on the Length and the Resistance of Iron Bars. 459 



each individual wire was stretched on the average by a weight of 

 312 grms. These wires were surrounded by all four spirals. 

 In these four series of experiments the following results were ob- 

 tained : — 



No. 



L. 



d. 



Tension. 



Resistance. 



Unmagnetized 



Magnetized. 



12 

 13 

 14 

 15 



2700 

 27000 

 26980 

 44480 



1-3 

 0-36 

 0-36 

 0-24 



grms. 













 312 



0-3417 

 59-527 

 58-500 

 160-870 



greater 

 59,563 

 58-534 



160-920 



In these and in the previous series, wires were used which in 

 previous experiments shortened on being magnetized (10, 11, 

 15), such as lengthen (12, 13, 14), bundles whose ends were 

 within (14), and those whose ends were outside the ends of 

 the spirals (12, 13): under all circumstances an increase of 

 resistance was observed; its magnitude z is seen from the 

 following Table : — 



No. 



Resistance 

 = w. 



z. 



z : w. 



10 



7-2990 



0-0025 



1 : 2920 



11 



4-2805 



00025 



1:1710 



12 



0-3417 



not measurable. 





13 



59527 



0-036 



1 : 1650 



14 



58-500 



0034 



1:1670 



15 



160-870 



0-050 



1 : 3220 



Wires 10 and 15 were taken from the same coil, as also were 

 wires 11, 13, and 14. The increase in resistance for the first 

 amounts on the average to 1 : 3070 (Thomson found 1 : 3000) ; 

 for the second, 1 : 1680. 



I think these experiments justify the conclusion that magneti- 

 zation in all cases increases the resistance of iron in the direction 

 of the magnetic axis. I think, further, that I can account for the 

 process which brings about this increase. One consequence of 

 magnetization must be an approximation of the molecules. If 

 the molecular distances in all parts of an iron bar were the same, 

 a necessary consequence of this approximation would be a de- 

 crease in resistance. But the molecular distances are not every- 

 where the same ; the molecules are rather arranged in groups — 

 in a line, for instance, thus : — 



and 



