CONTEMPORARY ADVANCES IN PHYSICS 149 



This makes it desirable to consider at some length how resonance may 

 be detected. There are the following ways: 



(a) When the target is thick, one may vary the energy Kq which 

 the particles possess when they strike the target-face Kq (usually by 

 varying the density of gas between the target-face and the source 

 of the alpha-particles) and plot the integral distribution-in-range curve 

 for many different values of Kq. Let us suppose that there is a certain 

 proton-group evoked only by alpha-particles having energy between 

 Ka and Kh, the notation being so chosen that Kh < Ka < Kq. Then it 

 will be found that as Ko is lowered, the step and plateau which reveal 

 the group will remain unaltered until Ko drops below a certain critical 

 value (to be identified with Ka) after which they will fade out. 



(b) In the foregoing conditions, one may use a very thin ionization- 

 chamber and plot instead of the integral distribution-in-range curve a 

 curve of the sort in Fig. 10, or the sort described on page 129 of which 

 the ordinate stands for the number of fragments producing more than 

 a certain chosen amount of ionization in the chamber. There will be 

 various peaks in the curve corresponding to various groups, and if any 

 of these is produced by "resonance" it will at first remain unaltered 

 and then gradually disappear as Ko is lowered. 



(c) When targets thin enough to be completely traversed by the 

 alpha-particles are available, one may leave Kq unchanged and increase 

 the thickness t of the target. The energies of the impinging particles 

 in a target then vary from i^o down to a minimum value Ki which 

 depends on /. If curves of any of the foregoing kinds be plotted for 

 various values of Ki, and if any of the groups is produced by resonance, 

 then the step or the peak corresponding to this group may be absent 

 when Ki is high (i.e. with the thinnest target) and will then make its 

 appearance when Ki is lowered past a certain critical value (again to 

 be identified with Ka). 



(d) If the target is so very thin that the loss of speed suffered by 

 the alpha-particles in going through is negligible, and Ki is sensibly 

 equal to i^o. then when Kq is varied the groups should appear and 

 disappear when it becomes equal to Ka and Kb, respectively. 



(e) Without subjecting the fragments to any analysis, one may 

 simply measure the total number thereof (or rather, the total number 

 having ranges superior to some fixed minimum) as function of i^o- 

 Suppose the target to be thick; then, if all the proton-groups are 

 evoked by resonance, the curve should display a sequence of steps 

 and plateaux; if in addition to such there are groups which are evoked 

 by particles of any energy over a wide interval, the steps need not 

 vanish, but the plateaux should slope upward and may be curved. 



