Magneto-Electric Induction. 453 



It appears from these numbers and from their graphic 

 representation in line 1, fig. A, that the torsion varies directly 



Fig. A. 



as the rate. So exactly is this the case that we may in future 

 with perfect confidence reduce from one rate to another or to 

 any arbitrary common rate when the rate varies between ex- 

 periments, to test the effect of variation of other kinds. And 

 we have, if the magnets and torsion keep constant, a very 

 exact and not inconvenient instrument for measuring the rate 

 of rotation of machinery. 



II. The Relation between Distance and Deflexion. — Two series 

 of experiments were made to determine this relationship. 

 We need only reproduce here the longest and most complete. 

 The apparatus used was that shown in tig. 1, Plate XIII., which 

 is drawn to a scale one third true size. 



The copper disk, c, was replaced by one of ebonite, to the 

 lower surface of which the metal plate could be attached by 

 a drop of weak gum. The copper disk used was only 0'27 

 millim. thick ; so that all its parts might be considered at the 

 same distance from the magnet. The magnet used was 3j 7 

 x 2 x A inch. The upper end of the torsion-thread was at- 

 tached by a screw-clip to the brass* rod, and graduated in 

 millimetres. This could be slid up and down, so as to vary 

 the distance between the magnet and the disk. To get the 

 absolute distance, the mica screen between them was removed, 

 and a piece of copper exactly 1 millim. thick laid on the 

 magnet, and the disk lowered till it just began to rest on the 

 copper. Then the copper was replaced by the mica, and the 



