222 LINUS PAULING 



carbon atoms in the cyclohcxenc molecule, for example, are connected by a 

 double bond. Every molecule has a structure, as was pointed out for the first 

 time by Butlerow, and the properties of the substance consisting of these mole- 

 cules are determined by their structure : we may say that the cyclohexenc mole- 

 cule is a system that can be shown experimentally to be resolvable into six carbon 

 nuclei, ten hydrogen nuclei, and fort>'-six electrons, and that can be shown to 

 have certain other structural properties, such as values 1-33 Â, 1-54 Â, etc., for 

 the average distances between nuclei in the molecule in its normal state; but it 

 is not resolvable by any experimental technique into one carbon-carbon double 

 bond, five carbon-carbon single bonds, and ten carbon-hydrogen bonds — these 

 bonds are theoretical constructs, ideahzations, with the aid of which the chemist 

 during the past one hundred years has developed a convenient and extremely 

 valuable theory. The theory of resonance constitutes an extension of this classical 

 structure theory of organic chemistry; it is based upon the same idealiza- 

 tions, the bonds between atoms, as classical structure theory, with the important 

 extension that in describing the benzene molecule or the amide group two 

 arrangements of these bonds are used, rather than only one. The theory of 

 resonance in chemistry is an essentially qualitative theory, which, like the classical 

 structure theory, depends for its successful application largely upon a chemical 

 feeling that is developed through practice; the theory is a part of chemical 

 structure theory, which has an essentially empirical (inductive) basis; it is not 

 just a branch of quantum mechanics. 



The pyrimidine and purine groups of nucleic acids provide another interesting 

 application of the theory of resonance. The molecules of adenine, guanine, 

 thymine and cytosine, and of other pyrimidines and purines, have their carbon, 

 nitrogen and oxygen atoms in a single plane, as is predicted for molecules with 

 structures that can be described as involving resonance among valence-bond 

 structures in which the double bonds occupy a number of different positions, 

 so that the property of planarity characteristic of the atoms adjacent to a double 

 bond becomes associated with nearly all of the bonds in the molecule. The 

 observed inter-atomic distances in these pyrimidines and purines [9] are also 

 found to correspond not to 0% or 100% double-bond character, but to inter- 

 mediate amounts of double-bond character. 



I think that it is likely that, as further investigations of haemoglobins A and S 

 are carried out, it will be found that the apparently very small difference in amino- 

 acid composition of these two haemoglobins, amounting to perhaps a difference 

 in only two residues per molecule, leads to a large difference in the way in which 

 the polypeptide chains are folded, and that this difference in folding changes the 

 properties of the molecules significantly, in particular the ionization constants 

 of some acidic or basic groups. 



The discovery of the abnormal varieties of human haemoglobin has led to 

 some quantitative information about the rate of evolution of human beings. 

 It was shown by Allison that the sickle-cell heterozygotes, with a mixture of 

 haemoglobins A and S in their red cells, are protected against malaria. We may 

 imagine that in a highly malarial region a mutation that converted the gene for 

 A into the gene for S occurred, that after some generations there were a number 



