BRIDGMAN. — THERMODYNAMIC PROPERTIES OF LIQUIDS. 59 



was made of the condition of the Hquid after the end of the first run, 

 so it cannot be told whether the effect was present then or not; prob- 

 ably not. In any event the error so introduced is probably very 

 small, because not more than a very small impurity of PCI3 could 

 have escaped attention in the weighing. There is, however, a slight 

 possibility that the reaction was catalytic, in which event the error 

 might be greater. The close agreement of the two sets of readings 

 makes this unlikely, howe^'er. 



The mean discrepancy in the piston readings for compressibility 

 was about 0.005 inch on a total stroke of 2.0 inches. The first com- 

 pressibility readings were made at 40° The liquid does not freeze 

 until it has been considerably subcooled, so that it was possible to 

 cover the entire pressure range at 40°. But in order to avoid the 

 possibility of the liquid freezing the second time at a less degree of 

 subcooling than at first, the second run was made at 60°, and then 

 reduced to 40° for comparison with the first run. The average 

 discrepancy^ of the displacements for thermal dilatation was 0.0013 

 inch on an average of 0.070 inch. 



The reduction factor from the mathematical formula was 1.049, 

 showing that acetone is more compressible than the average. 



The boiling point of acetone is 56.5°, so that for this reason the 

 initial point of the 80° curve is taken as 1000 kgm. The initial values 

 at 60° were obtained by extrapolation, disregarding the boiling, and 

 strictly apply only to a zero of a few kgm. 



The density of acetone at atmospheric pressure and 0° was assumed 

 to be 0.8136. The constants of the dilatation formula were: a = 

 O.O2I324, h = O.O538O, and c = -O.OgSS. These constants are taken 

 from data of Zahnder, who gives for the density at 0° 0.8125, and for 

 the boiling point 56.3°, values slightly different from those given 

 above. 



The change of volume at 40° between 1 and 500 kgm. was taken 

 as 0.0541 from Amagat. His value for the volume at 40° and 1 kgm., 

 however, is 1.0575 against 1.0585 given by the formula above. The 

 probable accuracy of the low pressure measurements of acetone may 

 be judged from a comparison of the values at 20° with those of Amagat. 

 For the successive pressure intervals 1-500, 500-1000, 1000-1500, and 

 1500-2000 atmos. the present work gave the following changes of 

 volume; 0.0526, 0.0308, 0.0251, 0.0208, while Amagat gives 0.0483, 

 0.0325, 0.0245, and 0.0196. 



The volume of acetone as a function of temperature and pressure is 

 shown in Table X and in Figure 14. 



