420 Ferguson and Merwin — Melting Points of 



unsuitable for such long heats; the ordinary platinum 

 furnace is not suitable, therefore one of different design 

 built on the well-known cascade principle was evolved and 

 did duty throughout the entire investigation. 



A vertical cross-section of this furnace is shown in 

 fig. 1. The outer heater, of platinum wire 0-8 mm. in 

 diameter, is wound on a helically grooved alundum tube 

 and held in place with alundum cement ; the outer insula- 

 tion consists of magnesia powder surrounded by a heavy 

 fire clay shell. The inner heater, of Pt-20% Rh wire 

 0-5 mm. in diameter, is wound on a helically grooved 

 magnesia tube and held in place by tie-wires of the same 

 alloy. The two heaters are insulated from each other 

 with magnesia powder and the furnace is so constructed 

 as to facilitate the addition of more magnesia powder at 

 any time. 7 



The quenching apparatus 8 with the Pt:Pt-10% Rh 

 thermoelement and its leakage wire is shown in fig 2. 

 The upper cap is cemented to the Marquardt porcelain 

 tube but all the other parts are snug fitting and held in 

 place by the wires. 



The operation of the furnace was very simple. , Enough 

 current was sent through the outer coil to maintain a 

 temperature of 1450° in the furnace without the use of 

 the inner coil. The furnace was then heated to the 

 required higher temperature by passing the necessary 

 current through the inner coil. The temperatures were 

 measured by means of the usual potentiometer set-up, 

 single shielding being used, and there was no trouble 

 from electrical leakage. 



The determinations were made in the <f ollowing man- 

 ner: a small amount of the sample was wrapped up in 

 platinum foil and tied with a platinum wire to the ther- 

 moelement in such a way that it hung close to the 

 thermoelement junction. The thermoelement tube with 

 the charge was then inserted into the furnace which had 



7 In the earlier runs several such additions were necessary during an 

 afternoon. The magnesia did not pack evenly and between runs the caked 

 material was broken up by means of a special iron drill. This assured a 

 fairly uniform insulating layer and had it not been done there would have 

 been spaces in which there was no insulating material except air between 

 the heaters. 



8 The usual form of quenching apparatus could not be used as experimenta- 

 tion showed that the furnace was too small and the platinum too soft. The 

 charge stuck to the globules of platinum which lined the furnace tube and 

 which resulted from the fusing of the fine wire holding the charge. 



