228 On the Passage of tlie Galvanic Current through Iron. 



constant can be written in the form 



tW_ 

 w 



it follows that every increase of w by the passage of the cur- 

 rent must itself lessen that constant. Indeed Robinson* has 

 confirmed this by showing that t diminishes with i increasing. 

 But in iron, apart from the beating, a further increase of w 

 takes place particularly through the magnetizing. Hence, 

 with equal values of i, the constant must come out smaller for 

 iron — indeed, since the true value to be taken into account is 

 not 



tW 



but 



tW 



w 



and since, according to the experiments in § 4, — rises to 



1*03, as much as 3 per cent. less. I have given at the com- 

 mencement of this memoir (p. 3) the average numbers which 

 result from Lenz's experiments. Their differences may most 

 probably be attributed to errors of observation and the want 

 of accuracy of the method. If, however, we try to bring the 

 above reflections into harmony with Lenz's number for iron, 

 this can be done perfectly ; for indeed the number for iron is 

 the smallest, and about 3 per cent, below the mean of the 

 three other numbers. 



The earlier view, that the temperature-coefficient u of the 

 resistance of a metallic conductor might be the same for all 

 pure metals, has not been verified by experience; inter alia, 

 the deviations from the mean value O0037 are very consider- 

 able. Various quantities, too, must have an influence upon 

 the value of « — for example, the specific heat, and, just as 

 much, also the specific magnetism. As, according to G. 

 "Wiedemann f, the magnetizability increases with the tempera- 

 ture, and since, as we have shown, the resistance increases 

 with the magnetizing by the current, the resistance of iron 

 must increase with the temperature more rapidly than that of 

 other metals. The value of a is. according to my data, in the 

 mean about 



u =0-0045. 



This value is in fact higher than the mean value for the rest 



* Trans. Roy. Irish Acad. xxii. (1) p. 3. 



| Ppgg. Ann. cxxii, p. 4G-3; Galvanismus (2), ii. (1) p. 604. 



