CONTEMPORARY ADVANCES IN PHYSICS 113 



whole of it) was certainly due to the fact that the faster particles 

 went farther into the layer and struck more nuclei. There is no need 

 to labor the point: it is obviously desirable to do the experiments with 

 a film so thin that each oncoming proton either strikes a nucleus with 

 its full and unabated initial energy, or else goes through the film and 

 away without any impact at all. This ideal was closely approached 

 by Oliphant and Rutherford, when they got countable numbers of 

 fragments from films of lithium and boron (deposited on blocks of 

 steel or iron) which were so thin as to be invisible, and of which the 

 latter was known to consist of only seven-tenths as many atoms as 

 would suffice to cover the iron surface with a single monatomic layer. 

 (The curves of Fig. 16 were obtained with these films.) 



This is a success which proves it possible to investigate films con- 

 sisting each of only a single isotope of the element in question; for 

 feeble as are the ways of separating isotopes in all but a few very 

 favorable cases, they yet are powerful enough to produce pure mon- 

 atomic layers. This article will amply show how valuable will be the 

 privilege of getting data from a single isotope, of lithium or boron for 

 example; already there are several cases of important antagonistic 

 theories, the decisions between which will be given once and for all 

 by such data. 



The apparatus devised by E. O. Lawrence and developed in his 

 school at Berkeley is of a singular ingenuity, inasmuch as in it ions 

 are accelerated until their energies are such as would be derived from 

 an unimpeded fall through a potential-difference of literally millions 

 of volts, and yet the greatest voltage-difference at any moment between 

 any two points of the apparatus is only a few thousands. It owes 

 its elegant compactness to the lucky fact that when a charged particle 

 is moving in a plane at right angles to a constant magnetic field, and 

 consequently is describing a succession of circles, the time which it 

 takes to describe a single circle is the same whatever its speed. One 

 sees this readily by writing down the familiar equation, 



mv"l p — Ilevjc, 



in which e, m, v stand for the charge (in electrostatic units), mass, and 

 speed of the ion and p for the radius of curvature of the circle, and 

 on the right we have the force exerted by the magnetic field H upon 

 the ion and on the left the so-called "centrifugal force" to which it is 

 equal. The radius p varies directly as v, but the time T = Irp/v 

 which the ion takes to describe a circle is independent of v. This is 

 no longer true if the ion is moving so fast that the foregoing classical 

 equation must be replaced by its relativistic analogue, but fortunately 



