180 HAROLD H. PLOUGH 



drop are formed, but such a suggestion does not account for the 

 later behavior of the curve. 



Heat effect at one stage in the ovary only 



The data given in the two day interval series with regard to the 

 effect of heat on crossingover is happily very much more striking, 

 and extremely easy to interpret. Figure 5 gives the analysis of the 

 offspring of flies which were hatched from the same bottles as those 

 which made up the control series. The heat was applied immedi- 

 ately after mating and was continued for seven days. The flies were 

 kept in the same tubes throughout this period. The percentage 

 of crossingover indicated is that of the initial high point of the 

 control. The next count taken represents the total for the 

 period from the seventh to the ninth day, and the results are 

 9 and 5 units higher than the previous values respectively, and 

 12 and 9 units higher than the corresponding control values. 

 This high percentage is maintained for eight days at the end of 

 which the curves drop abruptly 14 and 7 units respectively, and 

 at the end of the seventeenth day show values even below the 

 corresponding ones for the control. This low level is then 

 maintained throughout the series. 



Figure 6 gives a similar picture as a result of slightly different 

 treatment. The Fi females of this series were hatched at 31.5°C. 

 but mated and kept at 22 °C. The percentages of crossingover 

 begin at a point 4 to 5 units above the control and this high 

 level is approximately maintained for eight days. The curve 

 then drops to the normal point and stays there throughout the 

 remainder of the series. (The exception in the ten to twelve 

 day value for the purple-curved region apparently is insignifi- 

 cant statistically, and especially so when viewed in contrast to 

 the corresponding black-purple value.) At the tenth day three 

 of the pairs were again subjected to a temperature of 31.5°C, and 

 the results are shown by the broken line. The same values given 

 by the untreated pairs are maintained for five days, at the end 

 of which the curve jumps a second time to the original heat- 

 induced point. 



