Measurement of Mechanical and Electrical Forces. 89 



as the cosine of the inclination ; therefore the beam will set 

 itself at such an angle that the tangent of the inclination is 

 proportional to the force. If, therefore, the point p of the 

 beam can be pulled downwards by a force which is propor- 

 tional to the strength of the current, an electric meter will be 

 the result. The coils shown in the diagrams, which belong to 

 the energy-meter, must be removed and replaced by an elec- 

 tromagnet and armature of peculiar construction. Let there 

 be an electromagnet with pole-pieces a certain distance apart, 

 and let there be between them a wedge of iron at its narrow 

 end increasing in thickness or width rapidly, and towards its 

 thicker parts much more slowly ; then, on moying such a 

 wedge forwards between the poles, but without touching them, 

 it will at first facilitate by its movement magnetic induction 

 at a great rate ; and as it fills up the space, even though the 

 induction through it is greater, yet the increase of that induc- 

 tion is less. Now, as the rate at which magnetic induction is 

 increased by movement measures the force with which such a 

 wedge is pulled forwards, the wedge will, if suitably formed, 

 experience a force with a given current-strength which is less 

 as its entrance is greater, except over a small space near its 

 starting position, where the force should, if possible, be infi- 

 nite. Also, if the wedge is fixed in position and the current 

 made to vary, it will, so long as the magnetic limit is not ap- 

 proached, experience a force which varies as the square of the 

 current ; therefore, if the motion of the wedge is resisted ex- 

 ternally by a force which varies as its displacement, it will 

 enter to such an extent that the amount of its entrance is pro- 

 portional to the current. Let such a wedge be carried by the 

 beam B, so that when it is at its zero position the beam is 

 horizontal ; then the inclination of the beam will be greater 

 when the current is greater, and, except with very weak cur- 

 rents, may be made to vary so that its tangent is proportional 

 to the current. The same principle might be applied in an- 

 other way by causing rotation instead of linear movement of 

 the armature to facilitate induction, and by resisting such 

 motion by a pendulum-weight as before, or by a hair-spring. 

 The armature then would have to be shaped something like 

 an S, and adjusted experimentally so as to give, except with 

 very small currents, a deflection whose tangent is proportional 

 to the current. In Sprague's or Edison's electrolytic meters 

 (the only ones of which I have heard which can lay any claim 

 to being called current-meters at all) a portion only of the 

 current is sent through the meter, and the rest passed by 

 in a shunt. Now, if it were certain that the same proportion 

 of the whole current always passed the meter, there would be 



