150 



HAROLD H. PLOUGH 



X- 



15.25 



series composed of sibs made up at exactly the same time, and 

 under environmental conditions which were not sufficiently 

 different to cause a change in crossingover, we get the result 

 shown in table 4. The X 2 value is here 3.1 and the probability 

 of the two values being the same is 38 in 100, which is well 

 within the probable error. It seems probable, therefore, that 

 some slight difference in the environment must have had some 

 effect on the crossingover percentages. It was in order to test 

 whether the cause of such constantly recurring discrepancies 

 was to be found in slight paralleling changes in the environment 

 that the following research was undertaken. 



II. MUTANT STOCKS AND METHODS 



Before describing' the methods used for testing the effect of 

 variations in the environment on linkage, a word is necessary 

 about the mutant stocks used in the experiments. In all my 

 earlier work the triple recessive stock called black-purple-curved 

 ■ — b pr c — was used. This stock has black body color, purple 

 eyes, and curved wings. The percentages of crossingover given 

 in the published data for these factors are approximately: 

 black-purple, 5.5 per cent, and purple-curved, 19.0 per cent. 

 These percentages are small enough so that double crossingover 

 within each is relatively rare, a fact which makes it certain 

 that any deviation in the amount of crossingover will be detected. 

 In addition, the characters are very easy to determine, so that 

 the possibility of mistakes in counting is very slight. Later 

 star — S' — a dominant character involving a derangement of 

 the ommatidia of the eye; vestigial wings — vg — a recessive 

 character; and speck — sp — a recessive showing a black spot at 



