Properties of Platinoid and Manganine. 303 



tested several specimens of commercial copper wire of different 

 sizes against plates of electrically deposited copper and that the 

 electromotive force of every one of these combinations was 

 extremely small or insensible. When such results as these are 

 compared with those obtained fifteen or twenty years ago,* it 

 is evident that great progress has been made in the manufac- 

 ture of copper wire for electrical uses and that commercial 

 copper has now a pretty definite thermo-electric position. 



The platinoid wire which 1 used was of two sizes (nos. 19 

 and 28 of the American Gauge) and was bought about four 

 years ago of Messrs. Elliott Brothers of London. 



The so-called " manganine " wire was of seven sizes bought 

 at different times through Messrs. Gillis & Gleeson, of Messrs. 

 Bradford, Kyle and Co. of Plymouth, Mass. The largest size 

 (no. 18) was hard drawn, the other sizes were all annealed. 

 Three specimens (nos. 18, 26 and 30) showed when examined 

 qualitatively only a trace of nickel in combination with copper 

 and manganese. 



Each of the wire thermal junctions when in use was im- 

 mersed close to the bulb of a thermometer in a bath containing 

 about a litre and a half of heavy paraffine oil and furnished 

 with a stirring apparatus. Each oil bath was set into and 

 jacketed by a vessel of about 10 litres capacity into which ice 

 or warm water or steam could be introduced or which could 

 be used as an air bath. The electromotive forces of the dif- 

 ferent combinations were determined by compensation on a 

 potentiometer. When, for the purpose of obtaining a rough 

 check on the other results, quicksilver was opposed to plati- 

 noid or to manganine, the lower end of a very large test tube 

 containing about 40 grams of redistilled mercury was deeply 

 immersed in each oil bath and the contents of the two tubes 

 were connected by a long glass siphon filled with mercury. 

 The bulb of a thermometer and an end of a piece of wire of 

 the kind to be examined were immersed in the mercury of each 

 test tube and obvious precautions were taken to prevent undue 

 loss of heat from the contents of the tubes. The following 

 table gives the results of observations made in this way with 

 platinoid wire no. 19 and with hard drawn manganine wire no. 

 18. In each case the current passes across the hot junctiou 

 from the first named metal to the second, the electromotive 

 forces are measured in microvolts and the temperatures of the 

 junctions are given in centigrade degrees. 



* See Everett's Physical Constants, p. 151. 



