26 BIOLOGICAL EFFECTS OF RADIATION 



rule holds for each of the other series. When the impinging beam 

 consists of photons instead of electrons, the hke is observed. Photons 

 are not capable of producing any line of an X-ray series, unless their 

 frequency is superior to that of the absorption edge corresponding to the 



series. 



However, the spectrum of the X-rays produced by fast electrons 

 impinging against a metal does not consist exclusively of the lines of 

 which I have just been speaking. Indeed it consists chiefly of a con- 

 tinuum ; there are X-rays of every frequency from the least which can be 

 detected up to a sharply defined maximum v^^,, which is related to the 

 kinetic energy U of the electrons by the famihar-looking equation 



/iv„.. = U (18) 



max 



This is interpreted as follows: The impinging electrons, as they penetrate 

 into the metal, swiftly lose their energy; some by spending it in dislodging 

 electrons from the atoms and thus bringing about the eventual emission 

 of X-ray spectrum hnes, but others— and the majority— simply by 

 making collisions (presumably with atoms or electrons) in which they are 

 abruptly slowed down or "braked," and the kinetic energy which they 

 lose is transformed immediately into photons. Different electrons may 

 lose different fractions of their energy in different collisions, but no 

 electron can lose more than the total of its initial energy [/ in a single 

 collision, and therefore no photon can have an energy greater than U. 

 The X-rays of the frequency v^,,, which constitutes the upper limit of 

 the continuum, are due to the relatively few cases in which impinging 

 electrons lose the whole of their energy in one single operation the very 

 moment they strike the metal; while all the other X-rays of the con- 

 tinuum are due to cases in which impinging electrons lose their energy in 



steps. 



The X-ray spectrum of an ordinary X-ray tube thus consists of a 

 continuum, on which various spectrum lines are superposed. It is quite 

 possible to choose a particular energy U for the electrons bombarding a 

 target of a particular metal, such that U/h is greater than the frequencies 

 of one or more of the K lines, but less than the frequency pk of the 

 K absorption edge. In such a case the continuum runs smoothly through 

 the frequencies of the K series without exhibiting a single peak at any of 

 them; the peaks spring up above it when, and only when, U is increased 

 past hvK and the upper limit of the continuimi has moved on past vk. If 

 we select a very narrow range of frequencies around some particular 

 frequency v, measure the energy E of the X-rays lying within this range, 

 and plot it as a function of U, we find that as U is increased, E remains 

 zero until U reaches hv, after which it rises very sharply and rapidly. If 

 V coincides with the frequency of some line of the K series, the rate of 



