484 WEBER OX THE NATURE OF DIAMAGNETISM. 



The same comparison yields for the even series, where the 

 commutator occupied the position B on removing the bismuth 

 from the annular conductor, always a somewhat smaller mean 

 value. 



2. Series 514-02 = 5l7-4-3-3S 



4. „ 517'04 = 5l7-2— 0-16 



6. „ 51'J-90 = 523-0-3-10 



It should be borne in mind that the state of the galvanometer 

 observed before the commencement of each series was not ex- 

 actly that of rest. To avoid the uncertainty arising froin this, 

 the reading made previous to each series may be wholly excluded 

 from the calculation, and the comparison restricted to the mean 

 values of the several series. The comparison of the mean value 

 of the 2nd to the 6th series, with the mean from the immediately 

 preceding and succeeding series, then gives the following re- 

 sults : — 



2. Series 514-02 = 516-S2 — 2*80 



3. „ 518-72 = 515-53 + 3-19 



4. „ 5l7-04 = 520-36-3-32 



5. „ 522-00 = 518-47 + 3-53 



6. „ 519-90=525-43-5-53 



We see then also from this, that in the uneven series, in which 

 the commutator occupied the position A while the bisnmth was 

 withdrawn from the annular conductor, the state of the gal- 

 vanometer was constantly somewhat higher, and that the reverse 

 occurred in the even series in which the commutator had the 

 position B on the removal of the bar of bismuth. The dif- 

 ferences are somewhat greater for the last than for the first series, 

 which is easily explained from the change of induction being 

 gradually accelerated. 



Corresponding experiments were now made for the purpose 

 of direct comparison, the bar of bismuth bemg exchanged for a 

 slender bar of iron. The induced current was then so powerful 

 that no repetition could be made as in the case of the bismuth, 

 and that only the extreme end of the iron bar could be inserted 

 in the annular conductor. And even then the induced current 

 was so powerful that the deviation of the needle could not be 

 observed on the galvanometer, but merely the direction, whether 

 the position of the galvanometer rose, i.e. went from lower to 

 higher divisions of the scale, or the reverse. 



