480 Genetic Theory and Evolution 



ments could be understood as statistical consequences of the classic 

 theory of the gene and its mutation, together with the work of 

 selection or drift upon the genotypes in interbreeding populations, 

 especially by selection of modifier systems. There is no doubt that 

 population genetics can explain all evolutionary happenings within 

 a species, its diversification, fluctuation, adaptation to local environ- 

 ment. From the point of view of theoretical genetics, discussed in 

 this book, population genetics and its results do not require further 

 consideration. The actual problems (e.g., the relative importance of 

 drift in S. Wright's sense) are evolutionary problems, frequently 

 more or less problems of ecology and not problems of genetic theory. 

 Genetic results and points of view enter the picture in special cases: 

 the role of heterochromatin, the theory of polygenes, the consequences 

 of chromatin rearrangements, the role of cytoplasmic diversification. 

 We have already referred to these special problems in their proper 

 places. As far as genetic theory is concerned, population genetics 

 deals only with elementary Mendelian genetics based upon the classic 

 theory of the gene and its mutation. 



However, problems of genetic theory are bound to enter the 

 picture when the question is raised whether or not the genie per- 

 mutations of all kinds, together with selection and drift, suffice to 

 explain evolution beyond the level of adaptive or non-adaptive 

 diversification within the species in interbreeding populations. Neo- 

 Darwinian evolutionists take it for granted that the findings on the 

 intraspecific level can be simply extended to cover all evolution. This 

 means that the diversification by selection of small mutant deviations 

 and their accumulation slowly builds up species from subspecies, 

 genera from species, and so up to phyla. I have repeatedly discussed 

 this conclusion and tried to show the numerous difficulties with which 

 it meets (see Goldschmidt, 1940, 1945Z7,c, 1948c, 195 la). We are not 

 concerned here with most of these objections. If we look at the 

 problem purely from the point of view of theoretical genetics, we 

 realize that the classic theory of the gene in itself, that is, without 

 additional assumptions, would limit possible genie permutations by 

 reshuffling of mutant loci within the confines of the species. The 

 maximum that could be accomplished would be the reproductive 

 isolation of one or the other permutation, which then might be called 

 a difiFerent species. However, this new species could only supply more 

 of the same permutations and certainly nothing really new. It was the 

 realization of this situation which made me renounce my former Neo- 

 Darwinian ideas, when the analysis of the adaptive subspecies of 



