444 HAROLD F. BLUM 



that careful measurement of the effectiveness of incident monochro- 

 matic beams of high purity cannot yield direct information unless 

 many factors that are not feasible to measure are taken into account; 

 whereas in other cases, exact measurements may yield valuable infor- 

 mation more directly. Thus, enthusiasm for very exact physical 

 measurements should be tempered according to the problem in hand, 

 particularly when the desire for perfection may inhibit obtaining less 

 exact but more feasible measurements. 



Some, but not necessarily all of the above mentioned errors, may 

 be avoided when dealing with photochemical reactions in extracts 

 from living organisms. Many useful applications of action spectra 

 have involved such systems, e.g., the identification of enzymes by 

 Warburg and others. 



It must be admitted that in some instances very rough determina- 

 tions of the limits of the action spectrum may be almost as valuable 

 as more exact ones. These can often be obtained with cut-off filters, 

 by simply determining the effectiveness of an appropriate source 

 through short wavelength and long wavelength cut-off filters. Choice 

 of filters and source depends upon the part of the spectrum concerned, 

 and the intensities required. When very high intensities are needed 

 to elicit a biological response, and also when it is necessary to expose 

 more than small areas, the use of monochromatic radiation may not be 

 feasible, and in such cases determinations with filters are particularly 

 useful. When the incident radiation consists of a series of discrete 

 lines, as in the case of the ubiquitous mercury arc, filters may yield 

 as accurate estimates of action spectrum limits as a monochromator, 

 in some cases possibly more accurate because they permit greater 

 intensities to be used. 



When a substance of kno^vn absorption spectrum is suspected of 

 being the light absorber, an approximate test may be made by esti- 

 mating the relative effectiveness to be expected when different filters 

 are used. For a given wavelength, X: 



S = IsTeS. ' (22) 



where S is the relative effective intensity, h is the relative intensity of 

 the source, and T the transmission of the filter. The molecular ex- 

 tinction coeflacient might be replaced by k or k' . The relative ef- 

 fectiveness Sx through a given filter transmitting wavelengths Xi to 

 X2 is: 



