Chromosomes and Genes 177 



We are not interested here in a definition of allelism, which could 

 be derived from the elementary facts of cytogenetics, but in the 

 diflFerence between the partners in a pair of alleles, and the problem 

 of why the relation between a pair of alleles is so diflPerent from 

 that between different loci. 



In classic genetics — apart from the presence-absence theory or 

 its later modifications — a pair of alleles is the result of one gene 

 mutating to its allele, and allelism is thus a matter of descent. Rather 

 different meanings have been assigned to this relation. First, it was 

 the absence of the gene which was allelic to its presence (Bateson). 

 When multiple alleles were discovered, a partial absence was as- 

 sumed. This is not different from the idea ( Goldschmidt, 1917fl, 

 1920a) that a gene is a group of molecules and that mutation con- 

 sists in changing the quantity of the gene, thus making alleles 

 different quantities of the same substance. We have discussed the 

 recent revival of this idea by Serra in the form of gene quanta, sub- 

 sections of the gene with specific structure which individually can 

 produce haptene groups; mutation, then, denotes this production by 

 only a part of the substructures. In the course of our discussion we 

 shall meet, or have already met with, facts which favor the quanti- 

 tative idea of allelic relationship: the dosage experiments; the behavior 

 of loci opposite a deficiency; the deficiencies caused by irradiation; 

 some, though not all, facts relating to multiple alleles; and many 

 facts of phenogenetics, as discussed in detail in Goldschmidt (1927, 

 1938a). 



Opposed to this quantitative view of allelism is the qualitative 

 one, which has been expressed in different ways by various authors, 

 too many to be all recounted. One position is that the chromosome 

 is a giant protein molecule; the genes, side chains or prosthetic 

 groups; and mutants, replacement of the prosthetic group by a similar 

 but different group. Another view is that the gene is a single molecule, 

 and a mutant either a stereoisomere or the replacement of one 

 radical by a different one. A variant is that doubling of the gene re- 

 quires a template and that a mutant is the result of a small failure in 

 the exact duplication of the molecule by a little mishap in the tem- 

 plate mechanism. The result might be the absence of a radical or pros- 

 thetic group or a stereoisomeric pattern change within the molecule. 

 An interesting variant has been mentioned by Haldane (1954), that 

 between two copying actions the genie material has changed so that 

 it cannot be copied exactly with the available materials. Somewhat 

 related, though different in detail, is the concept based on the idea 



