HYDROGEN INTO ATOMS 



177 



slope of the line would then be q/^.S7, from which, by (49), we could 

 calculate ^1. The slope of this line, however, would be the same if we 

 plot the logarithm of any quantity proportional to K, instead of the 

 logarithm of K itself. If we could assume that Ta = T2 and that B and 

 E were independent of the temperature, then we see from (35) that W^z) 

 should be proportional to K. The values of gi that we might thus obtain 

 by plotting log W-i? would not be very accurate, because the above assump- 

 tions are not strictly correct. However, by using Equation 38, which gives 

 us a relation between K and Wmax. we avoid these difficulties. By (32) and 

 (34) we see that the factor BE is independent of Ta and T2, so that K is 

 strictly proportional to Wmax.- Another important advantage in using this 

 equation in determining q is that it does not involve a knowledge of the 

 pressure within the bulb. It should be remembered that in the experiments 

 at pressures of 10 mm., and above, the pressures of the hydrogen as given 

 in Table III were not measured while the filaments were heated, but were 

 the pressures at which the bulbs were sealed at room temperature. The 

 actual pressures in the bulbs during the experiment must have been con- 

 siderably higher. ^'^ 



It is clear, however, that this uncertainty in the pressure does not affect 

 Wmax. nor the accuracy of K calculated from it by (38). 



By examining Table III we see, as the pressure increases, that W/> in- 

 creases to a maximum and then decreases. In the following table are the 

 maximum values of W© corresponding to various temperatures as given 

 in Table III : 



By plotting 2 log Wmax. against i/T we obtain in fact an almost per- 

 fectly straight line. From the slope of this line we calculate by (50) that 

 q = 85500 calories. If we take T2 = 2400 (the center of the range of 

 temperatures in the experiments), we then find by (49) 



qx — 40300 calories. 



(51) 



*'' In some of the experiments at a pressure of 100 mm. of hj-drogen, the bulbs 

 became so hot at the higher filament temperatures that the glass softened and sucked 

 in slightly. In another experiment the bulb was immersed in water to keep it cool. 

 This cooling did not have any perceptible effect on the value of Wd. 



