186 Nature of the Genetic Material 



cations changing the chromosomal unit (e.g., stocks with a part of 

 the X attached to the 4th chromosome, if homozygous viable or 

 with pericentral inversions) might lead to unexpected results. The 

 work thus far done in Drosofjhila (reviewed in Patterson and Stone's 

 book) did not consider possibilities of the kind presented here and 

 was concerned only in the effects upon crossing over in pericentric 

 inversions, and in phylogenetic possibilities in the case of chromosomal 

 fusions and translocations. The only argument at present available is 

 that such fusions would hardly have persisted in phylogeny if they 

 did not produce some new genetic effect. 



D. CONCLUSIONS: THE MODERN THEORY OF THE GENE 



The foregoing discussion contains all the conclusions which, in 

 my opinion, should be drawn from recent developments in genetics 

 regarding the nature of the genetic material. Since the presentation of 

 numerous details obscures the simple sequence of ideas, the argu- 

 ment will be worked out once more, together with a few additional 

 facts and deliberations. The oldest idea concerning the gene and its 

 mutation is Bateson's presence-absence theory, meaning that a 

 mutant is the absence of a gene. After much discussion the theory 

 was abandoned, mainly because of the discovery of the ubiquity of the 

 existence of multiple alleles (also because of the alleged frequency of 

 reverse mutations). Whatever these were, the existence of different 

 grades of absence could not at that time be defended. This is a valid 

 argument for the classical theory of the gene if one accepts the idea, 

 for example, that the gene is a single molecule. 



But there is also the possibility of making different assumptions 

 about the nature of the gene which would still allow the working of 

 the presence-absence theory. If the gene should consist (as I formerly 

 supposed) of a definite number of molecules, the loss of one or more 

 could be the mutation process. It might then still be claimed that a 

 mutant is a deficiency — either a total or a partial one, the latter 

 assuming a divisible gene or a gene with some substructure — and this 

 has actually been done by Stadler (see 1933). The facts which have 

 been discussed above put the old Batesonian idea in a different light. 

 There is no longer a gene molecule but a definite molecular pattern in 

 a definite section of a chromosome, and any change of pattern 

 (position effect in the widest sense) changes the action of the chro- 

 mosomal part and thus appears as a mutant. Multiple alleles are no 

 longer different conditions of a single gene but changes in pattern 

 within small sections of a chromosome which are not necessarily of 



