220 



CHARLES ZELENY 



described by Tice in 1914, and that the change from bar to ultra- 

 bar has come only three times in 85,000 individuals, it is probable 

 that the absence of the jump from full to ultra-bar is merely a 

 matter of random samphng. 



3. The frequency of reverse mutations. The special feature of 

 the bar sei-ies is the frequency of reverse mutations (fig. 3). 

 Each of the three possible reverse jumps occurs with a frequency 

 much greater than that of the original mutations. The coeffi- 



\/ 



\/ 



< 



> 



-> 



^ red ■^ 

 -^ coral 



^ blood 



-^ cherry 

 eosln ^ 



-^ ivory 



-^ tinge 

 buff 



-^ ecru 



^ white '^ 



/\ 



^ ultra- 

 white? 



Fig. 4 The white-eye series of multiple allelomorphs arranged in order of 

 darkness of color according to Muller ('20). All the observed kinds of jumps 

 are indicated. 



cients of the reverse mutations are 0.00035, 0.00061 and 0.00058 

 as opposed to 0.00003, 0.00000 +, and no recorded mutations. 

 Since the original observation of reverse mutations by May they 

 have been noted in the white eye series of Drosophila where 

 several returns of white to eosin and one of white to red have been 

 noted. In none of these other cases do the reverse mutations 

 occur with the great frequency noted in the bar series. 



It is obvious that the presence and absence theory cannot apply 

 to these changes The closest analogy seems to be to chemical 

 changes involving a rearrangement of atoms. In the bar series 

 the rearrangement in one direction is more easily accomplished 



