30 METHODS OF INVESTIGATING 



proportional to the ratio of the quantities of the two elements 

 present, so that by measuring the intensities of both lines and 

 reference to a calibration graph obtained by the use of known 

 mixtures in which the quantities of the two elements are varied 

 the desired determination can be made. 



For measuring the intensities of the spectral lines various 

 methods have been devised, but this is now generally effected 

 by means of the microphotometer. In this instrument a narrow 

 beam of light passes through the photographic negative of the 

 spectrum, and then falls on a photoelectric cell, with the result that 

 a current is induced which is measured by a galvanometer. By 

 means of a rack and pinion the plate is moved very slowly over 

 the beam of light so that this passes in turn through the clear 

 plate and the spectral line. The difference in the galvanometer 

 deflexion obtained for the clear plate and the spectral line gives 

 a measure of the intensity of the line and hence of the amount 

 of material. The principles of spectrographic analysis have, of 

 course, been given here in the broadest and simplest terms. 

 Actually such analysis is full of difficulties and many pre- 

 cautions have to be taken to ensure reliable results. A descrip- 

 tion of these details is outside the scope of this book, and those 

 interested should consult works on spectrographic analysis, 

 particularly the publications by F. Twyman (1935, 1938a, 

 19386), published by Adam Hilger, Ltd., which aim at keeping 

 information on this subject up to date, and the four major works 

 by Lundegardh(1929, 1934, 1936, 1938), which are of particular 

 value to workers with, biological material. The same remarks 

 apply to the use of the microphotometer and the method of 

 calculating results. 



It is to be noted that in the method developed by Lundegardh 

 solutions are analysed, whereas in most other procedures solid 

 samples are used. 



It has already been stated that the spectrograph affords the 

 most sensitive method of measuring small quantities of many 

 elements. This is undoubtedly the case when an arc or spark is 

 used, but results obtained with the flame method indicate that 

 the latter, as used up to now, is capable, broadly speaking, of 

 yielding about the same degree of sensitivity as the polarograph 

 and absorptiometer. 



