RECENT ADVANCES IN SCIENCE 381 



few of the elements of high atomic number. In fact, we have 

 here a revelation of the whole series of elements from hydrogen 

 to uranium, all producing spectra of remarkable similarity, at 

 least so far as their K and L radiations are concerned, but 

 scattered regularly through the whole frequency region, from 

 the ultra-violet, where the isT-lines (Lyman series) for hydrogen, 

 are found up to frequencies for uranium (92 ) 2 or 8,646 times as 

 great. This similarity is due to the fact that the atoms are 

 all similar structures, in their inner portions at least, with nuclei 

 which are exact multiples of the " positive electron," surrounded 

 in each case by electronic orbits which have practically the 

 same relations in all the elements, the radii of these orbits 

 being inversely proportional to the central charge or atomic 

 number. 



A further result from the experimental facts summarised in 

 Moseley's law is that we are driven logically to some form of 

 quantum theory as soon as we begin to introduce energy rela- 

 tions ; for as the electrons are limited to orbits of particular 

 radii, there must be a sudden or explosive loss of energy when- 

 ever an electron is obliged, either by shock from a cathode 

 ray or by radio-active processes, to change its orbit. It must 

 radiate when it passes from an outer to an inner orbit the differ- 

 ence of the energies which it possesses when remaining in each 

 of these orbits, and this difference is a finite and definite amount 

 depending on the structure of the atom itself and not on the 

 influence which provoked the orbital change and the subse- 

 quent radiation. One must of course mention here, as another 

 powerful support to these views, the extraordinary success of 

 the atomic model devised by Bohr, shortly before Moseley's 

 work had appeared and forced us to adopt the essential elements 

 of Bohr's theory. In fact this model, based on the assumption 

 of non-radiating and fixed electronic orbits — an assumption 

 remarkably consonant with Moseley's results — has survived the 

 most critical tests to which any theory can be subjected ; by 

 means of it one can calculate from the values of other known 

 constants, such numbers as the Rdyberg spectroscopic constant 

 and the diameter of the hydrogen atom with remarkable accu- 

 racy (in the first number to *i per cent.). It also deduces 

 from the existence of the well-known Balmer's law for the 

 series of hydrogen lines, a dynamical principle which seems 

 likely to play a fundamental part in the new dynamics of the 



