capable of independent determination. Without these precautions it will 

 be impossible even to distinguish those forces and components which are 

 essential from those which are merel}^ incidental to his problem. 



Preparation of pure Minerals 



This conclusion might perhaps have been reached ft priori, but in com- 

 mon with man}^ conclusions of like fundamental significance it has waited 

 to be brought out by a long and painful experience of experimental dis- 

 appointment. But our attitude toward the new science must be con- 

 structive, and not destructive. If the accumulated experience of years 

 has gone to demonstrate that quantitative experimental petrology depends 

 on obtaining the minerals separately — that is, chemically pure — and de- 

 termining their individual and characteristic properties, and afterward 

 on our ability to combine them in known relations, the question which 

 presses hardest for answer is obviously whether or not chemically pure 

 minerals can be obtained for laboratory study by practicable processes. 

 To this most vital question all the experience thus far gathered by the 

 Geoph3^sical Laboratory goes to establish the affirmative answer. Not 

 much is to be expected from a more diligent search for purer natural 

 types than those already gathered; neither is there much encouragement 

 for the successful purification of natural minerals by laboratory process 

 with the single exception of quartz, but mineral synthesis from chemically 

 pure ingredients has been almost uniformly successful. 



Here again the difficulties encountered by the laboratory in the syn- 

 thetic preparation of minerals contain information for the student of 

 natural formations. ^NTot all molten minerals, as we have seen, can be 

 made to crystallize within the limited time available in the laboratory. 

 Such important minerals as quartz and the alkaline feldspars will not 

 crystallize in a furnace at any temperature unless volatile ingredients are 

 present to give the required molecular mobility, and thus to facilitate 

 molecular rearrangement. With the help of such volatile constituents, 

 pure quartz crystals can be obtained of a size and perfection sufficient for 

 determinations of high accuracy. It therefore suggests itself quite natu- 

 rally that we are approaching rather than departing from natural processes 

 in crystallizing quartz out of a solution containing volatile ingredients 

 which participate in the formation process, but do not appear in the fin- 

 ished product. There is abundant field evidence that vein quartz has at 

 some time possessed mobility sufficient to enable it to penetrate into the 

 minutest cracks, and yet but insignificant traces now remain of any other 



