126 CARNEGIE INSTITUTION OF WASHINGTON. 



gaseous at those temperatures where the formation process is active, 

 was initiated. Valuable results have already been obtained from 

 work on systems with water or CO2 as volatile ingredients in com- 

 bination with certain of the common components of igneous rocks. 

 The experiments are of rather limited scope thus far, and only the 

 simplest combinations have been tried, but the results are definite 

 and in full accord with existing physico-chemical theory as hitherto 

 applied to aqueous solutions of salts at ordinary temperatures and 

 comparatively low pressures. Three papers have thus far been pub- 

 lished upon this subject, of which brief reviews will be found under 

 numbers 19, 24, and 27. The laboratory technique of such studies 

 at high temperatures and pressures is extraordinarily difficult, as has 

 been explained in the annual report of this department for 1912, 

 but the definite and concordant results thus far achieved furnish a 

 more than sufficient reward for trouble already encountered and an 

 adequate incentive for further effort in this domain, which is of para- 

 mount importance to our knowledge of rock formation in nature. 



CALORIMETRY. 



Recent progress in the study of the calorimetry of the minerals 

 may be briefly summarized as follows : 



(1) Some new calorimetric devices, already briefly described in 

 print, have been installed and tested. They are intended especially 

 for obtaining high precision in protracted determinations. 



(2) The calorimetric platinum-wound furnace has again been 

 studied, remodeled so as to diminish some accidental errors trouble- 

 some in previous work, and recalibrated. 



(3) The specific heats have been partly determined for platinum, 

 various forms of silica, and 15 silicates, some crystalline and some 

 glassy, throughout the temperature interval from 300° to (in some 

 cases) 1500°. This work, which is still in progress, shows accidental 

 errors of a magnitude less than 0.0005 of the quantities measured. 

 The agreement with the best previous work of this Laboratory is 

 also entirely satisfactory. 



Among the more general results already obtained are the following : 

 The specific heat of platinum shows an increase, which to the full 

 precision of the measurements, is uniform from 700° to 1300°. This 

 seems likely to prove an important criterion of certain molecular 

 theories which indicate that it ought not to increase in this way. 

 There is a slight divergence from the linear law near 500°, which may 

 not be characteristic of pure platinum, and is to be further examined. 



Some of the silicate glasses show irregularities in the increase of 

 the specific heat with temperature. 



Melting and inversion are accompanied by specific heat variations 

 which promise to shed light upon some recent theories as to these 

 changes. 



