308 Profs. Dewar and Fleming on the Variation 



lead and tin, was described by M. von Aubel (see Phil. Mag. 

 1888, 5th series, vol. xxv. p. 191). He studied the electrical 

 resistance of certain specimens of pure bismuth in the form 

 of rods slowly cooled, and for temperatures between 17°* 2 C. 

 and 70 o, 8 C. found a negative coefficient for some of these 

 specimens. In the case of rods of pure bismuth which had 

 been tempered between temperatures 15°'2 C. and 69°' 7 C. 

 he found a positive temperature coefficient within these limits ; 

 that is to say, in the first case heating the bismuth decreased 

 its resistance, and in the second case heating the bismuth 

 increased its resistance between the stated limits of tempera- 

 ture. In none of the samples of bismuth examined by us 

 have we found a negative temperature coefficient for tem- 

 peratures above 0° C. 



For a certain specimen of compressed bismuth wire 

 M. von Aubel found the resistance practically constant be- 

 tween 1() 0, 8 C. and 76° C; but on melting, cooling, and 

 annealing the same specimen it acquired a positive tempera- 

 ture coefficient between these limits. M. von Aubel examined 

 the electrical resistance of certain bismuth-lead-tin alloys,, 

 and he came to the conclusion that the cause of these anoma- 

 lies in the temperature coefficient of bismuth was the varying 

 physical condition of the bismuth, and not the presence of 

 impurities. In 1894 Mr. J. B. Henderson published a paper 

 " On the Effects of Magnetic Fields on the Electric Con- 

 ductivity of Bismuth " (see Phil. Mag. lcS94, 5th series, 

 p. 488), and he gives a series of curves showing the tempera- 

 ture variation of bismuth when taken in magnetic fields of 

 different strengths. For a certain specimen of bismuth wire, 

 and for a certain magnetic field-strength of about 10,000 

 C.G.S. units, his curves indicate a point of minimum resist- 

 ance for the bismuth. Taking the bismuth at a certain tem- 

 perature, the temperature coefficient at that point in a zero 

 magnetic field was found to be positive ; but on gradually 

 increasing the strength of the magnetic field in which the 

 bismuth was immersed the temperature coefficient finally 

 became negative. One point of interest, then, in connexion 

 with these investigations is whether the specific resistance 

 of bismuth always presents a minimum value, and whether the 

 temperature at which this minimum value occurs depends 

 upon the strength of the magnetic field in which the 

 bismuth is immersed as well as upon the physical condition 

 of the metal. In the case of the two wires of pure bismuth 

 " A " and " B ' used by us, the diameters of the wires were 

 different, and the pressures under which they were formed, 

 and therefore the physical condition in the interior of the 



