GEOPHYSICAL LABORATORY. 99 



upon petrogenesis. Several papers by Dr. Walter P. White (of which brief 

 reviews are appended) are devoted particularly to a detailed exposition of the 

 character of mineral melting-points and specific heats which may be readily 

 observed and the necessary apparatus prepared in almost any good physical 

 laboratory without extravagant expenditure. We count this one of the most 

 important steps in the progress of the present undertaking, 



The work of the past year has been especially successful in three principal 

 directions : 



(i) The specific heat of several pure mineral types has been successfully 

 measured up to 1500° C. with an accuracy comparable with that obtained in 

 similar measurements at ordinary temperatures and under the most favorable 

 conditions. 



(2) A two-component mineral system has been successfully worked out in 

 which temperatures as high as 2100° C. (more than 300° above the plati- 

 num melting-point) were necessary. 



(3) Apparatus has been developed for exposing minerals to measured 

 pressures of 17,000 atmospheres and temperatures as high as 700° at the 

 same time. The methods of pressure measurement in these preliminary tests 

 are somewhat crude and are capable of considerable refinement, but the 

 development of apparatus in which so much energy could be concentrated in 

 a small space, maintained under control and measured, offers greater hope 

 of success in approximating to the more inaccessible conditions of rock form- 

 ation in nature than was at first anticipated. These measurements have been 

 prepared for publication in current numbers of the scientific journals and 

 will be reviewed briefly under the published work of the year. 



GEOLOGIC THERMOMETRY. 



The question whether more exact and extensive knowledge of the char- 

 acteristics of the minerals, both alone and in combination, will enable the tem- 

 perature of formation of individual portions of the earth to be established 

 with certainty, is one of immense geologic importance. The accumulation 

 of data of the kind and quality required for this purpose, upon a great num- 

 ber of minerals, is necessarily slow, but the experience of the Laboratory so 

 far appears to indicate conclusively that it is only a matter of time when 

 positive conclusions of this character can be made. 



The reasons for this statement can be given briefly. One in particular 

 has long been recognized in geologic literature and is already familiar. The 

 order of segregation, or differentiation as it is frequently called, of individual 

 minerals and groups of minerals from a given magma (solution) can be 

 established in the laboratory and identified in natural rocks, first for simple 

 cases and afterward for more complicated ones. This may not be done just 

 in the way in which it has been attempted heretofore, by inference from the 

 size of the grains, the character of the inclusions, etc., although this is all 



