CONTEMPORARY ADVANCES IN PHYSICS 637 



fall on only a part of the screen ; if a magnetic field is applied in the 

 proper sense to the region which they traverse, the spot on which they 

 fall moves sidewise; but the flashes are so infrequent that the shift is 

 not obvious, and only by lengthy countings can one be sure that more 

 of them appear in one place and less in another when the field is on 

 than when it is off. Rutherford however managed to make countings 

 enough to prove that the shift is of the order of magnitude to be ex- 

 pected, if the particles are protons having the speed inferred from their 

 range; and incidentally that they are positively charged, something 

 which has been taken for granted but which requires proof. It was 

 with the long-range corpuscles expelled from aluminium, from phos- 

 phorus and from fluorine that he achieved these results. 



The problem was then taken up by Stetter in Vienna; he tried the 

 scheme developed to so high a pitch by Aston in his famous series of 

 experiments on isotopes — a scheme of which I shall say only that al- 

 though it involves both electric and magnetic fields, they are so ar- 

 ranged that corpuscles having a common value of charge-to-mass ratio 

 are brought to a common focus irrespective of their speeds (so long as 

 these are not dispersed over too wide an interval) ; therefore, by locating 

 the focus, one may recognize the kind of corpuscle. In Fig. 3 appears 

 a part of his apparatus: the source of alpha-particles at Q, the sheet of 

 transmutable substance at 5 or S' , and beneath it the system of long 

 narrow parallel channels which Stetter arranged so that only a beam 

 of corpuscles following almost perfectly parallel paths should enter 

 the deflecting fields below. 



Shifting from place to place the microscope with which he examined 

 the screen beyond the deflecting fields, and counting the scintillations, 

 Stetter found three foci which in the curve of Fig. 4 appear as three 

 peaks (the ordinate being the number of flashes in unit time over a 

 given area, the abscissa the distance of the midpoint of this area from a 

 point taken as zero). From the positions of these foci on the screen 

 it followed that the one on the right was due to corpuscles having the 

 charge-to-mass ratio of protons; the one in the middle, to alpha-parti- 

 cles; the one on the left of which only a part appears, to corpuscles 

 having a charge-to-mass ratio half as great as that of an alpha-particle. 

 Two then are proof of alpha-rays deflected by the metal at 5, some of 

 which had lost one-half of their positive charge through picking up an 

 electron somewhere in their careers; the third is evidence of protons, 

 and strong evidence, for Stetter estimates the uncertainty of his 

 measurement of charge-to-mass ratio as no greater than five per cent. 

 The curve of Fig. 4 was got with aluminium as the metal which the 

 alpha-rays bombarded. Curves were obtained in the same way with 



