[ 303 ] 



XXV. The Variation in the Electrical Resistance of Bismuth, 

 ivhen cooled to the Temperature of Solid Air. By James 

 Dewar, M.A., LL.D. y F.R.S., Professor of Chemistry in the 

 Royal Institution, fyc, and J. A. Fleming, M.A., D.Sc, 

 P.R.S., Professor of Electrical Engineering in University 

 College, London *. 



[Plate V.] 



IN the course of investigations on the Electrical Resistance 

 of pure metals and alloys at low temperatures made 

 during the years 1892 and 1893, the results of which were 

 communicated to the Philosophical Magazine f, extensive 

 examination was made by us of the variation in specific 

 electrical resistance exhibited by most of the ordinary metals 

 when cooled to temperatures approaching the absolute zero. 

 In the course of this research the abnormal behaviour of 

 bismuth with regard to its variation of specific resistance with 

 temperature was noted, and in a further research on the 

 Thermo-electric powers of metals and alloys at low tempera- 

 tures, the results of which were communicated to the Philo- 

 sophical Magazine recently {, it was noticed that the curves 

 representing the thermo-electric power of bismuth, both for 

 pure and for commercial specimens, presented a discontinuity 

 at temperatures lying between about —70° and —100°. We 

 were therefore recently led to make a more careful examination 

 of the variation of electrical resistance of bismuth when 

 cooled down to the temperature of solid air, some of the 

 results of which are of sufficient interest to communicate, 

 although the investigation is still in progress, and much 

 remains to be done to complete it. A sample of very pure 

 bismuth was kindly prepared for us by Mr. George Matthey, 

 by the reduction of the oxide, an analysis of which showed 

 very small traces of iron as the only impurity. This bismuth, 

 which was the same as that used bv us in our thermo-electric 

 experiments, was pressed into wire by a steel press, two sizes 

 of wire being prepared, one having a diameter of about 

 0*067 centimetre, and the other a diameter of 0*049 centi- 

 metre. Lengths of these wires were mounted up in an appro- 

 priate form for taking the electrical resistance when measured 



* Communicated by the Authors. 



t Dewar and Fleming, Phil. Mag. 1892, 5th series, vol. xxxiv. p. 326, 

 "On the Electrical Resistance of Pure Metals, Alloys, and Non-Metals at 

 the Boiling-point of Oxygen ; " also Phil. Mag. 1893, 5th series, vol. xxxvi. 

 p. 271, " The Electrical Resistance of Metals and Alloys at Temperatures 

 approaching Absolute Zero." 



% Dewar and Fleming, Phil. Mag. July 1895. " Thermo-Electric 

 Powers of Metals and Alloys between the Temperatures of the Boiling- 

 Point of Water and the Boiling-point of Liquid Air." 



