204: Richards — Transition Temperature of Sodic Sulphate. 



partly dried in a desiccator. When placed in a wide test-tube 

 surrounded by a Beckmann air jacket (a still wider test-tube) 

 and allowed to remain for an hour in a thermostat at 32*9°, it 

 melted with absolute constancy at a temperature indicated on 

 the arbitrary scale by 4*758. On the following daw upon being 

 replaced in the thermostat the "melting" point appeared to be 

 precisely the same. It was found advantageous to melt the 

 crystals partially into a pasty mass, by means of a gas-lamp, 

 before immersing them in the thermostat, and the mass was 

 stirred gently by a ring-shaped platinum stirrer. Upon raising 

 the temperature of the bath to 36° not the least change was 

 observed, although this application of a higher temperature was 

 allowed to continue for half an hour. A lower temperature, on 

 the other hand, had a less satisfactory effect. When the bath 

 was kept at 29° the thermometer in the sodic sulphate indicated 

 a constant reading of 4*746, a depression of 0*012°. In order 

 to verify this phenomenon, a purer specimen, which melted at 

 4*760° with the bath about 33°, was surrounded by an environ- 

 ment at 30°, and then indicated 4*750° on the arbitrary scale. 

 Thus it is evident that the most satisfactory results are to be 

 obtained by melting the salt slowly, and not by allowing it to 

 solidify. The reason for this is obvious. The anhydrous salt 

 is far less apt to form supersaturated solutions than the 

 hydrated salt, and the crystals instantly melt upon being 

 heated above their transition point, absorbing an appreciable 

 amount of heat in the process. On the other hand, in order 

 that the mass may solidify, the anhydrous salt must dissolve 

 while the supersaturated solution is depositing its hydrate, both 

 of these operations being much less prompt than the preced- 

 ing. In all subsequent determinations the thermostat was 

 kept about half a degree above the point sought, and it 

 was found that the mixture would then melt so slowly as to 

 last for hours. 



Second Problem — Concerning the Purity of the Salt Required. 



Lowenherz* has found, under van't HofFs direction, that an 

 admixture of foreign substances lowers the transition tempera- 

 ture of sodic sulphate just as it does simpler melting points. 

 Hence a study of the purification of the salt is of great import- 

 ance for our present purpose. 



The mother-substance used in preparing the salt above was 

 found to melt at a temperature of 0*06° below that which had 

 been twice recrystallized in porcelain. Four further crystalli- 

 zations in platinum vessels changed the transition temperature 

 only 0*002° above that corresponding to two recrystallizations, 

 the arbitrary scale indicating 4*760°. 



*Loc. cit. 



