382 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCES VOL. 11, NO. IG 



the energy emitted per transformed atom, the number of alpha parti- 

 cles emitted per transformed atom, etc. If his observations extend 

 over a time t, he wishes to know the number of atoms transformed 

 during that time; i.e., 'KNt where X is the transformation constant, 

 and N is the total number of atoms present. He is primarily inter- 

 ested in XA^, rather than N. 



When only a single element is being considered it is a matter of 

 indifference whether the result is expressed in terms of grams, or of 

 N, or of \N. But when similar measurements have been made for 

 two different elements, and it is desired to compare the results obtained 

 these must each be expressed in terms of the quantity, XA^, involved, 

 if fundamental relations are to be obtained. For example, if it is 

 desired to compare the heating effect produced by the transformation of 

 radium with that produced by the transformation of rad'um emanation, 

 then these effects are expressed in gram calories per hour per gram of 

 radium, and per that amount of radium emanation that can exist in 

 equilibrium with one gram of radium ; \N is the same for each of these 

 quantities. This is quite satisfactory when only related elements in an 

 unbranched portion of a family are being considered ; but if a lateral 

 branch arises between the first and the second element, then a more 

 complicated description of the reference amount of the second element 

 is required. For example, uranium-II branches into uranium-Y 

 and ionium, some 4 per cent going into the former ; consequently the 

 heating effect of one gram of uranium-II is not strictly comparable with 

 the heating effect of the amount of ionium that can exist in equilibrium 

 with one gram of uranium-II, but with 1/0.96 of the latter amount. 



When the effects of elements belonging to unrelated families are to 

 be compared, they will likewise be referred to quantities having the 

 same value for \N; i.e., to quantities so chosen that the same number 

 of atoms take part in the production of the effect. For one element 

 the result may be expressed in terms of the effect per gram, but the 

 corresponding quantity of the other element cannot be very directly 

 expressed with our existing terminology. 



This difffculty would be removed if there were a name to denote the 

 amount of a radio-element corresponding to a fixed value of XA^^. 

 For example, if XA^ for one gram of radium is equal to k, and that 

 amount of any element for which XA^ = fe is called an r, then an r 

 of any element will be radioactively comparable in amount to an r 

 of any other element, whether related or not. A gram of radium will 

 contain one r of radium, a curie will contain one r of radium emanation, 



