on the Electrical Resistance of Bismuth. 199 



Table (continued). 



(e) Compressed Bismuth. (From Prof. Spring.) 





Temp. 



W. 



Wm. 



Compressed bismuth ..A 



The same compressed bis- \ 

 muth, melted, cooled 

 very slowly in a > 

 sand-bath, and then 

 operated upon again. J 



16-8 

 424 

 76 



f 16-2 

 ] 42-3 

 [ 71-7 



0-1148 

 0-1140 



01150 



0-1861 

 0-2051 

 0-2263 



01157 

 0-1145 

 0-1153 



01871 

 02053 



02263 



VI. Inferences drawn from the preceding Tables. 



1. The Influence of Temperature. — If the preceding results 

 be examined, it is easy to see at once that the different 

 samples of bismuth examined behave quite differently. Some 

 of them give an increased resistance, and others, which 

 is remarkable, a diminished resistance, when the temperature 

 rises. This last fact, noticed for the first time by M. Righi, 

 is not due, according to this physicist, to the presence of 

 arsenic as an impurity in the bismuth, but ought to be 

 attributed to the tin, which, even in very small quantities, is 

 able to produce this result. 



A number of experiments have been made by us to dis- 

 cover the reason of this diminution of resistance. We studied 

 first alloys of bismuth and tin, in which the latter metal was 

 present in comparatively large quantities. On examining 

 Table (c), it will be seen that these alloys always give an 

 increase of resistance with the temperature, while the bismuth 

 which enters into their composition produces a diminution of 

 resistance for a rise of temperature. Besides, the chemical 

 analysis has shown us that not one of oar samples of bismuth 

 contains any tin as an impurity. It is then sufficiently 

 demonstrated that it is not the metal that is the cause of the 

 observed anomaly*. A glance at the results of the chemical 

 analyses and at the electrical measurements will equally prove 

 that the reason for the fact ought not to be looked for in the 

 presence, as impurities, either of arsenic or of iron. 



Mr. 0. W. Kayserj has shown that expansion on solidifi- 



* See also the memoir of Mr. W. Nemst (Annalen der Physik, 1887, 

 p. 783), which has appeared during the course of our researches. 

 t Beriehte der deutsch. chem. Geselhch. Berlin, 1869, p. 309. 



P2 



