304 PETER GUTHRIE TAIT 



is required for a given charge the greater is the capacity of the receiver. Conversely, 

 the damage which can be done by the discharge, being equal to the work required 

 to produce the charge, is proportional to the square of the charge, and inversely to 

 the capacity of the receiver. Or, what comes to the same thing, it is proportional 

 to the square of the potential and to the capacity of the conductor directly. Thus 

 a given quantity of electricity gives a greater shock the smaller the capacity of the 

 conductor which contains it. And two conductors, charged to the same potential, 

 give shocks proportional to their capacities. But in every case, a doubling of the 

 charge, or a doubling of the potential, in any conductor, produces a fourfold shock. 



The only other point I need notice is the nature of the distribution of electricity 

 on a conductor. I say on a conductor, because it is entirely confined to the surface. 

 The law is it is always so arranged that its attractions or repulsions in various 

 directions exactly balance one another at every point in the substance of the conductor. 

 It is a most remarkable fact that this is always possible, and in every case in one 

 way only. When the conductor is a single sphere the distribution is uniform. When 

 it is elongated the quantity of electricity per square inch of its surface is greater 

 at the ends than in the middle ; and this disproportion is greater the greater is the 

 ratio of the length to the transverse diameter. Hence on a very elongated body, 

 terminating in a point, for instance, the electric density that is, the quantity per 

 square inch of surface may be exceedingly great at the point while small every- 

 where else. Now in proportion to the square of the electric density is the outward 

 pressure of the electricity tending to escape by forcing a passage through the 

 surrounding air. It appears from experiments on the small scale which we can 

 make with an electrical machine, that the electric density requisite to force a passage 

 through the air increases under given circumstances, at first approximately as the 

 square root of the distance which has to be traversed, but afterwards much more 

 slowly, so that it is probable that the potential required to give a mile-long flash 

 of lightning may not be of an order very much higher than that producible in our 

 laboratories. 



But from what I have said you will see at once that under similar circumstances 

 an elongated body must have a great advantage over a rounded one in effecting 

 a discharge of electricity. This is easily proved by trial. [The electric machine 

 being in vigorous action, and giving a rapid series of sparks, a pointed rod connected 

 with the ground was brought into the neighbourhood, and the sparks ceased at 

 once.] In this simple experiment you see the whole theory and practical importance 

 of a lightning conductor. But, as a warning, and by no means an unnecessary one, 

 I shall vary the conditions a little and try again. [The pointed rod was now 

 insulated, and produced no observable effect.] Thus you see the difference between 

 a proper lightning-rod and one which is worse than useless, positively dangerous. 

 There is another simple way in which I can destroy its usefulness namely, by 

 putting a little glass cap on the most important part of it, its point, and thus 

 rendering impossible all the benefits it was originally calculated to bestow. [The 

 pointed rod was again connected with the ground, but furnished with a little glass 

 cap. It produced no effect till it was brought within four or five inches of one of 

 the conductors of the machine, and then sparks passed to it.] You must be strangely 



