144 A. L. DAY MINERAL RELATIONS FROM LABORATORY VIEWPOINT 



question, in entering a new field, whether progress is most effectively 

 made by pushing forward rapidly with hastily gathered approximate data 

 or by a slower procedure, with greater attention to the magnitude as well 

 as the kind and direction of the forces operating. The first plan may, 

 and indeed often has, led the investigator far astray ; the second, with its 

 slow advancement, taxes his patience as well as that of others who are 

 waiting to make application of his results. Perhaps the choice should be 

 left to the temperament of the investigator himself. Be that as it may, 

 it has seemed wise, in organizing a perm-anent laboratory for the investi- 

 gation of geophysical problems in a broad way, to adopt the slower proce- 

 dure, on the general ground that the geologist Avho will use the data ob- 

 tained by the laboratory will be unwilling to take the chance of being led 

 astray by approximations, knowing that at the time when the approxima- 

 tions were made measured data could have been obtained with a little ad- 

 ditional effort and the resulting confusion and revision of inferential 

 conclusions avoided. Furthermore, errors in the investigation of min- 

 erals have the habit of becoming cumulative, and may very soon vitiate 

 any kind of productive conclusion if allowed to remain large or if care- 

 lessly treated. The plan determined on was, therefore, to undertake a 

 quantitative and thorough investigation of the properties of the chief 

 component minerals and of the forces which are operative in rock forma- 

 tion, drawing on the resources of the exact sciences to any extent which 

 might prove necessary. The present paper is devoted to the consideration 

 of some of the phases of this problem. 



Temperature Measurements are now^ Trustworthy and of ade- 

 quate Range 



The study of the crystallization of minerals from the magma has 

 usually been begun by determining the melting or solidifying tempera- 

 tures of the original component minerals on the supposition that the 

 order of the melting points^ and the character of those ph5^sical and 

 chemical properties which can be studied in the vicinity of the melting 

 point will provide one of the clues to the order of crystallization. Barus* 

 recognized that the first step toAvard such a study was to provide an accu- 

 rate high temperature scale in terms of which these melting temperatures 

 could be ascertained and expressed, and to it he devoted several years of 

 great activity. "With the same purpose, five years of continuous study 

 have been given to this problem in our own laboratory, and the results 



3 In the normal case of pure elements, or compounds, in equilibrium, melting and 

 solidifying temperatures are identical. There are many apparent exceptions among the 

 minerals, some of which will be treated on a later page. 



* Bulletin No. 54, U. S. Geological Survey. 



