upon the Dimensions of Iron and Steel Bars. 

 Experiment 3 (continued). 



83 



•From the last column of each of the preceding tables we 

 may, I think, safely infer that the elongation is in the duplicate 

 ratio of the mag7ietic intensity of the bar, both when the mag- 

 netism is maintained by the influence of the coil, and in the 

 case of the permanent magnetism after the current has been 

 cut off. The discrepancies observable will, 1 think, be satis- 

 factorily accounted for when we consider the nature of the 

 magnetic actions taking place. "When a bar experiences the 

 inductive influence of a coil traversed by an electrical current, 

 the particles near its axis do not receive as much polarity as 

 those near its surface, because the former have to withstand 

 the opposing inductive influence of a greater number of mag- 

 netic particles than the latter. This phaenomenon will be 

 diminished in the extent of its manifestation with an increase 

 of the electrical force, and will finally disappear when the cur- 

 rent is sufficiently powerful to saturate the iron. Again, when 

 the iron, after having been magnetized by the coil, is aban- 

 doned to its own retentive powers by cutting off the electrical 

 current, the magnetism of the interior particles will suffer a 

 greater amount of deterioration than that of the exterior par- 

 ticles. The polarity of the former may indeed be sometimes 

 actually reversed, as Dr. Scoresby found it to be in some ex- 

 tensive combinations of steel bars. Now whenever such influ- 

 ences as the above occur, so as to make the difltjrent parts of 

 the bar magnetic to a various extent, the elongation will neces- 

 sarily bear a greater proportion to the square of the magnetic 

 intensity measured by tlie balance than would otherwise be 

 the case. 



