1018 The Excitation of soft char act eristic X-rays. 



220 volts, which may have been due to residual gas as it 

 was independent of the voltage above 220 volts, under the 

 working conditions. The curve shows the setting-in of new 

 radiation at 286 volts. The points marked O were taken 

 with rising, and those marked □ with falling potentials. 

 The observations are consistent with each other, showing that 

 the rise in the current is not due to a generation of gas as 

 the potential is raised. The voltages were measured with 

 reference to the negative end of the filament, and the 

 voltage drop along this was between 6 and 7 volts in all 

 the experiments. 



The deflexions due to this characteristic radiation from 

 carbon were rather small, and some of the points do not fall 

 very well on the curve drawn. Nevertheless, they are 

 sufficient to determine the critical voltage and, assuming 

 the quantum relation to hold good, the wave-lengths of the 

 radiation pretty accurately. Numerous independent tests 

 showed that there was no increase at 280 volts and a 

 measurable increase at 300. The reading at 290 is not so 

 certain. We believe, however, that it is impossible for the 

 uncorrected exciting voltage to lie outside the limits 280- 

 290 volts, and it is probably close to 286. There are two 

 corrections — one for the drop of potential in the hot filament 

 and the other for the work done on the electron when it 

 passes into the anode. These cancel one another in this 

 case to within about 1 volt, which is about as accurately as 

 they can be estimated. If we admit that these radiations are 

 soft X-rays excited according to the same laws as hard X-rays 

 and apply the quantum relation, we find that the wave- 

 length of the shortest member of the group must lie between 

 the limits 42*7 and 44'3 A.U., and is probably very close to 

 43*4 A.U. Extrapolating from K^ for Al, using Moseley's 

 relation, we find that K^ for carbon should have a wave- 

 length of 45' 5 A.U., and from a formula given by Uhler * 

 50'25 A.U. It is thus practically certain that we are dealing 

 here with the K X-rays from carbon, although, as has 

 already been pointed out, we are not yet able to supply a direct 

 proof that the radiations generated are actually K , X-rays. 



Molybdenum. — The target was cut from a sheet of pure 

 molybdenum. The false zero line attributed provisionally to 

 gas was lower and the vacuum steadier than in the work 

 with carbon, probably because of the higher temperatures 

 (above the melting-point of copper) attained in the clean- 

 ups. Two out of a number of sets of readings, all of which 

 were concordant, are shown in the figure. The voltage drop 



* Phys. Eev. vol. ix. p. 325 (1917). 



