58 PROCEEDINGS OF THE AMERICAN ACADEMY. 



180°. These values all lie roughly on a curve passing through 20.0 

 at 0°, 23.5 at 40°, 27.0 at 80°, 3L0 at 120°, and 3G.5 at 160°. The 

 increase of Cp with temperature is fairly rapid, and becomes more 

 rapid at the higher temperatures. 



Acetone. — Two fillings of the apparatus with this substance 

 were made. The high pressure measurements only were made with 

 the first filling, and both high and low pressure readings with the 

 second. Acetone was unique among the liquids used in that it froze 

 under pressure. This was not anticipated nor desired, since, for one 

 thing, it made impossible measurements at the lower temperatures 

 and higher pressures. Furthermore, the separation of the solid phase 

 is apparently accompanied or foreshadowed by complications in the 

 behavior of the liquid, which it was not desired to encounter at the 

 present stage. As a consequence of the freezing, the readings at 20° 

 run only to 8000 kgm. The curves showing the average properties of 

 acetone over the entire temperature range all show a break, therefore, 

 at 8000 kgm. Below 8000 kgm. the average is over the range from 

 20° to 80°^ but above 8000 the range is from 40° to 80°. 



No attempt was made to follow out the freezing curve, or to deter- 

 mine accurately the equilibrium pressure at any temperatiu-e. It 

 was found, however, that at 40° the freezing pressure is about 10000 

 kgm. The freezing point of acetone at atmospheric pressure is given 

 by Kaye and Laby at — 95°. This raising of the freezing point by 135° 

 seems to be larger than any previously recorded. 



Acetone also showed one other peculiarity. When the liquid was 

 examined at the close of the second run, it was foimd to be of a slight 

 rose color, and there was a small amount of a fine w^hite precipitate. 

 The rose color deepened in the course of several days to a dirty brown, 

 and the precipitate appeared to increase slightly in quantity. It was 

 thought at the time that this was a chemical reaction brought about 

 by pressure alone, but subsequent investigation showed that the effect 

 was doubtless due to the presence of a slight impurity of phosphorus 

 trichloride, left from the previous run. Phosphorus trichloride when 

 mixed with acetone and allowed to stand at atmospheric pressure was 

 found to produce very slowly the same discoloration and precipitate as 

 observed after exposure to pressure. The effect of pressure apparently 

 is merely to hasten the reaction. In a subsequent experiment, in 

 which every trace of phosphorus trichloride had been carefully re- 

 moved by prolonged heating, acetone was submitted to the pressures 

 and temperatures of the regular experiment for a day, with abso- 

 lutely no trace of discoloration. Unfortunately, no examination 



