68 BELL SYSTEM TECHNICAL JOURNAL 



to deduce the atomic weights of all the other members of the two se- 

 quences; thus, a radium atom is what is left behind after a uranium 

 atom has ejected three alpha-particles (mass, 4 apiece) and two elec- 

 trons (mass negligible) and its atomic weight should therefore be 226.18. 

 Here we meet a troublous fact. The value of the atomic weight of 

 radium, as measured by no less an expert than the celebrated Honig- 

 schmid, is 225.97 with an uncertainty believed not to exceed three 

 units in the last place. This might be explained by supposing that the 

 element uranium as found in nature is a mixture of several isotopes in 

 relatively large proportions, only one of which is the parent of the 

 uranium-radium series, while the others may be stable or perchance the 

 ancestors of the other series; indeed it is hard to think of any other 

 adequate explanation.^^ 



All three of the sequences terminate in isotopes of the element 82, 

 commonly known (but remember the caution on page 110!) as lead. 

 It is a curious fact that the most rare and precious of all substances 

 should die away by self-transmutation into the one which serves as the 

 symbol for everything which is commonplace, dull and cheap. The 

 atomic weights of the terminating isotopes of the radium and thorium 

 sequences may be guessed in the same manner as that of radium from 

 that of uranium. Starting from radium and from thorium respectively 

 and noting that an atom of radium is destined to eject five alpha- 

 particles and an atom of thorium six during the transformations where- 

 by they turn into atoms of RaG and ThD respectively, we calculate 

 the values 206.0 and 208.1 for the atomic weights of these two isotopes of 

 element 82. Now nearly every sample of lead that has ever served 

 for an atomic-weight determination has yielded a value near 207.2. 

 Yet, when the lead-content of certain minerals rich in uranium and its 

 posterity and deficient in thorium was extracted and investigated, the 

 atomic weights of these samples were found to lie extremely near to 

 206— some of the values recorded are 206.046, 206.048 and 206.08. 

 On the other hand, samples of lead extracted from various minerals 

 rich in thorium and poor in uranium displayed abnormally high 

 atomic weights, values attaining in some instances to 207.9. These 

 are data much more dramatic than the customary outcome of the 

 tedious process of determining an atomic weight; one wonders vainly 

 what chemists would have felt, if they had been published before 



^^ What is commonly called "uranium" contains not only the ancestor of the 

 uranium-radium series, but also one of its descendants, which however is not present 

 in sufficient amount to affect the atomic weight. This is the reason for inserting the 

 words "in relatively large proportions" in the above sentence. The fact that the 

 atomic weight of uranium is not integral might be taken to suggest that it is a mix- 

 ture of integral-weight isotopes. Aston 's latest experiments on stable elements of 

 non-integral atomic weight show, however, that the premise does not necessarily 

 lead to the conclusion. 



