1891.] Properties of Metals in relation to the Periodic Law. 351 



the amount of heat evolved by the mass after it had become solid, 

 and deducting this amount from the total heat transferred to the 

 water by the melted platinum, he obtained the latent heat of fusion 

 of the metal. 



He also determined* the specific heat and melting point (1045 C.) 

 in the case of gold, but he does not appear to have ascertained what 

 is the latent heat of fusion of the precious metal. The following 

 experiments were therefore made, in order to afford a basis for cal- 

 culating the theoretical lowering of the freezing point of gold which 

 a given addition of impurity should produce. 



A calorimeter of polished silver, 10*5 cm. diameter, and 15'5 cm, 

 high, was supported upon three points of cork within a bright 

 metallic vessel, blackened externally, and constituting an air-jacket. 

 The amount of water employed varied from 800 to 1088 grams. 

 The stirrer was a thin sheet of mica, mounted upon a silver wire, 

 bent at its lower end into the form of a hoop ; the mica also served 

 to catch the gold poured into the water. In experiments E to I, 

 the stirring was effected by silver vanes, of a form suggested by 

 Professor Riicker, actuated by a small electromotor. Quantities of 

 very pure gold, varying from 68 to 123 grams, were melted in a 

 11 clay crucible and poured into the calorimeter, 

 his portion of the manipulation was performed by my assistant, 

 r. Groves, whose long experience in melting gold enabled him to 

 select the latest moment before solidification at which the gold could 

 be poured. In Experiments A and B and E to I, the temperature of 

 the molten mass was measured, by the aid of the thermo-junction 

 previously described, which was placed directly in the molten gold, 

 up to the moment of pouring, and it is believed that the temperature 

 the mass was known to within 10 C. 



t the end of each experiment, the gold poured into the calori- 

 eter was carefully collected and weighed. The thermometer used 

 was a very sensitive one with fixed zero, and made by Hicks, of 

 Hatton Garden; it could easily be read to 0'02 C. The depth 

 of water employed was sufficient to prevent the evolution of steam, 

 and none was observed to escape in any of the experiments recorded. 

 The calculations are as follows : 



t P = weight of water in calorimeter. 

 p!d = water equivalent of calorimeter. 

 j9 2 c 2 = ,, thermometer. 



p = weight of gold employed. 

 c = average specific heat of gold (Violle). 

 t = initial temperature of the molten gold. 

 T = water. 



ti = final temperature of the water. 

 * ' Comptes Rendus,' vol. 89, 1879, pp. 702-703; 92, 1881, pp. 886-8. 



