54 PRESIDENTIAL ADDRESS SECTION B. 



tive charge of the nucleus. The number of external electrons can 

 be estimated experimentally from the scattering of X-rays because 

 each electron is, comparatively speaking, far apart from its neigh- 

 bours and thus acts as an independent agent. Barkla thus found 

 by trying different elements that the number of electrons is always 

 a little less than half the atomic weight of the scattering element. 

 Again, calculation from the deviation of a -particles (mentioned 

 just above) when they are passed through different metals, leads 

 to a result for the positive charge of the nucleus amounting to 

 the electrical unit multiplied by about half the atomic weight of 

 the element used as a screen. These two experimental lines thus 

 converge to the same result. Again in 1913 Moseley*, a young 

 genius who was killed in the war, applied the principle of the 

 X-ray spectrum, which has proved very fruitful in the case of 

 salt crystals, to the lower elements, and discovered the remarkable 

 law, N = K \~i, in which X is the wavelength of the chief X-ray 

 line given by an element, and N is a number which is a natural 

 integer, going up by 1 as the element investigated goes up the 

 Periodic Table. N is in fact an "atomic number" which depends 

 only on the place of the element in the Periodic Table. Curiouslv 

 enough, when the results are plotted, the bottom of the series is 

 found at helium not at hydrogen, so that .V is not the Chemist's 

 atomic number but one less. It is seen at once that N is also about 

 half the atomic weight, so that three linen of evidence convereie 

 to prove that the nuclear positive charge (and in consequence the 

 number of external electrons) are determined in the case of each 

 element by its position in the Periodic Svstem and are numericallv 

 identical with the "atomic number" as defined above. 



The Periodic Table is thus explained away (except for its 

 arrangement in 8 or 10 columns) since for example sodium metal 

 has 11 plus charges in its nucleus and 11 electrons outside, and 

 the next element (magnesium) has 12 of each, thus putting up its 

 "valency" by one. Valency is of course, as mentioned on page 47, 

 the criterion of chemical properties, which have nothing to do 

 directly with the nucleus of the atoms, but are solely conditioned 

 by the external electrons, and, to speak more particularly, are 

 almost solely conditioned by the number and arrangement of the 

 few outermost electrons which are left over from fitting the atom 

 with a spherical shell. As already indicated on page 51, sodium 

 is monovalent and aluminium trivalent because the completed shellf 

 of all the elements N, O, F, Ne, Na, Mg, Al contains 10 electrons, 

 leaving respectively 1 and 3 over for valency electrons. Similarly 

 the negative valencies 1, 2, and 3 of fluorine, oxygen and nitrogen 

 are due to their possessing respectively 9, 8, and 7 external 

 electrons, whereas 10 are required to make an inactive shell of the 

 same shape as that of neon. The latter is inactive because it has 

 10 external electrons exactly (being the 10th element in th^ 

 Periodic Table) and therefore none over or under the number 



* Phil. Mag., 1913, 1024 and 1914, 703. 



+ The " completed shell" is, it may be repeated, the ionised form of the element. 



