﻿Expansion of a Drawn Tube of Vitreous Silica. 000 



results depends to a slight extent on the value assumed for 

 the expansion of: the 2 cm. tripod, which has not yet been 



completely determined. It is possible, however, to state that 

 the axial coefficient exceeds the radial by '20 X 10 ~ 6 over the 

 range 0° to 100° G. Assuming the value *465 x 10" 6 (the 

 mean of the results above quoted for the linear coefficient) 

 as being the value of the axial coefficient of the tripod, it is 

 clear that the cubical coefficient would come out *995 X 10~ 6 , 

 which agrees as closely as could be expected with the result 

 deduced by Harlow and Eumorl'opoulos independently 

 assuming the values of Callendar and Moss for the absolute 

 expansion of mercury. 



It may be objected that, although the axial and radial 

 coefficients differ, the mean of the three must agree with the 

 value for an isotropic specimen. Taking Chappuis' value 

 •oOxlO" 6 as representing an isotropic specimen, the axial 

 coefficient of the tripod would have to be *63xl0~ 6 . In 

 view of this objection, the axial coefficient of the ring, which 

 is probably the same as that of the tripod^ was measured on 

 three days, and found to be *455 x 10~ G ; but no great stress 

 is laid on this result (further than as confirming the value 

 above assumed) because the ring is only &5 mm. wide, and 

 was not intended for this determination. 



The principal objection advanced by Eumorfopoulos against 

 the results of Callendar and Moss at low temperatures, where 

 the observations are admittedly more difficult, appears to be 

 that, according to his weight thermometer, the cubical ex- 

 pansion of silica would vanish between 0° and 15° C This 

 apparently impossible result is confirmed by t lie observations 



