18 PEOCEEDINGS OF THE AMERICAN ACADEMY. 



15.9% boiled between 80° and 81° C. 



17.9% boiled between 81° and 87° C. 



31.4% boiled between 87° and 102° C. 



Chief impurity probably phosphorus oxychloride formed by 



combination of the trichloride with the moisture in the air, 



also some pentachloride is almost always unavoidably present, 



due to excess chlorine. 

 Ethyl Bromide {Eivier and Amend). 



The whole sample boiled from 38.0° to 38.4° C. and with the 



exception of not over 2% at 38.4° exactly. Neutral to litmus, 



an excellent sample. 

 Analyses could not be made of the ethyl chloride and the ethyl 

 iodide, because these samples w^ere completely used up in the experi- 

 ment. Both of these were obtained in sealed glass bulbs from Kahl- 

 baum, were used immediately after opening, and were both colorless 

 when used. 



These analyses show that probably none of the liquids contained 

 enough impurity to have any perceptible effect, except the phos- 

 phorus trichloride, and to a less degree the acetone. The phosphorus 

 trichloride was very noticeably simpler in its behavior at high pres- 

 sures than the other liquids; probably due to the fact that it is a 

 mixture of different substances, and so can not be expected to show 

 the well marked behavior of a single pure substance. 



A few measurements were made on one liquid which are not tabu- 

 lated here. These were measurements on chloroform, which was 

 found to freeze at fairly low pressures, about 6800 kgm. at 40°, and 

 10000 kgm. at 80°. Further investigation of this liquid has been 

 postponed until a systematic investigation of freezing curves can 

 be taken up. 



III. Methods of Computation. 



The methods of computation were slightly different from those 

 used in the work on water, in part because of the somewhat different 

 experimental method. They are, in consequence, here described in 

 detail. As in the computations for water, the first step was to pre- 

 pare a table giving the volume at regular intervals of pressure and 

 temperature. From this table the other thermodynamic quantities 

 were then computed. 



The Table of Volumes. In preparing the table of volumes, the first 

 computation was of the volume as a function of pressure at 40°. The 

 first step was to plot the piston displacement against the displacement 



