324 Prof. W. D. Harkins on the 



Prediction that the Number of Isotopes of Elements 

 of Odd Atomic Number will in general be Con- 

 siderably Less than for Elements of Even Atomic 

 Number. 



The general theory of nuclear structure developed by the 

 writer in the years 1915 to 1917 clearly indicated that in 

 general a smaller number of isotopes is to be expected for 

 elements of odd than of even atomic number. That this is 

 the case among the radioactive elements was pointed out by 

 N. F. Hall in the year 1917. While there are one or two 

 very small regions in which the relations indicate that the 

 isotopes may be more abundant for elements of odd atomic 

 number, it seems likely that with these unimportant 

 exceptions the rule will hold remarkably well. 



In this connexion a classification of the elements into the 

 region of relatively few isotopes (also the region in which in 

 general one isotope is far more abundant than any of the 

 others, when isotopes exist at all) from atomic number 2 

 (He) to 27 (Oo), the region of many isotopes (for elements of 

 even atomic number) from atomic number 28 (Ni) to atomic 

 number 83 (Bi), and the radioactive region between atomic 

 numbers 84 and 92 ; in which, while there seem to be & 

 moderate number of isotopes, many of these exist in such 

 small quantities that they would not be discovered by the 

 positive ray method. 



In the region of abundant isotopes as defined by the 

 writer in the year 1915, the positive ray investigations made 

 thus far confirm the prediction in a remarkable way. Thus, 

 for elements of even atomic number Aston has found about 

 six isotopes for krypton, about seven for xenon, and indi- 

 cations of about the same number for mercury. In sharp 

 contrast w T ith this he has found only one isotope for arsenic, 

 one for iodine, and two for bromine, all elements of odd 

 atomic number, or approximately only one-fifth as many for 

 elements of odd atomic number as for even. It is, of course, 

 quite likely that the ratio may not remain as large as five 

 to one w T hen the region is thoroughly investigated. 



It has been found that neon in the region of relatively 

 few isotopes consists of 2 or 3 isotopes, magnesium of 3,. 

 and silicon of 2 or 3 ; all in elements of even atomic number. 

 While the adjacent elements of odd atomic number have not 

 been investigated by the positive ray method, their atomic 

 weights indicate that each of the elements sodium and 



