10 PROCEEDINGS OF THE AMERICAN ACADEMY. 



containing vessel and for the auxiliary liquid. These corrections have 

 been discussed in great detail in the preceding paper on water. The 

 corrections to be used in this paper are exactly similar to the former 

 ones, with the exception of the correction for the effect of the mercury. 

 It will be recalled that two sets of measurements are necessary to 

 obtain all the data needed in making the corrections; the first is with 

 the steel Inilb filled with the liquid under investigation, and the second 

 with the liquid and bulb replaced by a cylinder of Bessemer steel. 

 The quantities of kerosene are approximately the same in the two 

 sets of measurements. If the quantity of mercury used in the two 

 sets were exactly the same, it would not be necessary to apply any 

 correction for it, but since it was not easy to use exactly the same 

 quantity, a correction for the difference had to be applied. This cor- 

 rection was exceedingly small, being merely the change with pressure 

 and temperature of the small difference, and could be obtained with 

 sufficient accuracy for this purpose from measurements already made 

 on mercury up to 12000 kgm, at 0° and 20°.* The variation of 

 compressibility and dilatation with temperature is so small at 

 high pressures that it is perfectly safe to extrapolate to 80°. The 

 quantities of mercury in the two determinations seldom differed by 

 as much as 0.3 gm., in which case the correction can be entirely 

 neglected. 



The altered design of the containing vessel made necessary a modi- 

 fication of the procedure in filling the cylinder ready for a run. The 

 bulb was filled by boiling the liquid into it at reduced pressure at 

 room temperature. During the filling the liquid came into contact 

 with a gum stopper for a few seconds. The bulb was then discon- 

 nected from the air pump, and gently warmed by holding it in the 

 hands, thus forcing a slight quantity of liquid out of the stem. After 

 the bulb had come to the temperature of the hand, the outside was 

 carefully wiped dry, and the stem immediately inverted into the cup 

 of mercury. In this way the complete exclusion of all air bubbles 

 was ensured. After the bulb had once more come to the temperature 

 of the room, the combined weight of the filled bulb, mercury, and steel 

 cup was found. The weight of the liquid in the bulb can then be 

 found immediately by subtracting the known weight of the mercury, 

 cup, and bulb. The quantity of the liquid used in the majority of 

 the experiments was about 12 c. c. The mercury cup with the bulb 

 still in place was now slipped over the manganin resistance coil, 



4 Bridgman, These Proceedings, 47, 319-438 (1911). 



