1890.] on WekUncj hi/ Electricity. 191 



similar length and area of silver would, however, destroy only some 

 615 foot-pounds of electrical energy in the same time, giving only 

 two-thirds of a unit of heat. Iron (which is nearly as good a con- 

 ductor as silver) is therefore, as you will see, happily placed between 

 silver and copper on the one hand, and German silver on the 

 other. 



It is extremely likely that at the temperature at which welding can 

 be performed, the resistance of iron to the passage of an electric current 

 is increased to very much more than eleven-fold that which it had 

 at 32'^ Fahr., because, probably, the welding temperature is about 

 3000° Fahr., while, as we saw from Dr. Hopkinson's curve, at 

 1532^ Fahr., we get eleven times the resistance there is at freezing- 

 point. But assume the electrical resistance to be increased only to 

 eleven times that which the iron had when cold. What follows ? 

 Why this: tl:at, a piece of iron 1 foot long, and having a section of 

 1 square inch, would, under these circumstances, destroy in a second 

 of time 40,700 foot-j)ounds of electrical energy. But, as you see, the 

 bar which is being heated, is much shorter, than a foot. It is only 

 about 2 inches, and thus it only destroys about one-sixth of this, or 

 about 6600 foot-pounds of electrical energy j)er second of time, equal 

 to about nine units of heat, or a little more. But the specific heat 

 of wrought iron being only -11379 — water, as you know, beiug unitv 

 — these ten units would raise one pound weight of iron 90 deo-rees in 

 each second. But the portion heated up is only about two-thirds of a 

 pound, and it would be heated, therefore, 135° Fahr. each second • 

 but, as I have told you, as the temperature increases, the resistance, 

 and therefore the heating effect, increases In a lengthened trial 

 with a machine dealing with pieces of good bar iron having a 

 sectional area of about 1 square inch, the maximum heat developed 

 per second of time was 18 units, and the welding heat was reached in 

 22 seconds. 



Now I shall have to refer, as an illustration of electrical phe- 

 nomena, to a very old friend for this purpose, viz. water. Suppose it is 

 a question of working a hydraulic lift, or anything of that kind. If 

 you have 100 gallons of water, multiplied by 50 lbs. of pressure, you 

 get 5000 gallon-pounds. If you multiply 50 gallons by 100 lbs. of 

 pressure, you equally get 5000 gallon-pounds. Similarly, if you 

 multiply 100 amperes of electrical current by 50 volts of j^ressure of 

 electrical current, you get 5000 watts, which is the equivalent in this 

 illustration of the gallon-pounds of the water ; and if you multiply 

 50 amperes by 100 volts, you equally get 5000 watts. From what I 

 have told you as to the resistance of metals, it is clear that 

 for welding purjjoses we want the electrical energy in the form of 

 large quantity and of low pressure. So that if I have at my disposal 

 a total energy of 5,000,000 watts, it may for some purposes suit 

 me to have it in the form of 1000 amperes of quantity by 5000 

 volts of pressure, but for welding I should undoubtedly jDrefer to 

 have it in the form of larger quantity and low 2^i't3ssure, say, 



