BRIDGMAN. — THERMODYNAMIC PROPERTIES OF LIQUIDS. 39 



pressures. The accuracy of the compressibihty measurements may 

 be estimated from the fact that the mean discrepancy of the piston 

 displacements in the two sets of high pressure readings was 0.0035 

 inch, the maximum displacement being 2.07 inches. For the dilata- 

 tion, the average discrepancy in the piston displacement for a rise 

 of temperature of 20° was 0.0011 inch, the average displacement being 

 0.070 inch. 



In computing the volumes at atmospheric pressure the density at 

 15° was taken to be 0.7960, from Kaye and Laby. The constants of 

 the dilatation formula from Landolt and Bornstein are as follows; 

 a =0.021186, 6=0.05156, c =0.08911 This gives for the density at 

 0°, 0.8100. The quantity of methyl alcohol to which the tables and 

 the diagrams refer weighs, therefore, 0.8100 gm. Since the boiling 

 point of methyl alcohol is 64.7°, the volume listed in Table II for 80° 

 and atmospheric pressure is, therefore, merely an extrapolation by 

 means of the formula. 



The "reduction factor" by means of which the transition was made 

 from the mathematical formula for volume in terms of pressure at 

 40° to the experimental curve was 1.009. 



The change of volume from 1 to 500 kgm. at 40° was taken as 

 0.0483, following Amagat. It should be noticed, however, that 

 Amagat gives for the volume at 40° and atmospheric pressure 1.0438, 

 against 1.0483 of the tables of Landolt and Bornstein. In this work 

 the value 1.0483 was taken as the volume at 40°, but Amagat 's value 

 for the change of volume 1-500 kgm. was adopted without correction. 

 At low pressures (20°) the present experimental values for the changes 

 of volume were as follows: 1-500 atmos., 0.0530; 500-1000, 0.0294; 

 1000-1500, 0.0242; 1500-2000, 0.0199. The corresponding values of 

 Amagat are 0.0480, 0.0300, 0.0239, 0.0194. The agreement is fairly 

 good, ej^cept for the lowest pressure interval, where as has been pointed 

 out, the present method can only indicate the probable result by an 

 extrapolation. The newly published result of Richards is 0.0430 

 for the change of volume at 20° for an increase of 500 kgm. of pressure 

 as against 0.0415 listed in the tables of volume. 



The volume of methyl alcohol is shown as a function of pressure 

 and temperature in Table II and in Figure 8. 



The compressibility, (8, of the first five alcohols has been measured 

 by Pagliani and Palazzo,^ who have collected their results into formulas 

 of the type, /3< = jS^ (1 + a^ + bf)- Their pressure range was 1-4 



6 Pierre, Ann. chim. et.phys., 15, 325 (1845). 



7 Pagliani and Palazzo, Mem. R. Ace. Lin., 19, 279 (1883/84). 



