284 BURGESS AND SALE: PURITY OF PLATINUM WARE 



Quite satisfactory results may also be obtained by simply 

 touching the Pt wires to the crucibles at e and / without any 

 soldering. With the apparatus once set up, a test may then 

 be made in a few seconds and the crucible remains, of course, 

 absolutely intact. The homogeneity of the crucible may also 

 be determined by this method. 



By means of a commutator at a b, measurements may be 

 made alternately of the temperature at e and of the EMF de- 

 veloped across the crucible when e is at this temperature. The 

 commutator may also be so designed as to reverse e and / if it 

 is desired to make/ the hot junction. 



It is of course essential that the two platinum wires be made 

 of strictly pure platinum. For this purpose, use is made of 

 Heraeus normal thermoelement wire drawn down; this platinum 

 has shown itseK to be a standard, uniform product, which 

 is easily controlled by the electric resistance method above 

 mentioned. 



The wires are attached at e and / by the well known arc- 

 soldering method, which consists in making the crucible one 

 terminal of an electric circuit of about 40 volts and a sharp- 

 ened graphite pencil the other, with a rheostat in series. The 

 operation of soldering consists in drawing a minute arc be- 

 tween pencil and crucible and at the same instant touching 

 the wire to the crucible at this point. The end of the wire is thus 

 fused to the crucibles. With a little practice this operation 

 may be made so that there is hardly any noticeable effect on the 

 appearance of the crucible after the wire is removed. 



In the table are given the results of a series of measurements 

 at about 1050° of the EMF of pure platinum against "plati- 

 num" crucibles of various makes. In figure 2 are shown the 

 isothermal curves, at 900°, 1000° and 1100°C, for iridium con- 

 tent of platinum in terms of EMF against pure platinum. There 

 is also shown the iridium equivalent of the impurities in the 

 crucibles c, h and k of the table, in terms of their stated iridium 

 content. 



It will be noted that all impurities are, for convenience, ex- 

 pressed in terms of iridium content. Of particular interest is 



