32 THE ADDRESSES, LECTURES, ETC., OF 



but also with this leakage, which leakage we may regard as ti 

 separate discharge, bearing a certain proportion of the principal 

 discharge, or as offering a certain proportionate resistance to the 

 flow of the liquid. By increasing the length of the tube, the 

 leakage will increase or its resistance to leakage will diminish in 

 the same proportion. 



Now, if at the outlet end we had a water-wheel, and made that 

 water-wheel turn by the flow of the liquid, we should then have a 

 case analogous to working an electrical instrument at the end of 

 our line, and it is at once apparent that we should endeavour to 

 lose as little of our current on the road as possible, either by con- 

 duction, or by induction, or by absorption into the material com- 

 posing the pipe, in order that we may obtain the greatest possible 

 effect from a given source of power and a given size of conductor. 

 This illustration serves to bring home to our minds the nature of 

 the forces and resisting influences which we have to deal with in 

 submarine telegraphy. 



At the root of all electrical measurement is what is called Ohm's 

 law, which says that the current, or the amount of the flow, is 

 equal to the electromotive force divided by the resistance. That 

 is a law which applies not only to electricity, but also to heat and 

 other forces. It stands to reason that the amount of electricity 

 flowing through a conductor is as the electromotive force (or the 

 height of the column), and, inversely, as the resistance. All sub- 

 stances do not, however, offer the same resistance to the electric 

 current. If this pipe was rough in the inside it would offer 

 greater resistance to the flow of water than it would if it was 

 smooth. So, also, if we have a conductor of one metal, such as 

 iron, we find greater resistance than when we have the same 

 length, and the same area of conductor of another metal, such as 

 copper. 



That brings us to the measurement of electrical resistances. 

 There is a table here showing the conductivity of different metals, 

 which vary in an extraordinary ratio. Thus, pure copper conducts 

 fifty-five times better than mercury ; whereas, the copper of com- 

 merce very often conducts only about five or six times more than 

 mercury. 



MEASUEE OF ELECTEICAL RESISTANCE. In order to measure 

 resistance, we must have a measure or unit of resistance, just as 



