March i;, 192 i] 



NATURE 



81 



electrons at the surface of a hot body that if the 

 latter is made negative in potential relative to its 

 surroundings there is a current discharge which 

 may sometimes be measurable in arnperes. Of 

 course, such a current can pass only one way, 

 negatively from the hot body, or positively to- 

 wards it. So we get the basic principle of the 

 "valve," and so Coolidge provides the electrons 

 for projection against the target in the X-ray 

 bulb which he has designed. At this point we 

 find already the adaptation of our new knowledge 

 of electrons to apparatus of extraordinarily great 

 use to mankind. 



If now we plunge a little deeper into our 

 subject we come to certain most fascinating 

 regions of it, where exploration is still in full 

 progress. In one of these we find the most re- 

 markable connection between moving electrons 

 and electromagnetic waves. One, it seems, 

 can always call up the other, and the action obeys 

 certain precise numerical laws. 



Let us take as an example the production of 

 X-rays in a Coolidge bulb. A plentiful supply of 

 electrons is provided at the cathode by heating a 

 fine spiral of tungsten wire to a high temperature. 

 A high potential difference between cathode and 

 target is provided by some approximate means, 

 and the electrons are hurled at the target, each 

 possessing an amount of energy equal to the pro- 

 duct of the electron charge and the applied 

 potential. Where the electrons strike, some of 

 their energy is converted into electromagnetic 

 waves of very high frequency, the so-called 

 X-rays. Suppose that we measure the energy 

 supplied to each electron — not an easy matter 

 with the usual arrangements, but very easily done 

 if, as in certain experiments of Duane and Hunt 

 at Harvard University, the potential is derived 

 from a great storage battery of 40,000 volts. 

 Suppose, further, that we analyse by the X-ray 

 spectrometer the X-ray radiation that issues from 

 the target. We find that the frequencies of the 

 emitted rays may have a wide range of values, but 

 that the upper limit of the frequencies is always 

 proportional to the energy of the electron, and, 

 therefore, to the potential imposed on the tube. 

 This ratio remains the same no matter what the 

 intensity of the electron discharge, and no matter 

 what the nature of the target. This ratio of elec- 

 tron energy to maximum frequency is a number 

 which has turned up in previous cases where the 

 emission of radiation energy has been measured : 

 it is known as Planck's constant, and denoted 

 by "/i." Its value is 6-55x10-27. Although the 

 constant has been met with before, there is prob- 

 ably no instance where the transformation of 

 energy which it governs is so simply displayed or 

 so easily measured as in the case just described. 



In certain measurements made bv Duane and 

 Hunt and illustrated in Fig. i, the X-ray spectro- 

 meter was set to observe the presence of a certain 

 frequency as soon as it appeared. The potential on 

 the tube was then increased by degrees. The rays 

 of the given frequency appeared as soon as the 

 NO. 2681, VOL. 107] 



energy supplied to the electron was equal to the 

 frequency multiplied by h. As the potential was 

 increased still further these rays increased in in- 

 tensity, as the figure shows. 



It is to be observed that the production of 

 X-rays is no aggregate of individual efforts by 

 separate electrons : each electron produces its own 

 train of X-rays when it strikes the target. There 

 is no sign of any combined action, as, indeed, is 

 evident from the fact that the intensity of the 

 cathode-ray stream is without influence on the 

 frequencies of the X-i4^'s produced. 



The crucial point is that when the energy of 

 an electron is handed over in whole or in part, 

 the frequency of the X-ray waves that take over 

 the energy is determined by the quantity of energy 

 handed over. This explains why there is a limit 

 to the frequency of the X-rays : it is because there 

 are some electrons, though only a fraction of the 

 whole number, which give up all their energy to 

 the formation of X-rays at the moment of strik- 

 ing, before they have lost energy in collisions. 



24 26 28 30 32 34 36 38 40 

 KiLOVOLTS 



FiG. 1.— From Daane and Hunt, Physical Reaieiv, 1915, p. 166. Each 

 curve represents the growth in intensity of a certain wave-length as the 

 voitage applied to the X-ray bulb is increased. The wave-lengths are^: 

 left to right, 0-488, 0-424, 0-377, o'345) o'S'S, 0-308, all in Angstrom 

 units (10 8 cm.). 



The rest of the rays, all those which have lower 

 frequencies, will come from electrons that have 

 lost speed in this way, or possibly have trans- 

 ferred only part of their energy. The atom of 

 the target is playing the part of a transformer, 

 and does not determine the frequency, so far as 

 these effects are concerned. 



All this is wonderful enough ; but the marvel is 

 greatly increased by the discovery that the effect 

 is reciprocal. Just as the swiftly moving electrons 

 excite X-rays, so X-rays when they strike any 

 substance lose their energy, which now appears 

 as the energy of moving electrons. And, again, 

 we find the same variation in the result and the 

 same limit to that variation. Among the electrons 

 so set in motion we find, examining them as soon 

 as possible after their motion has begun, every 

 variety of energy-content up to a certain critical 



