from an Arc and from Incandescent Metals. 255 



Current heating Leakage 



the wire Positive charge Negative charge 



I ampere No leak No leak 



II amperes No leak 12 divisions in 30" 

 12 amperes No leak 70 divisions in 30" 

 1*3 amperes 40 divisions in 30" 350 divisions in 30" 



Wire melted when further heated. 



(24) A point in connection with these experiments with 

 incandescent wires, which is worthy of notice, is that the positive 

 carriers are present in the gas near the wire just as soon as it 

 becomes luminous. If a conductor charged negatively is near the 

 wire, and the temperature of the wire gradually raised, we always 

 notice that leakage of the charge begins just when the wire 

 becomes luminous. This point having been observed when work- 

 ing with a platinum wire, a number of experiments were made 

 with platinum, German-silver, and brass wires, and it was always 

 found that just when the luminosity of the wire could be observed 

 a negatively charged body near it began to leak. It is difficult to 

 see why the power to produce ionisation should thus accompany 

 the luminosity, but the agreement can hardly be merely a coinci- 

 dence. Perhaps we should look on the luminosity as due to the 

 ionisation of the layer of gas close to the wire. 



(25) The conductivity of the gas drawn away from an incan- 

 descent wire increases very rapidly as the temperature of the wire 

 is raised. In the following table the current used to heat the 

 wire is given, and also the conductivity measured by forcing the 

 gas through the apparatus shown in Fig. 1, and observing the 

 leakage of the terminal B after it has been charged to a high 

 potential. 



Current through Conductivity of gas 



the wire proportional to 



2-02 amperes 12 



2-25 „ 60 



2-44 „ 135 



2-61 „ 330 



2-75 „ 450 



2-89 „ 635 



The resistance of the platinum wire when these different 

 currents are used was obtained by measuring the difference of 

 potential between its ends, and in this way its temperature was 

 estimated. The numbers given in the above table would cor- 

 respond to a range of temperature of from about 1300° C. to 

 1800° C. 



The gas taken from the wire can discharge negative electricity, 

 as we have seen, as soon as the wire is luminous, but before 



18—2 



