DIRECTION AND FREQUENCY OF MUTATION 231 



IJf.. Is the rate of mutatio7i affected by the ternperature? Muller 

 and Altenburg ('19), in a study of the frequency of appearance of 

 lethal factors in the cap chromosome of Drosophila, raised some of 

 their famihes at about 66°F. (18.9°C.) and others at about 80°F. 

 (26.7°C.). The 445 famihes raised at the lower temperature 

 produced five lethals, or one in ninety, and the 517 raised at the 

 higher temperature gave thirteen lethals, or one in forty. This 

 gives a positive value of two to three for Qio. 



The present data on mutation of bar to full furnish some 

 evidence on the temperature effect. The two selection lines 

 were kept at 27°C. All the other stocks were kept at 20° to 22°, 

 except that the temperature dropped somewhat lower on 

 a few occasions when the flies were kept in a cooled chamber 

 during the summer. The difference between the temperature 

 of the two sets is about 6°C. At the lower temperature there 

 were forty-four mutations from bar to full in 70,681 individuals, 

 or a mutation coefficient of 0.00062. At the higher temperature 

 there were eight similar mutations in 14,327 individuals, or a 

 mutation coefficient of 0.00056. The difference between the 

 two cannot be considered as significant and is not even in the 

 same direction as that obtained by Muller and Altenburg. The 

 present data are subject to the criticism that the comparison is 

 between stocks not undergoing selection on the one hand and 

 those in process of selection on the other. The objection, how- 

 ever, is not a valid one because high and low selection lines during 

 the process of selection, high and low stocks derived from these 

 lines, and unselected stocks, all have approximately the same 

 mutation coefficients. It has seemed desirable, however, to 

 test the temperature relations to rate of mutation more fully 

 and a carefully controlled series of experunents is under way. 



