Till- INDUCTION OR ELECTRIC STRAIN 51 



very easy mode of indicating the direction of electric strain. 

 Suspending a small elongated conductor, say a piece of a match, a 

 straight cotton fibre, or a short piece of wire, by a cocoon fibre 

 passing as nearly as possible through its centre of gravity, the two 

 ends will be oppositely electrified unless the con- 

 ductor is at right angles to the direction of strain. 

 When so electrified the ends will be acted on by 

 forces constituting a couple, which will set the con- 

 ductor with its axis along the direction of strain as 

 a position of stable equilibrium. The position at 

 right angles is evidently unstable. Such a conductor 

 may be termed an "electric pointer." 



Finally, we may indicate the direction of strain 

 by the direction of the force on a very small posi- 

 tively charged body, say a pith ball hung by a silk 

 Its charge should IK- so small that surround- 

 ing charges are not sensibly affected by its presence. 



Direct experiments in the field round a charged Fio. 41. 

 globe or round the knob of n charged Leyden jar 

 show that tin- three methods (1) by two proof planes, (2) by an 

 electric pointer, (3) by a charged pith ball give the same direction 

 for the strain. 



We may use any of tin -se tests for the direction to prove that 



The electric strain just outside a conductor is normal 

 to the surface of the conductor. Perhaps the simplest 

 proof of this is attained by suspending a small electric pointer from 

 tin-end of a thin ebonite or shellac rod and bringing it near the sur- 

 face of an insulated conductor, when it will always take the direction 



IK- normal. A very striking verification is obtained by fixing 

 the pointer at some little distance from the conductor and then 

 introducing another conductor, say the hand, into the field and 

 gradually bringing it up to the pointer. When the second con- 

 ductor is brought quite close to the pointer it will be normal to 

 this second surface. If pith balls be fixed by short silk fibres to 

 the sides of a conductor, when the conductor receives a charge they 

 will all hang in positions agreeing with the supposition that the 

 strain is normal to the su i >r if two proof planes be used, they 



give the maximum charge on separation after being held parallel to 

 the surface and close to it, and no charge when they have been held 

 perpendicular to the surface. This last method enables us also to 

 show that the direction of the strain is from a positively electrified 

 surface and towards one negatively electrified. The law thus proved 

 may be regarded as a case of the more general law that positive 

 electrification t< ncls to move in the direction of strain and negative 

 electrification against it. Had the strain a component parallel to 

 the surface of the conductor, the electrification would move either 

 with or against that component. Such motion of the charge does 

 actually occur \\hile a body i* receiving its electrification. The 



