based upon Electromagnetic Theory. 591 



penetrate to any great depth into a solid body like the earth. 

 Such forces will not appreciably affect the gravitational force. 



2. How can such an atom as pictured for hydrogen 

 possibly radiate the characteristic lines of the hydrogen 

 spectrum ? Or how can an electron ever become detached 

 from it, due to the passage of a stream of bombarding 

 electrons through hydrogen gas, barring an actual collision, 

 which seems most improbable of occurrence ? 



The possible answer to these questions is taken up in detail 

 in a subsequent pfiper, but it seems to be required to make 

 a preliminary reply here. A collision is not necessary to 

 account for the radiation. Electromagnetic theory is able 

 to show that, when the bombarding electron exceeds a 

 certain critical velocity in passing near to the atom, its effect 

 will be to detach a single electron from the hydrogen atom. 

 The action drives the electron out in a straight line from the 

 nucleus to a certain maximum distance, whence it returns 

 again to the nucleus. The acceleration of the electron is not 

 constant, but contains a single periodicity, one of the 

 characteristic lines of the spectrum. The maximum dis- 

 tance to which the electron is driven is always the same 

 fraction of the wave-length of this particular periodicity, 

 and the electron goes out further for long waves and low 

 frequencies than for short waves and high frequencies. The 

 energy radiated in this straight-line motion during one com- 

 plete excursion out and back has been calculated and found 

 to be equal to a constant limes the frequency. This is 

 Planck's quantum relation, and the constant is the equiva- 

 lent of the constant //, which by the use oE this model 

 receives a physical interpretation in the case of light radia- 

 tion from gases and X-radiation from solids. 



The action between an electron thus detached from the 

 nucleus and the part of the atom that if left behind is to a 

 first approximation analogous to the complex case of the 

 action between two or more gyroscopic bodies, for each 

 charge is supposed to be in rotation about an axis. It has 

 been shown that such mutual forces may give rise to just 

 such line series as are met with in the hydrogen spectrum, 

 but the exact calculation of this problem in detail must 

 eventually tax the highest type of mathematical analysis that 

 has yet been devised. 



There seems no room for doubt that new ideas are neces- 

 sary before the real significance of Planck's quantum relation 

 can be understood, and it has seemed well worth while 

 following to the limit the clue suggested by the new electro- 

 magnetic theory. Already this course has led to a model 



