232 CHARLES ZELENY 



IV. SUMMARY 



1. The bar-eye allelomorphic series of Drosophila melano- 

 gaster is espelcially favorable for studies of the nature of mutation. 



2. There are three definite stopping-places without inter- 

 mediates. 



3. All possible kinds of jumps between these stopping-places 

 have been observed except full to ultra-bar. 



4. Reverse mutations are more frequent than the original 

 direct ones. 



5. Recent origin is not an explanation of frequency of muta- 

 tion, because full derived by mutation from bar or ultra-bar does 

 not revert. 



6. The longer jumps are as frequent as the shorter ones. 



7. The direction of selection for facet number has no effect 

 upon the direction or frequency of mutation. 



8. The presence or absence of accessory factors causing high 

 or low facet number in bar stocks does not affect the direction 

 or frequency of mutation of bar. 



9. The direction of origin of bar has no effect. Bar stocks 

 derived by reverse mutation from ultra-bar show the same direc- 

 tion and rate of mutation as bar stocks derived directly from 

 full. 



10. There is no evidence of periodicity in the mutations. 



11. Mutations occur in the germ plasms of both males and 

 females. 



12. Mutations are not confined to a single period in the germ- 

 cell history. Some appear early and others late in oogenesis or 

 spermatogenesis. 



13. Frequency of observed mutations at the bar locus is greater 

 than in all accessory factors combined. 



14. The changes in the gene which produce the somatic series 

 full to bar to ultra-bar are probably not of a quantitative nature. 



15. The different components of the bar series" are definite 

 entities comparable to definite chemical compounds or physical 

 states. 



16. There is no difference in rate of mutation from bar to full 

 between stocks kept at 20° to 22°C. and those kept at 27°C. 



