BERNARD L. STREHLER 



245 



in chemical reactions as indicated in Fig. 15, similar to a diagram in 

 our book (Johnson, Eyring, and Polissar, 1954). Thus thermodynamics 

 will be satisfied if 



kn 



W2 ^ AF 



A'l + A'2 fli fiv 



The problem is to find actual systems where ^2/(^1 +^2) is this large. 

 I know of none. I would rather believe this high efficiency is obtained 

 by processes in which the free energy of additional reactions is used 



Fig. 15. 



to pile up sufficient potential energy for the emission without absorb- 

 ing kinetic energy from the heat reservoir by a not thermodynamically 

 impossible but by a nevertheless unknown specialized mechanism. If 

 one thinks of a Planck energy density versus frequency distribution 

 curve transformed so that each abscissa, representing a frequency, is 

 lengthened by the free energy donated by the chemical reaction to 

 the emission process, the ordinates give the maximum intensity to be 

 expected for the changed frequencies. The actual distribution curve 

 can only be predicted from a detailed knowledge of reaction mechan- 

 ism. 



One qualitative argument against this type of efficient mechanism is 

 to be found in the fact that kinetic energy of reaction is never passed 

 along in an efficient manner to make thermal chains. It seems 

 enormously more likely, as I have already stated, that enough reactions 



