406 Scientific Proceedings, Royal Dublin Society. 



exhausted tube until it was nearly filled. After cooling, the vapour- 

 density tube was placed in a vertical position, and a little more mercury 

 added so as to completely fill it ; it was then inverted in a large trough of 

 mercury. The liquid to be investigated was weighed in a small bulb provided 

 with a hooked capillary neck ; and the bulb was passed up into the tube 

 immediately after being weighed. 



Mercury was then poured into A until it filled the glass cylinder and 

 also the large perforatiou in the india-rubber cork. The vapour-density 

 tube was then sloped so as to cause more mercury to euter it from the 

 trough. The open end was then closed by the finger and the tube was 

 cautiously removed, and the open end placed in the mercury in the wide 

 upper part of the perforation in the cork, wliich served as a small temporaiy 

 trough. The vapour-density tube was then pushed down through the cork as 

 far as possible, the excess of mercury escaping through the tube D. By forcing 

 air through the tube 0, more mercury was driven out of A through D, 

 until the requisite level was reached. The stopcock on D was then closed, 

 and the rest of the apparatus fitted up. 



The pressure of the vapour requires to be corrected for the heated column 

 of mercury, and also, at the higher temperatures, for the vapour-pressure of 

 mercury. 



2. For pressures higher than one atmosphere, tlie pressure-apparatus 

 (fig. y) was employed. As with the modified Hofmann apparatus, pressure, 

 volume, and temperature can all be altered at will. 



The volumes, as in all other cases, were corrected for the expansion of 

 the glass tube ; the corrections for pressure differed from tliose required for 

 determinations of vapour-pressure in two respects : — 



{a) There was no column of liquid over the mercury. 



(6) The vapour-pressure of mercury must be taken into account ; but 

 the estimation of this correction is difficult owing to the slowness with 

 which mercury vapour diffuses through a compressed gas. The question 

 is fully discussed in the paper on isopentane.' 



Direct Method. 

 1. By means of the Pressure- Apparatus. — It has been shown (p. 398) how 

 (a) the ratio R of the specific volume of saturated vapour Vv to that of 

 liquid V;, and (i) the specific volume of liquid, were calculated from four 

 readings of the actual volumes of liquid and saturated vapour at each 

 temperature, the weight of substance in the tube being known. 



1 Proc. Phys. Soc, xiii., p. 621, 1895 ; Zeitschr. physik. Uhem. xxix., p. 210, 1899. 



