I 



soMi- COM i-Mi'oi^.iKy .ini.iw i:s ix I'lnsics // 171 



through air. loiiiid i>iil> tlinv inst.uici-s of di-lkrlions cxrivdiiiu !1()°; 

 (". T. R. Wilson, ph<)t()v;r.i|)liin^; tlu- paths of 503 fast olL-ctroiis through 

 air. found forty-four instanivs of di-Hi-rtions i-xcccding !K)°. Wliili-. 

 in i;i-neral, it would he liardK- fair to make such comparisons without 

 allowing for the relative energies of the two kinds of particles, the 

 <!ilTerencc iti order of lu.igniliide is so great that we ma\- accept it as 

 t\pica!. 



.Mort^)ver, the ekrtron will he deflected hy the atom-electrons 

 as well as by the nucleus, and will not disarrangt- the atom-electrons 

 so hadly on its wa\- through the atom-s\stern. TIicm' dclleclions 



Fig. 10 



will he superposed upon the deflection produced hy the nucleus, and 

 will modify the distrihution-in-angle function F from form (7) into 

 s<jme other form. Such modifications ha\-e been suspected by several 

 investigators; for example, by Crowther and Schonland in their 

 stufly of the deflections of ^■ery fast electrons hy metal atoms. It 

 has been argued by Wentzel, howex'er, that the distril)ution-in-anglc 

 function observed in their experiments departed from the form (7) 

 not because the atom-electrons were interfering with the fast electrons, 

 but l)ecausc some of the deflecterl electrons had been de\iati-(l h\' 

 several atom-nuclei in succession. 



C. Davisson and C. H. Kunsman, in the laboratories of the Western 

 Electric Company, made the first definite attempt to produce elcctron- 

 deHet tions under conditions in which the distrihution-in-angle function 

 would disclose the influence of the atom-electrons. To do this it 

 was desirable to use, not the fast electrons from radioactive atoms 

 which previous experimenters had employed, but slow electrons of 

 controllable spectl. .A diagram of their apparatus is shown in Fig. ID 

 and a photograph in F"ig. 11. The electrons proceed from a hot 

 filament at Fi. strike the metal target at 7", and are detlecteil through 

 various angles; the shielded collector ZJ,, swinging from one angle to 



