THE ATOM OF ACTION 183 



does not suggest to us any particular lump of action. 

 We must turn to a form of energy which has a definite 

 and discoverable period of time associated with it, such 

 as a train of light-waves; these carry with them a unit 

 of time, namely, the period of their vibration. The 

 yellow light from sodium consists of aethereal vibrations 

 of period 510 billions to the second. At first sight we 

 seem to be faced with the converse difficulty; we have 

 now our definite period of time; but how are we to cut 

 up into natural units the energy coming from a sodium 

 flame? We should, of course, single out the light pro- 

 ceeding from a single atom, but this will not break up 

 into units unless the atom emits light discontinuously. 



It turns out that the atom does emit light discontin- 

 uously. It sends out a long train of waves and then 

 stops. It has to be restarted by some kind of stimula- 

 tion before it emits again. We do not perceive this 

 intermittence in an ordinary beam of light, because there 

 are myriads of atoms engaged in the production. 



The amount of energy coming away from the sodium 

 atom during any one of these discontinuous emissions 

 is found to be 3-4. io -12 ergs. This energy is, as we 

 have seen, marked by a distinctive period 1-9. io~ 15 sees. 

 We have thus the two ingredients necessary for a 

 natural lump of action. Multiply them together, and 

 we obtain 6-55. io~ 27 erg-seconds. That is the quan- 

 tity h. 



The remarkable law of Nature is that we are con- 

 tinually getting the same numerical results. We may 

 take another source of light — hydrogen, calcium, or any 

 other atom. The energy will be a different number of 

 ergs; the period will be a different number of seconds; 

 but the product will be the same number of erg-seconds. 

 The same applies to X-rays, to gamma rays and to other 



