4 Lankshear, Quantitative Absorption Spectra. 



region recourse must be had to relatively long exposures. 

 Experiments are in progress for the making of a plate at 

 once sensitive and fast in the region X = 3,000 to 2,000. 



The plates are treated by the ordinary i)hotographic 

 methods. 



Exposure. Exposures of from 15 to 20 seconds 

 usually give a readable plate, but, as is shown elsewhere, 

 a readable plate does not necessarilj' mean a correct plate. 



Plotting Results. In taking the photograph, a suitable 

 concentration of the substance being studied is chosen, 

 and exposures taken through thicknesses such as 50, 40, 

 35) 30, 25, 20, 17, 15, 12, 10, 8, 6, 5 mm. If necessary the 

 solution is then diluted down to one-tenth the strength, 

 and the process repeated. Between each exposure the plate 

 is racked forward sufficiently to prevent the overlapping 

 of the successive spectra. 



After development the plate is read. This is done 

 by placing on the plate a wave length scale and reading 

 off the wave length of the last line visible in the given 

 exposure. In many instruments there are arrangements 

 for photographing a wave length scale directly on to the 

 plate. The curve can then be plotted. It is usual to 

 plot oscillation frequencies as abscissae and logarithms 

 of relative thicknesses in millimeters of M/io,000 solution 

 as ordinates. 



Types of Curves In general, it is found that two dis- 

 tinct types of absorption occur, {a) general, {b) selective. 



The general absorption curve may be either straight 

 or may have one or more bends in it, which in some 

 instances take the form of steps. Up to the present time 

 this form of curve has had only a restricted significance 

 for chemists, and is associated with aliphatic compounds 

 and comj)letely saturated aromatic compounds like hexa- 

 hydrobenzene. 



