432 Wright and Larsen — Quartz as a Geologic Thermometer. 



peratures are of interest for the sake of comparison and are 

 included under column I of Table 4. In the second column 

 the average result of the measurements of the present writers, 

 for the different temperatures is given, and a comparison of 

 the two columns shows agreement as close as could be expected 

 and well within the limits of error. The measurement of the 

 angle of rotation of quartz is less accurate at all temperatures 

 than that of the birefringence, and such close agreement between 

 individual readings cannot be expected, especially if plates of 

 different thickness are used, as in the present series of 

 determinations. 









Table 



IV. 







Aug] 



e of rotation 



Angle of rotation 





Temperature 



(Joubert 



and LeChatelier 



) (Wright and Larsen) 



Difference 







I. 





II. 





20° 





21-72° 





21-71° 



o-oi° 



100 J* 





21-98 





21-96 



0-02 



280 





22-68 





22-58 



o-io 



360 J 





23-04 





22-90 



0-14 



415 





23-40 





23-15 



0-25 



448 J 





23-46 





23-30 



0-16 " 



560 





24-30 





24-10 



0-20 



600 





25-26 





25-20 



0-06 



840 J 





25-26 













900 





.25-32 









1500 J 





25-42 









Table 4. In this table the measurements by Joubert and LeChatelier of 

 the angle of rotation of basal plates of quartz l mm thick at different 

 temperatures in sodium light are compared with those of Wright and Larsen 

 for the same temperatures. 



The actual point of change is still further emphasized by the 

 behavior of the basal crystal plates themselves at the inversion 

 temperature. In passing through 575° the optic phenomena 

 become temporarily disturbed and there is no position of total 

 extinction. As soon as this temperature is passed, however, 

 the abnormal behavior subsides and accurate measurements 

 can again be made. This effect, which is most pronounced 

 on thick plates, may be due in part at least to unequal distribu- 

 tion of the heat and consequent lag in inversion of one part of 

 the plate after another, while the optical effect observed is 

 that from the whole plate. — To test this behavior still further, 

 a plate parallel to the principal axis was taken and observed in 

 parallel position between crossed nicols. In this instance, 

 however, no lighting up of the field was observed at the inver- 

 sion temperature, thus indicating that the change which takes 

 place does not affect the vertical c-axis appreciably, it remain- 

 ing strictly parallel throughout the inversion. 



* Measurement by Joubert, Comptes Eendus, lxxxvii, 497-499, 1878. 



