CONTEMPORARY ADVANCES IN PHYSICS, XXVIII 583 



will be small and perhaps Imperceptible; if by a particle which is 

 approaching its maximum ionizing-power, the difference will be 

 appreciable and of one sign ; if by a particle which is coming to the end 

 of its range, the difference will be considerable and of the opposite 

 sign. So the differential chamber and its detecting device (in these 

 experiments, an oscillograph connected through an amplifier, reacting 

 appreciably to the passage of a single particle) are sensitive above all 

 to particles which are nearing the ends of their ranges; and if a small 

 number of such corpuscles be mingled with even a much greater 

 number of faster charged particles — be they alpha-particles, be they 

 protons, be they the fast electrons produced by gamma-rays— this 

 circumstance, which would cripple any other method, will be almost 

 without effect on it.^^ If the readings of the electrometer are plotted 

 against the air-equivalent of the thickness of matter between the 

 source (of alpha-particles) and the chamber, the resulting curve should 

 not be much distorted from the ideal distribution-in-range curve. 



The curve obtained in this way for the long-range particles of RaC 

 exhibits a notable peak at range 9.0 cm. ; to one side thereof a very 

 much lower hump at smaller range (7.8) ; to the other side a wavy 

 curve with four distinct maxima, which Rutherford and his colleagues 

 deem to be the superposition not of four peaks only, but of seven. 

 I show this portion (Fig. 5) to illustrate the analysis of such a curve 

 for groups. Even the tallest of the peaks just mentioned is a mere 

 molehill compared to the mountain which the principal group of RaC 

 — the 6.9-cm. a-particles which were formerly the only ones known — 

 would form if it could be plotted on the same sheet of paper; for the 

 abundances of the 7.8-cm., 9.0-cm. and 6.9-cm. groups stand to one 

 another as 1 : 44 : 2,000,000. 



These, however, are not the latest words concerning the a-spectrum 

 of RaC. The energies of these groups might be deduced from their 

 ranges, but for this purpose it is necessary to use an empirical curve 

 of energy vs. range which at the time of the foregoing spectrum- 

 analysis had been extended only up to range 8.6 cm. It was desirable 

 to measure the energies of some of these groups directly, not only for 

 their intrinsic interest but in order to carry onward that empirical 

 energy-ZJ.?. -range relation. Recourse must therefore be had to a de- 

 flection-method. 



Now in the usual form of magnet employed in deflection-experiments 



the field pervades the whole of the space between the solid disc-shaped 



faces of two pole-pieces. Were the pole-pieces to be so hollowed out 



" Also a proton near the end of its range can be distinguished from an alpha- 

 particle near the end of its range, on account of the difference in maximum ionization- 

 per-unit-length ("The Nucleus, Second Part," pp. 124-125). 



