and Phenomena of Magnetic Induction. 267 



necessarily transmitted in modification of the direct action of 

 the magnet on the nearest extremity of the bar. These diffe- 

 rent forces, indeed, may be easily calculated, but the manner in 

 which they act reciprocally under the different arrangements re- 

 mains to be determined. 



Notwithstanding what has now been stated, the results derived 

 from the foregoing experiments may be considered, perhaps, as 

 so far satisfactory, as affording a comparison of the quantity of 

 magnetism directly induced upon the nearer end of a bar, and 

 of that developed at the remote extremity, and indicating also 

 the proportion of resistance in the iron itself to the perfect trans- 

 mission of the inductive energy from end to end. 



Another mode of experimenting, however, which subsequently 

 occurred to me, is much more practical, satisfactory and useful, 

 affording, by direct experiment, the actual quantity of magne- 

 tism induced into any iron or steel bar or bars of different ca- 

 pacity, in reference to the energy of the inducing magnet itself, 

 together with the exact proportion of influence produced on each 

 at any given distance. 



The arrangement, in this case, which was peculiarly simple, 

 required only this adaptation, that the bars of iron or steel used 

 for trying the quantity of induction should be of the same 

 length, or nearly of the same length, as the magnet employed. 

 And when comparative experiments were made on the relative 

 capacities of different kinds of iron and steel, or iron and steel 

 of different tempers or degrees of hardness, it further required 

 (in order to obtain at once direct and final results) that each bar 

 should be of the same dimensions as the magnet made use of. 



Fig. 6. Plate II. represents this new arrangement. M is a 

 magnet placed in the direction of the east or west point of the 

 compass C at ani/ distance, — say at two Jbcal lengths from the 

 centre of the compass, when the deviation by either pole is care- 

 fully observed. An iron bar, I, of the same dimensions as the 

 magnet, is then placed exactly over it, separated by two little 

 blocks of wood, or other substances of equal thickness, adapted 

 to the distance required for the experiment. As the magnetism 

 now induced through the whole length of the iron bar is of the 

 opposite kind to that of the inductive magnet, its action on the 

 compass must also be opposite; the deviation is consequently 

 found to be diminished, and that (measuring by the tangents) 



