On the Coefficient of Expansion of Gas- Solutions. 113 



images, such comparison is not quite trustworthy. On each 

 of the four days, however, when I made observations in the 

 different homogeneous colours, I believe that I noticed a 

 greater darkness with the blue. 



XVI. On the Coefficient of Expansion of Gas-Solutions. By 

 E. L. Nichols, Ph.D. '( Gottingen), and A. W. Wheeler, 

 Fellows of the Johns Hopkins University, Baltimore*. 

 [Plate II.] 



IN the determination of the coefficient of expansion of 

 aqueous gas-solutions the methods used for ordinary 

 liquids by Pierre, Kopp, Matthiessen, and others are open to 

 serious objection. For the research about to be described 

 here, which is confined to the study of the expansion of aqueous 

 solutions of ammonia, the writers devised an apparatus better 

 adapted to their purpose. 



This apparatus consists of a bath, B (fig. 1), in which is 

 immersed the glass bulb of the dilatometer, D. The straight 

 portion, N, of the neck of the dilatometer dips at its lower 

 end into a larger glass tube, T, filled with mercury. $ is a 

 mirror-scale graduated to millimetres. The neck N can be 

 fastened securely to the scale at k. The mercury-tube is ad- 

 justable to different heights. The thermometer, t, shows the 

 temperature of the bath. H is the stirrer. 



With this apparatus there is no evaporation, nor escape of 

 gas, the solution being kept entirely from the air. Aside 

 from this advantage over the open-mouthed dilatometers is 

 that of very great sensitiveness, due to the great size of the 

 bulb compared with the capacity of the neck. The convenience 

 in filling is also of importance when one is experimenting 

 with volatile liquids. 



Our method of volume-measurement was as follows : — The 

 dilatometer was filled, then placed in the bath in the position 

 shown in fig. 1 (the bulb being wholly below the surface of the 

 liquid), and was securely fastened. The adjustable tube t was 

 then raised until the end of the neck of the dilatometer dipped 

 into the mercury. The filling was performed at a temperature 

 higher than that of the bath, so that the mercury might rise 

 somewhat in the neck. The bath was then cooled by inter- 

 vals of a few degrees each; and the height of the mercury in 

 the neck, after each cooling, was noted. From these readings 

 and the corresponding temperatures our results were calcu- 

 lated. To obtain from these readings the relative volumes of 

 the solution for different temperatures the following data were 

 necessary : — 



* Communicated by the Authors, having been read before the American 

 Association for the Advancement of Science, Boston, August 28, 1880. 



Phil Mag. S. 5. Vol. 11. No. m. Feb. 1881. 1 



