SMITH. — EXPANSION OF ETHER AND ALCOHOL. . 431 



the change of volume corresponding to a definite change of pressure 

 or of temperature one need only observe the distance which the 

 liquid meniscus in the capillary tube moved. For this purpose a 

 telescope mounted with its axis horizontal on the screw of a dividing- 

 engine was used. The position of the telescope of the dividing- 

 engine is indicated in Figure 3. The screw of the dividing engine 

 had a pitch of 0.05 cm. and was graduated to read to 0.000125 cm. 



Measurement of Teminrature. 



It has been already pointed out that both the piezometer and the 

 dilatometer contained a platinum thermometer for the measurement 

 of the temperature of the liquids. Both of these thermometers were 

 taken from the same piece of fine platinum wire, which was about 

 0.015 cm. in diameter. This wire was in each case annealed by pass- 

 ing through it a current of 2 amperes for about 6 seconds. The wire 

 was then wound in the form of a helix, 30 cm. long and 1.3 cm. in 

 diameter. To each end of the helix was hard-soldered a short piece 

 of rather stout platinum wire. This was found necessary in order to 

 prevent the wire from breaking off where it was sealed into the ends 

 of the glass bulbs, G and H (Figure 4) and and N (Figure 5). 

 Each of the platinum spirals, when inside the bulb, extended nearly 

 the whole length of the bulb, and the axis of the spiral coincided nearly 

 with the axis of the bulb. To the terminals of the spirals were 

 soldered copper lead wires, which, after being carried along parallel 

 to the capillary tubes of the dilatometer and the piezometer respec- 

 tively, passed out through the respective rubber stoppers M (Figure 4) 

 and L (Figure 5). It was found necessary to enclose these lead 

 wires in fine rubber tubing, in order to prevent battery-action between 

 them and the liquid in which they were subsequently immersed. The 

 total resistance of the lead wires on the dilatometer was 0.114 ohm, 

 while those on the piezometer had a resistance of 0.123 ohm. 



Since both platinum thermometers were taken from the same piece 

 of wire and had been treated in precisely the same way, it seemed 

 sufficient to calibrate only one of them. I calibrated the platinum 

 thermometer which was in the piezometer and used the temperature 

 coefficient thus obtained in the work with the one which was in the 

 dilatometer. In the calibration of the platinum thermometer the 

 bulb of the piezometer was immersed in a mixture of ice and water 

 or in the vapor of one of the following liquids, boiling under atmos- 

 pheric pressure, — water, benzol, chloroform, and carbon disulphide. 

 The temperatures, except those of melting ice and boiling water, were 

 determined by means of a well-tested Baudin thermometer. 



