GENETICS PURPLE EYE COLOR DROSOPHILA 299 



crossovers of a double no longer included the same average 

 length of chromosome, but were including a greater length. The 



S + + + 



coincidences of the case support this view, but in 



+ pr c sp 



neither case are the data conclusive. A calculation of the coin- 

 cidences shown by Plough's black purple curved two-day tube 

 cultures has provided data considerably more satisfactory, 

 though still subject to a high probable error. On the basis of all 

 the data, it is probable not only that coincidence varies with age, 

 but that the curve of age variation in coincidence is roughly 

 the mirror image of the curve of age variation in crossing over. 

 While it seems probable that at least part of the explanation of 

 the age variation both in crossing over and in coincidence has 

 been found in an internode variation as suggested, yet in any 

 case there is provided evidence of a common cause that should 

 repay further analysis. 



A second problem involving purple and very closely allied to 

 the age variation in conception, material, methods, and bearing, 

 is that of the temperature variation described by Plough ('17). 

 Since the genie constitution of a female showing the age variation 

 is constant throughout the course of this variation, the imme- 

 diate causes of the variations must be regarded as environmental 

 differences arising through rhythmic changes in the physio- 

 logical processes of nutrition and development. While the 

 crossing-over variations due to age and to specific genes were 

 affected through environmental changes arising internally, they 

 suggested the possibility that similar variations might be 

 initiated by environmental changes arising externally. Plough 

 found that exposure to abnormally high or low temperature 

 actually did produce linkage changes even more extreme than 

 those due to age changes. Black purple curved back-cross 

 cultures were raised at various temperatures from 9° to 32°C. 

 When the black purple crossover values were plotted, it was 

 seen that at a low temperature (9°), crossing over is very free 

 (14 per cent) and becomes even more free at 13° (18 per cent). 

 The amount of crossing over then falls away very rapidly, and 

 at 18° has nearly returned to the normal value. The normal 



