SMITH. — EXPANSION OF ETHER AND ALCOHOL. 



429 



K. 



The stopcock K4 was for convenience in case the mercury had for any 

 reason to be removed from the pressure gauge. 



Filling the Dilatometer and 

 the Piezometer. 



The greater part of the capil- 

 lary stem and the short vertical 

 tube of the dilatometer up to the 

 line E F (Figure 4) were filled 

 with a known amount of clean, 

 dry mercury. The mercury was 

 poured into the dilatometer, care 

 being taken to avoid, as far as 

 possible, any air bubbles. One, 

 however, could never be perfectly 

 sure that slight traces of air were 

 not left either in the mercury or 

 along the sides of the tube which 

 contained the mercury. Since 

 the coefficient of expansion of air 

 is only about twice the coefficient 

 of expansion of either ethyl ether 

 or of ethyl alcohol, these very 

 slight traces of air would intro- 

 duce no appreciable error in the 

 data on the expansion of these 

 liquids. The end of the capillary 

 tube being closed, I (Figure 4) 

 was joined by means of a short 

 piece of rubber tubing to a glass 

 tube bent so as to dip into a 

 flask containing the liquid to be 

 investigated. This liquid had 

 been previously boiled to remove 

 traces of air. By alternately 



heating the dilatometer above the boiling point of the liquid and then 

 cooling it, the bulb of the dilatometer was finally filled with liquid 

 which was free from air. When the liquid had assumed a temperature 

 nearly equal to the temperature at which observations were to be made, 

 the stopcock K was closed. 



It was desirable to have the mercury in the dilatometer for two 

 reasons. At low pressures the otherwise free surface of the ether or 



H 



FlGUSE 7. 



