216 LINUS PAULING 



molecule A is able to influence its environment so as to lead to the production 

 directly of the second molecule A. This possibility is appealing because of its 

 simplicity. The suggestion was made by Jordan [i] that the quantum-mechanical 

 resonance phenomenon might lead to a special stabihty of an aggregate of two 

 identical molecules, AA, relative to an aggregate of two similar but non-identical 

 molecules, AB. A simple analogy is with a system of one electron and two 

 identical nuclei, in which there is a strong interaction (formation of a one-electron 

 bond) as compared with a system of one electron and two non-identical nuclei, 

 in which there is only a weak interaction, the electron remaining attached pre- 

 ferentially to one of the nuclei. It was pointed out by PauUng & Delbrück [2], 

 however, that the special quantum-mechanical interaction energy between large 

 identical molecules, such as protein molecules, is negligible in comparison with 

 the interaction energy of either identical or non-identical large molecules, and 

 that this mechanism could not lead by a direct process to the synthesis of replicas 

 of a molecule. Instead, Pauling & Delbrück suggested that the process of dupli- 

 cation is a two-step process, involving the use of molecule A as a template for 

 the synthesis of the complementary molecule, A"i, and then the use of A'^ as 

 a template for the synthesis of a molecule complementary to it, and identical 

 with A. 



The idea of complementariness in structure as the cause of biological specificity 

 is an old one. The principal evidence for the theory comes from the field of 

 immunochemistry, and the first, rather vague, suggestion of complementariness 

 was contained in Ehrlich's lock-and-key explanation of immunological speci- 

 ficity. The idea that antigen molecules and the homologous antibody molecules 

 have complementary molecular structures was suggested by Breinl & Haurowitz 

 [3], Alexander [4], and Mudd [5], in the period 1930 to 1932, and a detailed 

 theory of the structure and process of formation of antibodies based upon the 

 principles of molecular structure was formulated in 1940 [6]. 



Already in 1940 there was available a considerable amount of evidence in 

 support of the theory of complementariness in structure of antigen and anti- 

 body, in the experimental results obtained by Landsteiner and collaborators 

 through the study of the serological properties of azoproteins containing haptenic 

 groups of known simple structure. A great amount of additional evidence in 

 support of the theory was then gathered by the execution of experiments designed 

 specifically to test the theory [7]. For example, it was found that the introduction 

 of a methyl group in place of a rneta hydrogen atom in the benzene ring of the 

 benzoate ion decreases the combining power of the hapten with antibody mole- 

 cules homologous to p-azobenzoic acid azoprotcin to about one-fifth of the com- 

 bining power of the benzoate ion itself. This fact is explained in a simple way as 

 the result of a resistance to fitting the larger methyl group, which has an effective 

 radius of about 2-0 Â, into the region of the antibody which in the process of 

 its synthesis was occupied by a hydrogen atom of the haptenic group of the 

 immunizing azoprotein, this hydrogen atom having an effective radius of about 

 1-2 Â. The only conclusion that can be drawn from this experiment is that the 

 antibody is closely complementary in structure to the haptenic group of the 

 azoprotein, and that the fit between the complementary structures is a close one. 



