426 HAROLD F. BLUM 



Most measurements of absorption spectra are made in solutions 

 that do not scatter light to an appreciable extent, and most modern 

 methods are adapted to this particular type of measurement. Such 

 methods have been adequately treated in various places {e.g., 5, 8, 9a), 

 and hence only a brief discussion of the general principles of absorp- 

 tion spectroscopy is needed here. The measurements consist essen- 

 tially in determining for each wavelength the ratio Iq/I for a sample of 

 the material of known thickness, l; from this information the ab- 

 sorption coefficient or extinction coefficient may be calculated. 



The most direct method — not commonly used now except for the 

 rotation-vibration region of the infrared — is to measure the intensity 

 of a monochromatic beam successively with and without the absorb- 

 ing material in the beam; the former gives /, the latter /o. The in- 

 tensity may be measured with an instrument whose response is di- 

 rectly proportional to the intensity, for example, a thermocouple; 

 or with an instrument in which correction can be made for any de- 

 parture from such proportionality. Measurement in absolute units 

 is unnecessary since only the ratio of intensities is required, and the 

 instrument need not be nonselective as regards wavelength. 



Another method employs the spectrograph. In this instrument 

 the spectrum of some source, say an iron arc, dispersed by means of a 

 prism or grating, is arranged to fall on a photographic plate. Photo- 

 graphs are made with and without the absorbing medium interposed, 

 and the intensities of the various wavelengths determined by analysis 

 of the photographic negative. This method has the advantages that 

 it provides a permanent record, and that a wide range of wavelengths 

 may be measured with a single pair of exposures. It has the dis- 

 advantage that the response of the photographic plate is not directly 

 proportional to the intensity and hence correction must be made in 

 terms of the blackening characteristics of the plate; this requires 

 appropriate equipment and is time-consuming. In some of the older 

 investigations this correction was not made, leading to serious errors 

 in interpretation. Such mistakes are to be found in the biological 

 hterature somewhat later than in the physical. 



The methods most in use at present employ spectrophotometry in 

 one form or another. By splitting the monochromatic beam into 

 two beams of equal intensity and interposing the absorbing material 

 in one, the reduction in intensity in passing through this material 

 may be determined by comparison of the two beams photometri- 

 cally. This gives the ratio h/I directly without determining the 



