554 Sir E. Rutherford and Mr. H. Robinson on Mass and 



high speed is identical with the value deduced from electro- 

 chemical data. While the approximate agreement between 

 the theoretical and experimental values for the charged 

 hydrogen atom, determined by Wien and Sir J. J. Thomson 

 from experiments on the anode rays, certainly suggests such 

 an equality, it is at the same time of the greatest importance 

 to test this conclusion as rigorously as possible*. In our 

 ignorance of the origin of mass and of its distribution 

 among the component parts of an atom, it is not a priori 

 impossible that the apparent mass of the helium atom which 

 has lost two electrons might be sensibly different from 

 that ol the neutral atom. General electromagnetic theory 

 would indicate that the electromagnetic mass of the charged 

 helium atom of speed 2 x 10 9 cm. per second would only 

 differ from that of the same atom in slow motion by 1 part 

 in 450. This effect, even if it existed, is too small to detect 

 with certainty by the present experimental methods. At the 

 same time, it should be borne in mind that the variation of 

 e/m with speed has so far only been experimentally tested in 

 the case of the negative electron, and thac no corresponding 

 observations have been made on the variation of the mass 

 with speed of a positively charged atom. In addition, several 

 observers have pointed out that the difference in the accepted 

 atomic weights between uranium and radium is greater than 

 would be anticipated if the radium atom is derived from the 

 uranium atom by the expulsion of three a. particles. A dis- 

 crepancy of opposite sign exists between the atomic weights 

 of radium and lead if the latter, as evidence strongly suggests, 

 is the end product of radium. 



For these reasons, we have made very careful investigations 

 to test whether the observed value of E/m for the u particle 

 is identical within experimental error with the theoretical 

 value for the helium atom. Experiments have been in 

 progress for the past two years to develop the experimental 

 method in order to give results of the accuracy aimed at, 

 viz. to 1 part in 400. 



The deflexions of a narrow pencil of a rays have been 

 measured in both magnetic and electric fields in vacuo by 

 the photographic method. The deflexion in a magnetic field 



* W. Hammer has recently (Annalen der Physik, xliii. p. 653, March 

 1914) published an account of a careful determination of e/m for hydrogen 

 Kanalstrahlen of velocity 2-53x105 cm. sec.-- 1 , the velocity of the rays 

 being measured directly by an oscillating condenser method. He finds 

 c/m = 9775, or about 2 per cent, higher than the calculated value 9570 for 

 the hydrogen ion in electrolysis (by an oversight Hammer quotes 9654 

 for the calculated value), and concludes that the difference can be 

 accounted for by the probable experimental error. 



