DIRECTION AND FREQUENCY OF MUTATION 223 



coefficients of mutation in unselected, high and low stocks are, 

 respectively, 0.00068, 0.00063, and 0.00069. There is thus no 

 evidence that the presence or absence of particular accessory 

 factors affects the frequency of mutation. 



The emarginate stock (fig. 1, E) is a very high stock derived 

 from bar by mutation in an accessory gene. As shown in table 

 2, it has an unusually low mutation rate to full. If an effect of an 

 accessory factor upon mutation at the bar locus is assumed in this 

 case, it is in the opposite direction from that usually postulated. 



8. Does direction of origin have an effect u-pon the direction or 

 frequency of mutation? There is a widely prevalent view that 

 biological states cannot be described in terms of present condition 

 alone, but that a knowledge of past history is necessary for the 

 prediction of their reactions. They are supposed to retain the 

 ear-marks of their past. Perhaps such views merely indicate a 

 general appreciation of the complexity of biological phenomena 

 and the feeling that for that reason there cannot be identity of 

 present structure and behavior when past histories are different. 

 This is especially true in the discussions of the probable course of 

 evolution and is one of the postulates of orthogenetic theories. 

 On the other hand, chemists and physicists are inclined to con- 

 sider a particular substance as being always the same, regardless 

 of its source, though an outsider may notice a present tendency 

 to scrutinize carefully supposedly identical substances derived 

 from different sources. 



The different components of the bar series are definite entities 

 comparable to definite chemical compounds or physical states. 

 This is indicated by the facts of their structure, their behavior 

 in heredity, and their tendency to mutate in the same manner 

 regardless of the character of their origin. Thus bar derived 

 from full is not different from bar derived from ultra-bar, and full 

 derived by reverse mutation from bar or ultra-bar is not different 

 from the original full. 



9. Is there a periodicity in the mutations? It is common belief 

 that mutations are periodic, occurring with great frequency dur- 

 ing a limited time and very rarel}^ or not at all at other times. 

 The studies of Drosophila as a whole have not yet demonstrated 



