244 N. L. Boioen — The Ternary System 







Table 



v. 





I 



II 





in 



Final 



IV 



Final 





Heating 



disappearance of 



disappearance c 





curve break 



pyroxene determined f orsterite deternii 



#MgSi0 3 



"Upper Point" 



by 



quenching 



by quenching 



78 f 



1424* 





1425 



1474 



73 



1427 





1430 



1484 



75 



1432 





1440 



1490 



80 



1474 





1477 



1515 



90 



1524 





1523 



1552 



95 



1543 





1543 



1558 



100 



1557 





1557 



1577 



*The figures given in column II are the values found in the earlier work 

 corrected in terms of the extended gas-thermometer scale, see Day and Sos- 

 man : The Melting Points of Minerals in the Light of Recent Researches on 

 the Gas Thermometer, this Journal (4), xxxi. 341-349, 1911. 



f The figures in column I are in terms of the components CaSi0 3 and 

 MgSi0 3 ; see lower scale of figure 20. 



A still higher break corresponding to the temperature at 

 which forsterite finally disappears in the liquid was not 

 obtained in the earlier work and it has - remained for quenching 

 to determine the existence and the limits of the temperature 

 interval during which the various mixtures consist of forsterite 

 and liquid (see Table V and figure 20). 



Summary of Comparison of Results. — The comparison of 

 the results of the earlier and the present work may now be 

 summarized. The method of heating-curves was employed in 

 the earlier work and the mixtures were crystallized in such a 

 manner that some of them consisted of pyroxene of variable 

 composition, forsterite and silica, and were not equilibrium mix- 

 tures. On this material two breaks were obtained, near each 

 other, at about 1375° and 1385°. The lower break, formerly 

 attributed to inversion, has been found by quenching to be 

 due to the melting-together of silica and pyroxene and the 

 higher break to the melting-together of forsterite and pyrox- 

 ene. ISTo change occurs at the temperature of the lower 

 break in any mixture when the material started with is homo- 

 geneous pyroxene, as it should be at equilibrium. No change 

 occurs at the temperature of the higher of these two breaks in 

 the more magnesian mixtures when they consist of similar 

 homogeneous material. The temperature recorded as the 

 upper point in the earlier work corresponds with the tempera- 

 ture of final disappearance (solution) of pyroxene, leaving liquid 

 and forsterite. The existence of a temperature interval during 

 which the mixture consists of forsterite and liquid was over- 

 looked in the earlier work but revealed by quenching. 



