Chromosomes and Genes 119 



other hand." To me it seems that the facts favor the assumption that 

 there is no difference in principle between the processes resulting in 

 so-called gene mutations and those resulting in rearrangements; the 

 differences observed are only in the size of the effect, that is, invisible 

 versus visible rearrangement. 



This conclusion must not be confounded with the ideas of Lea 

 (1947) and Herskowitz (1946) that gene mutations are accompanied 

 by a chromosome breakage nearby. If this is repaired, or healed, we 

 find only the gene mutation; if it unites with another break we have, 

 in their opinion, a rearrangement with the simulation of a position 

 effect, which actually does not exist. We shall see that the facts of 

 position effect completely rule out this hypothesis, which of course is 

 the opposite of the thesis we shall develop: both are position effects 

 of visible and invisible rearrangements. 



c. Position effect 



We come now to the second major group of facts which is 

 decisive for the understanding of the genie material within the chro- 

 mosomes. It is rather unfortunate that the historically justified term 

 "position effect" cannot be changed. It connotes a description of the 

 phenomenon exclusively in terms of a definite interpretation based 

 upon the classic gene. The term implies that the action of a gene 

 differs according to its position. The actual facts lend themselves to a 

 different and much more justified interpretation. The unprejudiced 

 description of the effect is the following. Chromosome breaks, result- 

 ing not only in rearrangements like translocations and inversions, but 

 also in deficiencies and duplications, very frequently have a pheno- 

 typic effect, which commonly is identical with that of the mutant 

 phenotype of a nearby locus. Cases in which no nearby locus is 

 known, the mutant effect of which is imitated by the rearrangement 

 break, suggest that such a locus is not known, because by chance 

 it never mutated, though it cannot be denied that an independent, 

 autonomous breakage effect may exist. The facts show only re- 

 arrangement breaks and phenotypic effects similar or identical with 

 a mutant allele at a nearby locus, though this locus was not affected 

 directly. An unprejudiced description of the actually observed facts 

 does not require a consideration of genes and their position. I think 

 it is important that the logic of this situation should be recognized 

 and that the problem be attacked without the theoretical prejudice 

 contained in the word "position." 



