﻿THE 

 LONDON, EDINBURGH, and DUBLIN 



PHILOSOPHICAL MAGAZINE 



AND 



JOURNAL OF SCIENCE. 



[FOURTH SERIES.] 



MARCH 1871. 



XXIV. Calorimetric Researches. By R. Bunsen*. 



[With a Plate.] 

 1 . The Ice- Calorimeter. 



THE calorimetric methods hitherto adopted are objectionable 

 on account of the relatively large quantities both of the 

 liquid needed for the calorimeter and of the substance examined, 

 which have to be employed for the purpose of diminishing the 

 error caused by heat unavoidably lost in the experiment. Satis- 

 factory estimations of specific heat can scarcely be obtained by 

 the more exact methods hitherto in use, unless the weigbt of 

 substance employed is at least from 10 to 40 grms. Rare 

 substances in a state of perfect purity can only be obtained in 

 such quantities after almost insurmountable difficulties ; and it is, 

 only on this account, conceivable that we do not yet know the 

 specific heats of all elements as yet obtained in the pure state, 

 although such estimations are of fundamental importance in con- 

 trolling the atomic weights. 



The instrument about to be described will serve to remedy 

 these objections. The principle on which it acts consists in 

 measuring the volume of ice melted by the contraction which 

 this ice undergoes on liquefaction. 



The instrument (Plate V. fig. 1) consists of an inner glass 

 vessel a, in the form of a common test-tube, which has been 

 fused at the blowpipe into the cylindrical vessel b. From this 

 vessel (b) proceeds the glass tube c } to which the iron collar (d) 

 is fastened above. The inner vessel (a) is filled from a to //,, and 



* Translated by Francis Jones, Esq., from Poggendorff's Annalen, No. 9, 

 1870. 



Phil. Mag. S. 4. Vol. 41. No. 272. March 1871. M 



