150 MORRIS LOEB 



own properties and those of their compounds before they 

 were actually discovered. Success has also attended the at- 

 tempts to correct atomic weights in several cases where the 

 elements appeared misplaced in the original tables and 

 were assigned to positions more in accordance with their 

 properties, but necessitating the assignment of new atomic 

 weights. The weakest point of the table lies in the position 

 of tellurium, which should fall into the sixth group, but is 

 found to have a higher atomic weight than iodine, which un- 

 doubtedly belongs to the same series in the seventh group. 

 Efforts to explain this discrepancy have so far been unavail- 

 ing. There are also a number of elements derived from the so- 

 called rare earths whose place in the system is not readily as- 

 signable. In the latter case, however, it may be said, as well 

 as in that of the atmospheric gases, argon, helium, neon, and 

 krypton, that their properties and atomic weights are not so 

 well established as to cast doubt upon the theory through 

 their failure to coincide with it. One interesting result of the 

 theory is that of limiting the probable number of chemical 

 elements to about 120, since the actual number of blank 

 spaces is limited, and since it is extremely unlikely that any 

 elements remain to be discovered with an atomic weight less 

 than that of hydrogen or greater than that of uranium. 



Among the physical properties which appear as periodic 

 functions of the atomic weight may be mentioned the densi- 

 ties of the uncombined elements and of their oxides, fusi- 

 bility, atomic volume, crystalline structure of the compounds, 

 coefficient of expansion, refractive index, conductivity for 

 heat and electricity, color, and velocity as ions. 



As an indication of some purely chemical periodicities the 

 following conspectus has been arranged, in which the elements 

 are indicated by their positions in the above table, and are 

 generally enumerated in such order that the one which shows 

 the property in the most marked degree has precedence. The 



