VEGA- EXFEDiriOXENS VETENSKAPLIGA AHCETEN. 269 



therefore be mentioned ver}^ briefly here. A glass tube of the 

 shape shown in fig. 1 on plate 20 is filled with water and 

 placed in an iron calorimeter containing a weighed quantitv of 

 mercury. The calorimeter is protected from the condensing 

 humidity of the air by means of an enclosing cylinder of metal. 

 The air enclosed between the calorimeter and the envelope is 

 kept dry by means of conc. snlphuric acid. The calorimeter 

 rests upon three points of ivory and is nowhere in contact 

 with the metallic capsule. The capsnle being cooled from 

 without by means of a freezing mixture of snow and NaCl, 

 the temperature of the calorimeter slowly sinks by radiation to 

 — 6° or — 8° C. During the whole time (the experiment ordi- 

 narily lasts 6 or 7 hours) the mercury is constantly stirred 

 and the temperature observed every half minute, as soon 

 as it has approached Vli or 2 degrees from the appointed 

 temperature of solidification, which the experimentor is at 

 liberty to fix beforehand, according to his pleasure. When 

 the temperature has sunk sufficiently, an assistant introduces 

 a thin thread of glass, which has been brought in contact 

 with a snow flake, into the stem of the tube, the water of 

 which is then immediately transformed into ice. The solidi- 

 fication is at — 6° or — 8° C almost instantaneous: a sudden 

 crack is heard, the tube is broken into pieces, which are kept 

 below the surface of the mercury in the calorimeter by a 

 network of iron. The latent heat developed by the freezing 

 of the water is measured by the rise of temperature observed 

 by means of a kathetometer on the Geissler thermometer Gjp 

 On account of the excellent conductivity of mercury relatively 

 to the fluids ordinarily employed in calorimetric experiments, 

 the eqiiilibrium of temperature is established in a very short 

 time and the curves representing the loss of heat b}" radiation 

 can be constructed with the utmost regularit3\ 



It will occur to the reader, that this method gives the 

 latent heat of the fluid enclosed in the spiral by a tempera- 

 ture some degrees below its ordinary melting point. Accor- 

 ding to the law of Person the latent heat will vary a little 

 with the tem^perature of freezing, variations, which may be 

 calculated from the difference of the specific heat in the liquid 

 and solid state. Let r denote the latent heat at the ordinarv 



1 In tbe Journal f. prakt. Cliem., Avliere an illustration of the progress 

 is insorted, the author has published his determinations of the latent heat 

 •of phosphorus at diflferent temjieratures below its real melting point. 



