150 A. L. DAY MINERAL RELATIONS FROM LABORATORY VIEWPOINT 



appearance of rounded corners as a criterion that the melting point had 

 been reached. It may have been passed long before. Furthermore, with 

 respect to this particular illustration, if he has had wide experience he 

 may reason that crystallization represents the operation of a force of 

 characteristic magnitude operating between the molecules to maintain a 

 given systematic arrangement (crystal form), while the rounding of the 

 corners of a thin fragment represents the tendency of all molecular sys- 

 tems to reduce to the form which has the smallest surface — that is, a 

 sphere. Bearing in mind that the effect of such a surface tension will be 

 greater the smaller the sliver and the sharper its corners, without regard 

 to its melting temperature, it will be conceivable that the rounding of the 

 edges, which he has arbitrarily determined upon as his criterion, may 

 represent the prevailing of one internal force over another, in which the 

 observed temperature bears no relation to the melting point or change of 

 state whatever. 



The classic experiments of Barus to determine the change of density 

 of a mineral upon solidification offer another illustration of the difficulty 

 of determining the physical constants of minerals which cool to glass 

 with little or no crystallization. All of his observations (save one) were 

 made on cooling diabase, and he has left an unqualified record that the 

 melting rock cooled to glass without crystallization in each case. There 

 was therefore no discontinuity of molecular structure, no characteristic 

 change of properties, no release of the latent heat of fusion — in short, no 

 solidifying point whatever on the cooling curve, and therefore no tem- 

 perature at which observations of density would have significance to the 

 student of rock formation.^ 



Their Effect upon the early Measurements 



All these observations of the peculiar properties of individual minerals 

 in the vicinity of their melting temperatures which have been revealed by 

 laborator}^ study subtract something, it is true, from the value of existing 

 records, in obtaining which it is known that certain necessary precautions, 

 of which an outline has been given above, were not recognized, and there- 

 fore not taken, but on the other hand they add immensely to our knowl- 

 edge of the behavior of minerals during the process of crystallization, 

 and offer positive reasons, where none existed before, for some of the 

 anomalous observations which have been made on natural rocks, from 



® The properties of mineral glasses were wholly unstudied at the time when Barus's 

 experiments were made, but the sudden change in density, amounting to 3 per cent, 

 which he obtained, Is not explained by any known property which they have been found 

 to possess. 



