OF CUBICAL EXFxVNSION OF ICE. 
405 
of tlie experiment, this tube reached as far as o only. The coefiicient of expansion of 
the glass was first determined by preliminary experiments with mercury. To 
partially fill the inner cylinder with water, a bulb provided with a small opening at 
the top was sealed on at c.. This bulb was filled with distilled watei', and after this 
had been boiled for some time the upper orifice of the bulb was sealed. During this 
operation the opening at o had been closed, but, after cooling, l)oth the upper sealed 
point in the ljulb and o were simultaneously opened. Water flowed into M, 
expressing an equal volume of mercury. The trd^e at o was again closed, and the 
Avater in the fine capillary at c was displaced by slightly warming the apparatus. 
This caused the mercury in Avhich the central capillary dipped to rise, aiid the bulb 
was then removed by sealing off at c. Finally the capillary tid^e fd o was sealed 
on at 0 , and the position of the end of the column of mercury marked on tlie tid)e 
after the Avhole apparatus had been reduced to 0° G. On fieezing, the water in 
the inner glass vessel expanded, and breaking the inner cylinder, relieved itself from 
constraint. 
T1 le mean residt foi' the increment in volume of unit of volume of Avater at 0° Cf, 
on changing to ice at 0°, AA'as ’09195, Avhich is equivalent to ’91507 for the density 
at 0°, Avhile ’0001585 Avas olfiained for the coefiicient of cidfical expansion. 
The next observations Avith Avhich it is necessary to deal are those of Dufour 
(‘ Comptes Rendus,’ vol. 54, p. 1080). Having previously experimented by finding the 
density of a mixture of alcohol and Avater in Avhich ice floated in neutral equilibiium, 
he published in 1802 an account of experiments in Avhich a mixture of chloroform 
and petroleum Avas used in preference to the former liquid, which dissoDes ice. By 
taking tlie mean of 16 experiments, he olfiained ’9178, Avdth a probable error of ’0005, 
as the specific graxhty referred to Avater at 0'^ C. He employed the Amine ’000158 for 
the coefiicient of expansion for reducing his results. The ice used Avas prepared 
from Avater boiled in vacuo, and although free from air bidAbles Avas “ opaline ” in 
appearance. 
Bunsen’s celebrated paper on Calorimetry appeared in 1870 
(‘Pogg. Ann.,’ vol. 141, p. 1, 1870). Amongst other researches 
included in this memoir is a determination of the density of ice at 
0° C., by a dilatometric method which, according to the illustrious 
author, completely eliminated the errors Avhich had rendered previous 
estimations uncertain. 
Bunsen’s dilatometer is shown in fig. 2. It consisted of a thick- 
Avalled U-tube of hard glass draAvn out at a, and this A-vas filled Avith 
mercury up to the leA-ml, h, Avhich Avas boiled for some time. Boiled 
Avater Avas sucked into the apparatus, and rested on the mercury 
at b. This water was then boiled in the tube for half an hour, the 
end a being, by means of a rubber tube c, led under the surface of 
Avater, Avhich Avas also kept boiling. The dilatometer Avas alloAved to 
3 o 
Fig. 2. 
9 
VOL. CXCVIII.—A. 
