November 9, 1917] 



SCIENCE 



447 



(uranium) ; and lead (82) is the end of the 

 series as now recognized. For our pur- 

 poses, however, we still call the atoms of 

 atomic numbers 1 to 29 the lighter atoms, 

 and from 30 to 92 the heavier atoms. The 

 following table indicates that when de- 

 fined in this way the lighter atoms are ex- 

 tremely more abundant. In the table the 

 weight percentages are given, but it is evi- 

 dent that if these same figures were calcu- 

 lated to atomic percentages they would show 

 even smaller values for the heavier ele- 

 ments. The table shows that although the 

 heavy atoms have been so defined as to in- 

 clude more than twice as many elements 

 as the light atoms, their total abundance 

 is so small as to be relatively insignificant. 

 The data are taken from estimates by 

 Clarke and by Farrington. 



TABLE VI 



IllustTating the Large Proportion in Various Ma- 

 terials of the Elements of Low Atomic 

 Numbers {1-29) 



Percentase of Elements with 

 Atomic Numbers 

 Material 1-29 30-92 



Meteorites as a -whole. 99.99 0.01 



Stone meteorites 99.98 0.02 



Iron meteorites 100.00 0.0 



Igneous rocks 99.85 0.15 



Shale 99.95 0.05 



Sandstone 99.95 0.05 



Lithosphere 99.85 0.15 



It is thus seen that so far as the abun- 

 dance OP THE ELEMENTS IS CONCERNED, THE 

 SYSTEM PLATS OUT AT ABOUT ELEMENT 30, 



and it is of great interest to note that it is 

 just at this point that other remarkable 

 changes occur. For example, up to this 

 point nearly all of the atomic weights on 

 the oxygen basis are very close to whole 

 numbers. On the other hand the elements 

 with higher atomic numbers (28 to 92) 

 have atomic weights which are no closer 

 to whole numbers than if they were 

 wholly accidental. Also, just at this 

 point the atomic weights cease to be those 



predicted by the helium-hydrogen theory 

 of structure presented in this paper 

 (Table III.). This does not mean, how- 

 ever, that the helium-hydrogen system 

 fails at this point, but that the deviations 

 in the atomic weights for the elements of 

 higher number are produced by some com- 

 plicating factor. This would be most easily 

 explained on the hypothesis that isotopes 

 are abundant among the elements of atomic 

 number higher than 28. Such a hypothe- 

 sis should, of course, be confirmed experi- 

 mentally before it is given much credence. 

 It is quite possible, too, that radioactive dis- 

 integrations have proceeded downward in 

 the system as far as iron, and that iron is 

 the end of a disintegration series. If this 

 were true, it would explain the great abun- 

 dance of iron in the meteorites. In what- 

 ever way we may average the analyses of 

 the materials found in meteorites or on 

 earth, the two most striking elements from 

 the standpoint of abundance are oxygen, 

 the most abundant of the elements of very 

 low atomic number (8), and iron, which 

 has the highest atomic number (26) of any 

 very abundant element. 



The fact that the elements which have 

 heavy atoms (atomic numbers 30 to 92, or 

 more than two thirds of the elements) have 

 been formed in such minute amounts 

 would be very much more striking to us if 

 we lived on an earth with a perfectly uni- 

 form composition. On such an earth, 

 formed without any segregation, it is prob- 

 able that almost none of these elements 

 would have been discovered. Quite cer- 

 tainly such elements as gold, silver, iodine 

 and arsenic would not be known, and 

 copper, lead, zinc and tin, if known at all, 

 would be in the form of extremely small 

 specimens. 



In this connection it may be remembered 

 that the earth has the highest density of 

 any of the planets. The data given in 

 Table V. show that in the meteorites, which 



