88 Nature of the Genetic Material 



trance of the effect that varies from less than 1 per cent to 100 per 

 cent. Selection experiments (see 1951; and Goldschmidt, 1953Z?) show 

 a considerable similarity to the ones reported by Mather, though they 

 deal with penetrance (and partly correlated expressivity) of a single 

 though variable phenotype, not with a numerical variation of a char- 

 acter. Selection, as a rule, does not lead very far, though it shows 

 the not unusual availability of minor modifiers. However, selection 

 from less than 1 per cent to 100 per cent penetrance can be accom- 

 plished in a few major jumps. Some of these sudden jumps were 

 shown to be due to the introduction of major enhancers; others must 

 have been based upon the liberation of recessive enhancers by cross- 

 ing over from a balanced condition. But there is no reason whatever 

 to assume special balanced blocks of polygenes. Individual enhancers, 

 balanced by association with lethals, suffice. 



A number of special features distinguish the podoptera effect 

 from ordinary mutant effects. In their totality, these special features 

 suggest that we are dealing not with the ordinary type of mutants, 

 but possibly with mutated conditions, whatever that may mean, of 

 the intercalary heterochromatin. There is no reason to compare such 

 conditions of the heterochromatin with standard point mutants. I am 

 inclined to think of a difference in quantity, without any proof for 

 such an interpretation other than the comparison with the effect 

 of quantities of block heterochromatin. These distinctive features of 

 the pod effect are as follows. (1) Genetic factors for the pod effect are 

 ubiquitous, and can be found in any Drosophila line which has been 

 properly tested. (2) The effect varies down to such a low penetrance 

 and expressivity that there is practically no limit between a low effect 

 and normality. (3) Pod factors are known in all chromosomes and 

 have different effectiveness. Their exact localization is difficult except 

 for one or two major ones in the first and second chromosome. (4) 

 These factors control definite and different phenotypes within the 

 general pod effect in different selected lines. (5) Within each line 

 these multiple factors collaborate in an additive or multiplicative 

 total effect, the details being rather complicated. (6) The pod factors 

 of different lines do not replace each other, are not allelic. (7) Never- 

 theless, there is some interaction between these non-allelic factors, 

 as their simultaneous, heterozygous presence gives a little pod effect. 

 We may therefore speak of a weak false allelism (a "false" allelism 

 because the term "pseudoallelism" already has a different meaning), 

 which is not confined to factors in the same chromosome. (8) The 

 degree of dominance of some dominants is influenced by the presence 



