84 " gi:fit bonniek 



use another mak' number. If there lias been only one lethal present 

 tlie number of males is multiplied by 2, in most of the cultures, 

 howevi'r, two lethals (// and /;•) at the distance of 8,9 units have been 

 present and in these cases the numljer of males is multi])lied by 2,2. 

 (For the derivation of the number 2,2 see Bonnier 1922). In other cul- 

 tures, finally, there have been lethals in both the Ä'-chromosomes (cf. 

 Bonnier 1923) and in these cases there has been chosen as male number 

 95 % of the number of females since the ])roportion in Drosophila 

 between males and females according to Wahren (1918) ordinarily is 

 95:100. In all doubtful cases I likewise have taken 95 % of the 

 females as the male-number. 



In order to avoid errors due to the influence of the environment 

 the rule has been to count only those cultures in which at least 40 

 females have hatched. But as a number of cultures has been made 

 up under rather unfavorable culture-conditions it has been necessary 

 in a few instances in order to have any result at all to consider also 

 cultures with a lower output. Finally concerning the tables I have by 

 the side of the number of regular males tabulated the theoretical num- 

 ber used for the calculation of the percentages and in order to make 

 it possible for the reader to judge the theoretical value of the experi- 

 ments I have stated for every culture the phenotype of the parents and 

 from which culture they are descendents. 



I am much indebted to Dr. M. E. Collett, who has been kind 

 enough to correct the English text and to Dr. G. Söderberg for her kind- 

 ness in drawing the figures. 



II. THE EXPERIMENTS. 



Let us now begin by enumerating the different a priori causes 

 which are not due to the environment but which may influence the 

 percentage of exceptions. 



1. There may be different kinds of Y-chromosomes (as mentioned 

 by Bridges). 



2. There may be a gene or genes in the autosomes. 



3. There may be a gene or genes in the Ä'-chromosome. 



4. There may be genes both in the autosomes and in the A'. 



5. There may be a cytoplasmic »factor» inherited only from the 

 females. 



6. There may be still other causes. 



From the trial mentioned in the introduction it was infered that 



