444 Mr Whiddington, Note on the Velocity of X-ray Electrons 



If the distance between T and *S is 1 cm., and we assume that 

 the distance between the target of a Coohdge tube and the sht 

 is (say) 8 cm., it follows that with the arrangement above, there 

 is an available intensity 64 times as great. 



Even if only 1 milliampere is used the available intensity will 

 be more than ten times as great. 



One disadvantage of the arrangement is that the coils producing 

 a field in B, produce a small but appreciable magnetic field in the 

 region STC. 



The result is that the cathode-rays from C are deflected. This 

 deflexion must be balanced out by an additional compensating coil 

 in series with the main coils. It is fortunate in this connexion that 

 it is the fastest rays arriving at T which are the most effective 

 X-ray parents and which are least affected by the stray field*. 



The photographic film used was 10 cm. long and the current 

 in the field coils was adjusted to give a range of velocities from 

 3 X 10» to IQi" cm./sec. approximately, yielding a dispersion of 

 0-071 X 10^ cm./sec. per milhmetre. 



It is possible to take a photograph with the apparatus in half 

 an hour although longer exposures are desirable. Using a platinum 

 or rhodium target, for example, and a copper foil over the slit, 

 the strongest lines on the film correspond to velocities 6-02 x 10^ 

 and 5-74 x 10^ cm./sec. 



If we apply the quantum relation to the X-ray K doublet of 

 copper we get the corresponding velocities of electrons carrying the 

 same energy to be 5-61 x 10^ and 5-33 x 10^ cm./sec. 



It is interesting to note that while the differences are precisely 

 the same in both experimental and calculated cases, the actual 

 values differ by about 10 per cent. Whether this diflerence is real 

 or due to a defect somewhere in the apparatus must be determined 

 by further investigation. 



It is worth remembering, however, that both the present writer 

 and Hullf found that velocities of parent electrons distinctly in 

 excess of the value demanded by the quantum relation were re- 

 quired for the production of a fluorescent radiation of any definite 

 wave length. 



It seems possible that the difference if it be real may be accounted 

 for by taking into account the sums of the individual energies 

 required for the simultaneous excitation of the radiations of both 

 K and L series. 



It is hoped to publish a full and extended account of this work 

 very shortly. 



* See also Whiddington, Proc. Roy. Soc. 1911. 

 f Physical Review, 1916. 



