L'ss PHYSIOLOGICAL GENETICS 



concepl may, however, Lead to a reconsideration of all these 



points, as \\ ill become clear later. 



We may point finally to an important point. Since the gene 

 must needs give rise to two daughter genes of the same constitu- 

 tion in cell division, it is generally assumed that the gene divides 

 into two like an organism. In Goldschmidt's theory, this is not 

 the case, hut the cell produces as a consequence of the auto- 

 catalytic action of the gene more gene molecules, of which an 

 equal number is adsorbed to the daughter chromosome at the 

 time of division of the chromosome. In a recent paper, Haldane 

 (1935), who is opposed to the theory of gene quantities, comes 

 to the conclusion that the gene is something specific which cannot 

 consist of many similar parts because it might be activated by a 

 single light quantum. Therefore, it cannot reproduce by 

 fission. From a biological point of view, one might speak of a 

 parent and two daughter genes. From a chemical point of view, 

 one must think of a model gene and a copy of it produced in the 

 cell. For an explanation, Haldane looks forward to wave 

 mechanics. Similar views have also been expressed by other 

 authors (Koltzoff, 1935, see page 301). 



B. Mutations as Changes of Side Chains or Residues 



A set of facts that has led to numerous speculations regarding 

 the nature of the gene is found in the occurrence of what are 

 called nowadays unstable genes. The facts were discussed by 

 De Vries (1903). They were later taken up and analyzed by 

 Emerson (1914, 1917, 1922) and Correns (1905, 1919, 1928) and 

 more recently by Anderson and Eyster (1928), Eyster (1924- 

 1929), Demerec, (1928jf.), Imai (1925/.), et al. Correns (1919) 

 seems to have been the first to draw inferences concerning the 

 nature of the gene. A recent review of the literature on unstable 

 genes has been given by Demerec (19356). The most typical 

 cases are found in plants where a peculiar variegation is caused by 

 unstable genes. Generally speaking, they are produced by a 

 single mutant gene, which must, however, undergo some changes 

 during individual development, which may be described in 

 general terms as somatic mutation. It occurs most frequently 

 as a reversal toward Wild type but also toward intermediate 

 conditions. The details are rather complicated, as the following 

 description of the phenotypes by Eyster (1928) shows: 



