34 THE ADDRESSES, LECTURES, ETC., OF 



round plate of metal of any given size, and oppose to it another 

 plate of metal of the same size, approaching them to the distance 

 equal to their own diameter, I have a unit measure of capacity for 

 electricity. I may take these plates as large as I like, or as small 

 as I like ; so long as I mako the distance equal to the diameter, 

 the amount of electrical charge contained by such arrangement is 

 always the same, taking atmospheric air to be the intervening 

 medium in every case. 



EFFECT OF TEMPEEATURE. We have next to consider the 

 effect of temperature upon a conductor. "We have hitherto 

 assumed that a wire of a certain length, of a certain diameter, of 

 a given material, produced a given resistance. That is only true 

 as long as the temperature remains the same ; but with change of 

 temperature, the resistance also changes in a fixed ratio, which is 

 perfectly well ascertained as regards metallic conductors. There- 

 fore, if we want to compare electrical tests, we must either make 

 the actual measurement at a fixed temperature, or we must intro- 

 duce a calculation by which we reduce the values to a standard 

 temperature. "With regard to metals, the ratio in which tempera- 

 ture affects conductivity, or electrical resistance, is a very simple 

 one ; but it is very different with the substances which we call 

 dielectrics, or insulators. There are no such things as absolute 

 insulators ; but the insulator which we mostly apply, namely, 

 gutta percha, increases in its conductivity with increase of tempe- 

 rature in a very rapid ratio, which may be represented by an 

 ascending hyperbolic curve, see Fig. 2, Plate 2 (whereas the ratio 

 applicable to the metals is represented by a straight, descending 

 line). A precise knowledge of these ratios is of great importance 

 to the art of testing cables. 



WHEATSTONE'S BALANCE. In measuring the resistance of 

 metal-conductors, we generally employ a method which is based 

 upon the "Wheatstone's diagram or balance. "We weigh one 

 resistance against another known resistance, and thereby deter- 

 mine its value. I have Wheatstone's diagram here, which is also 

 represented by a diagram (Fig. 3, Plate 2), and may be seen in 

 actual operation. If we send the current of the battery, B, into 

 this point of juncture, a, the current must flow in the two 

 directions, constituting the branches of the diagram, which 

 branches re-unite at the point I, whence the current is brought 



