PHYSICAL PRINCIPLES OF CHEMICAL REACTIONS 249 



electron from one part of the molecule to another (as in the ferro-ferric 

 transformation). Thus, certain complexes of hemin and muscle protein 

 form addition compounds with CO; these are dissociated and the CO 

 released not only by light absorption in the hemin, but also by light 

 absorption in the ultraviolet absorption band of the protein (Kubowitz 

 and Haas, 1933), which should be interpreted as energy migration by a 

 mechanism of the sensitized fluorescence type (Franck and Livingston, 

 1949). The actual mechanism of the CO release seems to involve an 

 internal conversion, for the binding of CO to the complex is of such a 

 nature that the molecule is only weakly coupled to the hemin and direct 

 dissociation is therefore unlikely. 



Another example of a localized action is the mutation of a single gene 

 by ionizing radiations. According to a commonly held view the mutation 

 itself may be interpreted as a molecular reorganization — a sort of isom- 

 erization — of a molecular aggregate which can be considered in many 

 respects as a single large molecule. Whether or not this view is correct, 

 it is instructive to examine the elementary processes whereby such 

 a reorganization may be effected. One possible means is the simul- 

 taneous excitation by the penetrating charged particle of a great number 

 of oscillatory degrees of freedom, the oscillation energy then fluctuating 

 between various modes until it is sufficiently concentrated in one or sev- 

 eral of them that reorganization to the appropriate second configuration of 

 the molecule can occur. This is extremely unlikely, however, because the 

 probabilities for vibrational excitation are simply not very great (Platz- 

 man, 1952b). It is common that interpretations correlate the isomeriza- 

 tion with an excitation or ionization, but do not explain properly how 

 energy communicated to the electronic system of a molecule can cause the 

 reorganization of its atomic positions (cf., for example, Schrodinger, 1945, 

 p. 66). Internal conversion provides the answer, and suggests, moreover, 

 that the conversion of the electronic excitation energy into "internal 

 thermal " energy of the molecule is not distributed over a large area, but is 

 rather concentrated into just a few oscillatory modes, and indeed, that 

 part of the electronic energy goes directly into the potential energy which 

 isomerization requires. It seems likely that many different modes of 

 internal conversion are possible, only one or a few of which lead to the 

 proper isomerization (and the others to lethal effects, or no effects at all). 

 Thus there exists a certain probability that excitation causes mutation 

 (unimolecularly) , and this will be less than unity and may be very much 

 less. 



Nonlocalized action must involve a mechanism of different nature, for 

 here a more delicate alteration at many points of a large area occurs. 

 As examples may be mentioned the denaturation of proteins, and very 

 likely the breaking of chromosomes. (There will clearly be many cases 

 that are compUcated by the fact that both locahzed and nonlocalized 



