the Galvanic Current through Iron. 219 



one diagonal of the Wheatstone square ; between B and D a 

 Daniel? s element (in the order of succession copper — zinc) 

 was inserted, between F and H two, and between Gr and E 

 three such elements in the same direction and one behind an- 

 other. The plug at b was never put in, those at a and h 

 always; if beside these there were now fixed in place only 



g and/ c, e,f c, g, d g and d e, e, d e and d 



then there were 



12 3 4 5 6 



elements inserted. 



The double and triple elements can thus be simply excluded; 

 the single one can, it is true, be only weakened by the secon- 

 dary closing at c ; I found, however, that in reality it is thereby 

 likewise excluded. 



The resistance in the elements hardly came into considera- 

 tion in comparison with that of the rest of the circuit ; the 

 intensity of the current therefore varied nearly as the electro- 

 motive force. I at first intended to make these experiments 

 with the wires already used for the others ; but they exhibited 

 a behaviour so abnormal that I was obliged to replace them 

 by fresh ones that had not been submitted to galvanic and 

 magnetic treatment. I will cite only a few instances of these 

 abnormities. Through a thin annealed iron wire (the magne- 

 tization-experiments made with which are not included among 

 those above selected) the current of 2D was sent, Avithout in- 

 termission, for one minute each time, first in one and then in 

 the other direction. The following resistances a and b were 

 found: 



a 14-57 14-42 14-29 14'23 14-14 14-11 14-06 14-06 14'03 

 14-32 14-03 14-00 13*98 13-96 13*95 13*94 13*94 



b 13*74 13*37 14-15 13*71 13*38 13*69 13*43 13*81 13*39 

 13*82 13*44 13*89 13-44 13-81 13*51 13-79 13*39 



AVhile, therefore, a apart from a gradual diminution, 

 amounting to about 4 per cent., connected with the gradual 

 lowering of the surrounding temperature, shows constancy, 

 the values of b form a toierablv regular zigzag line, the 

 maxima of which deviate on the average 3 per cent, of the 

 entire ordinates. Further, the next day the mean resistance 

 (the wire being differently fixed), on the employment of 



ID, 14=14-921 n-i 



>, «-!= 14-921 



3D, l ,- 3 =16*33J" 3 ^- U(0 



