l6'2 A. L. DAY MINERAL RELATIONS FROxM LABORATORY VIEWPOINT 



puzzling to the laboratory student to hear the petrologist ascribe the low 

 formation temperature and several physical characteristics of rocks to the 

 effect of pressure, when he knows that pressure has relatively little effect 

 upon a single pure mineral at any stage in its formation process. Thirdly, 

 as has been noted before, the laboratory observer reports that liquid silica 

 and many acid silicates are enormously viscous, while the field observer 

 finds abundant evidence of extreme fluidity during their crystallization in 

 nature. All three of these difficulties of deduction become clearer imme- 

 diately we take into account the effect of volatile ingredients. The melt- 

 ing temperature, as we have seen, is much lowered by such substances, the 

 viscosity is enormously reduced by them, and finally, the pressure is neces- 

 sary to hold these volatile ingredients in solution. The very fact that the 

 pure silica of the laboratory, or a pure natural silica, behaves in a way so 

 different from the obvious behavior of the same substance during its 

 initial crystallization in nature is of itself proof that its behavior has 

 been influenced by the action of other substances, and if these are no 

 longer present, or have left mere traces, we must assume that they were 

 volatile and endeavor to restore them. This is the next step in the labo- 

 ratory problem, and one in which supporting field evidence will be of the 

 greatest assistance. 



Single Minerals — Summary 



In the preceding pages the effort has been made to describe some of the 

 properties of single minerals near their melting temperatures as they have 

 been developed by laboratory study. It has been more or less inevitable 

 that emphasis should be laid on what may be called the disturbing ele- 

 ments like the effects of viscosity in preventing the establishment of 

 equilibrium during the change of state, and so delaying, or even prevent- 

 ing, crystallization from the liquid, in concealing the outward evidence of 

 melting, or in producing cryptocrystalline masses. The effect of small 

 amounts of impurity in lowering the melting temperature and preventing 

 the accurate interpretation of melting-point determinations and the effect 

 of the relatively poor heat conductivity in causing considerable local 

 irregularities of heat distribution in the mineral under observation have 

 been frequent causes of misunderstandings and misinterpretation in labo- 

 ratory study. Little has been said of the successful work of Barus and 

 others in establishing the temperature scale which is absolutely indis- 

 pensable to any quantitative research in this field. 



It has been shown that these difficulties which have been encountered 

 by the laboratory have merely served to develop important phases in 



