OF MUTANT CHARACTERS. 137 



shorter than usual (Morgan, 1911). This fly when bred to his wild-type 

 sisters produced about 10 per cent of offspring whose wings showed a 

 slight or moderate amount of truncation (plate 6, figs. 4 and 5). Some 

 of these truncated flies bred together produced in Fo nearly 50 per 

 cent of truncated. For several months it was impossible by selection 

 to raise the percentage of truncates much above 50 per cent, though 

 there was an increase in the shortness of the wings of the truncates 

 that did occur. But later certain cultures gave much higher percentages, 

 and selection started at this point and continued for about a year estab- 

 Hshed a stock in which the percentage of truncates was not far from 90 

 per cent. Above this level it was impossible to maintain gains. 



Along with this increase in the percentage of truncate individuals, 

 several other changes were observed to be going on. There was an 

 increase in the number of flies which were sterile and gave no off'spring ; 

 at times about 50 per cent of the pairs were sterile. There was a 

 marked decrease in the productivity of those pairs which did give 

 ofifspring. These two facts made the stock very difficult to carry on, 

 except in large mass cultures. In those cultures in which the per- 

 centage of truncates was high, the amount of the truncation was great ; 

 i. e., many of the flies had extra-short wings, some wings being even 

 shorter than the abdomen (plate 6, figs. 1 and 2). It was found that 

 if these short-winged flies were used in carrying on the stock, the per- 

 centage of truncates was higher. The flies whose wings were sHghtly 

 truncated gave under 90 per cent of truncated, while those flies whose 

 wings showed no truncation either gave no truncates or less than or 

 about 50 per cent of truncates. In all of these cultures more of the 

 females than of the males showed the truncate character; that is, 

 truncate is ''partially sex-Umited" in its expression. 



At this point Mr. E. Altenburg took up the selection of truncate and 

 confirmed the points just stated. Later, Altenburg and Muller (Mech- 

 anism of Mendelian Inheritances; also mss.) sought to analyze the 

 truncate stock by the method of linkage to find the cause of the con- 

 tinued appearance of normal flies and of the variations in the extent of 

 the truncation. Their tests showed that truncate is primarily due to 

 a dominant gene in the second chromosome. This gene is lethal when 

 homozygous, so that pure-breeding stock can not be obtained. When 

 heterozygous, this gene is capable of producing only a moderate trun- 

 cation and fluctuates in expression, so that most of the flies ha\ang the 

 truncate gene fail to show the character at all. But in the selected 

 stock there are certainly two (one in the first and one in the third) and 

 probably more genes whose effect is to increase the amount of trunca- 

 tion, whereby both a greater proportion can be detected and the wings 

 of those which do show truncation are shortened. 



Such intensifiers could theoretically only raise the amount of trunca- 

 tion to the extent that every fly carrying the truncate gene can be 



