4 Prof. W. L. Bragg and Messrs. James and Bosanquet : 



This expression has a maximum value—, when 



fie 



*-i (5 > 



The Determination of Q. 



3. The object of the measurements described here and 

 in the previous paper has been the determination of the 

 absolute value of the quantity F over a range of glancing 

 angles. F can be calculated if Q is determined experi- 

 mentally, since the other quantities in the expression for Q 

 are known. 



In the former paper. Q was found by determining the 



171 



reflecting power -=- for a crystal face, for which the formula 



-y- = w- holds good. The linear absorption-coefficient //, 



was measured directly, by passing the homogeneous beam 

 through plates of rock-salt of various thicknesses. It w T as 

 pointed out, however, that in the case of reflexions at small 

 glancing-angles, the value of //, determined in this way is 

 certainly too low. When X-rays pass through a crystal 

 in such a direction that the crystal reflects the rays, the 

 absorption of the transmitted beam is greater than that 

 for other directions. This effect was first noticed in the 

 case of the diamond by W. H. Bragg*, and the effect 

 also exists in the case of rock-salt j\ Fig, '2 shows some 

 determinations made with the apparatus arranged as in 

 fig. 1, but with the ionization-chamber placed so as to 

 receive the transmitted beam. The ordinates represent 

 the strength of the transmitted beam, plotted for a number 

 of crystal settings. It will be seen that the transmitted 

 beam is reduced by 20 per cent, when the crystal is 

 set at 4° 50' ±. the angle at which it reflects the X-rays- 

 most strongly. 



The value of /jl is therefore greater than its normal 

 value (10*7), for beams passing through the crystal in any 

 direction in which thev are reflected strongly, and the 

 values of Q, calculated using the value 10*7 for fi, will 



* W. H. Bragg, Phil. Mag-, vol. xxvii. p. 881 (iMay 1914). 



f Cp. the figures obtained by Kutherford and Andrade in their deter- 

 mination of the wave-lerjgth of y-ravs, Phil. Mag. ser. 6, vol. xxviii. 

 no. 164, p. 263 (August 1914). 



t Owing to a zero error in setting the crystal plate, this angle differs 



