404 Dr. L. Vegard on 



right spacings for the five faces. We easily find that the 

 lattice gives the following spacings : — 



1 „/ ^ _ a ' . /o 



^100— 9 a ' ^110 — "o" v 2, 6? m — 



>/•♦(# 



^ioi — / , /v =2> ^ooi = c r . . . . (8) 



Comparing these values with the relations (2), we see 

 that our lattice gives just the right spacing. 



Determination of the Parameters. 



§ 5. The general procedure for the determination of the 

 parameters would be the following : — 



We have first, for each face, to find the distribution 

 of point planes within each period equal to the spacing of 

 the face. 



This distribution would be a function of the parameters. 

 For a given distribution of planes we can calculate the dis- 

 tribution of intensities of various orders. Now we have 

 observed relative intensities, or we know the ratio of the 

 intensities of the spectra observed. We have observed four 

 orders for each of the five faces, which should give fifteen 

 ratios — corresponding to fifteen equations. These cannot all 

 be independent. Thus, in our case, the intensities of the 

 faces (100) and (110) only depend on four parameters, while 

 they would give six equations. We are not going to carry 

 out the calculation in this general way ; but we shall 

 adopt a procedure similar to the method of successive 

 approximation. 



We make use of the fact that the hydrogen atoms have a 

 quite small reflecting power, and as the first approximation 

 we can put the hydrogen atoms out of consideration. Then 

 we have merely the four parameters l , /, I', and Z x left for 

 determination. 



Now the problem is very much simplified by the fact that 

 he intensities of the faces (100) and (110) do not depend on 

 the parameters l 0i I, I', and we can determine l x separately. 



First of all, we must remark that the molecular element 

 may be placed in two different ways. 



Either the lines 00 (fig. 2 a) can be parallel to the sides 

 or to the diagonals of the base of the lattice. From the fact 

 that the reflexion from the (100) face shows a nearly normal 



