138 LINUS PAULING 



around these bonds such that N — H . . . O hydrogen bonds are formed, with the 

 nitrogen-oxygen distance approximately 279 A. 



In all of these configurations of polypeptide chains and in all of the structures 

 for simple peptides and amides that have been rehably determined there is one 

 structural feature that would not be predicted by the classical structure theory 

 of organic chemistry. This is the planarity of the amide group. It is found by 

 experiment that the planarity of the group consisting of the C atom, the N atom, 

 and the four atoms bonded to them is preserved to within a mean deviation less 

 than 0-05 A from the median plane. This structural feature was predicted in a 

 straightforward way [15] by use of the theory of resonance, which is an extension 

 of the classical chemical structure theory. According to the theory of resonance 

 the structure of the amide group, as of other molecules for which a single distri- 

 bution of chemical valence bonds does not provide a satisfactory representation, 

 can be described in terms of two distributions (in some cases more) of the 

 valence bonds. For the amide groups we write the two following electronic 

 structures : 



aC H aC H 



\ , ••/ \ , / 



C— N C'=N 



/ \ / \ 



O: aC :0: aC 



I II 



The group may be described as having a structure that represents the superposi- 

 tion of these two electronic structures : it may be described as a hybrid of these 

 structures, with the C — N bond a bond with partial double-bond character. 

 The observed distance C — N, 1-32 A (0-15 A less than the distance aC — N), 

 indicates about 40% double-bond character for this bond, so that the two struc- 

 tures I and II shown above are described as contributing in the ratio about 

 60% to 40% to the structure of the amide group. With 40% double-bond charac- 

 ter for the C — N bond, the group can be expected to have the property of 

 coplanarity to an extent that can be described quantitatively as about 40% of 

 that of a molecule containing the carbon-carbon double bond. 



The theory of resonance has been subjected to severe criticism; as I under- 

 stand the criticism, it is based largely upon the fact that the theory involves the 

 use of structures (such as the structures I and II given above for the amide 

 group, or the two Kekule structures for benzene) that do not have independent 

 existence in reality. It is true that these structures, use of which is made in the 

 theory of resonance, are idealizations, and do not have existence in reality. 

 However, if the argument were to be accepted as a vaUd argument against the 

 theory of resonance and the theory were in consequence to be abandoned, it 

 would be necessary also, for the sake of consistency, to abandon the whole 

 structure theory of organic chemistry, because the structural elements that are 

 used in classical structure theory — the carbon-carbon single bond, the carbon- 

 carbon double bond, etc. — also are ideahzations, having no existence in reality 

 [16, 17]. There is no rigorous way of showing by experiment that two of the 



