SOML CONIluMI'OR.IRV ADVANCIIS IN I'HYSICS-lll 309 



energy can i-otjununii-aU" no encr>;y at all to a merrury atom; and an 

 electron with ") or () et|uivalcnt volts of energy can transfer only 4.0 

 of them. It is conceivable that other conditions may he found to 

 j{o\ern the orbits of the electrons, so that the atoms shall have only 

 the prescribed enerj^y-Nalues and no others; it is even conceivable 

 that the conception of electron-(jrbits may be discarded; but 'the 

 interpretation of the terms in the formula (7j as energies will, in all 

 human probability, be permanent. 



The foregoing Rule is thus very strongly based; but let us neverthe- 

 less rephrase it in a somewhat milder form as follows: The idea that 

 radiant energy of frequency v is emitted in packets of the amount hv, 

 and the contemporary theory of atomic structure, between them give a 

 attractive and appealing account of spectra in general, and a convincingly 

 exact explanation of two spectra in particular. 



But what has happened meanwhile to the Vibrator, to the oscil- 

 lating electron, to the postulated electrified particle of which the 

 vibrations caused light-waves to spread out from around it like 

 sound wa\es from a bell? It has disappeared from the picture; or 

 rather, since the attempt to account for the frequencies of a spectrum 

 as the natural frequencies of an elastic framework was abandoned, 

 no one has tried to re-insert it. But there are some who will never be 

 quite happy with any new conception, until the vibrator is estab- 

 lished as a part of it. 



Ionization, the total removal of an electron from an atom, affords 

 another chance to see whether radiant energy beha\es as though it 

 could be absorbed only in complete packets of amount hf. That 

 it requires a certain definite amount of energy to deprive an atom 

 of its l(X)sest electron, an amount characteristic of the atom, may 

 now be regarded as an experimental result quite beyond question, 

 cUid not requiring the support of any special theory. Thus, a free- 

 Hying electron may remove the loosest electron from a free mercury 

 atom which it strikes, if its energy amounts to 10.4 equivalent volts, not 

 less; or the loosest electron from a helium atom if its energy amounts 

 to at least 24.6 equivalent volts. If radiant energy of frequency v goes 

 about in parcels of magnitude hv, the frequency of a parcel which 

 amounts just exactly to 10.4 equivalent \olts is j'o = 2.5.3. 10'*, corre- 

 sponding to a wave length of 1 188A. Light of inferior frequency should 

 be unable to ionize a mercury atom; light of just that frequency should 

 just be able to ionize it; light of a higher frequency v should be able 

 to ionize the atom, and in addition confer upon the released electron 

 an additional amount of kinetic energy equal to h (v—Vo). The same 

 could be said, with appropriate numerical changes, for every other 



