438 



G. W. DELUZ HAMLETT. 



relations are shown in Fig. 2, a, which shows the ratios of the 

 values obtained in these experiments to the standard values of 

 each of the major regions studied. 



At the same time that these crosses were being made, the 

 linkage in flies heterozygous for the abnormality was being tested. 

 The same stocks were used in making up the crosses, and the 

 characters used were introduced into the cross in as near the 

 same way as possible. Both the homozygous and heterozygous 

 lines were kept under the same conditions, so that any possible 

 variations in viability would affect both lines alike and allow us 

 to draw valid comparisons between them. As will be seen in 

 Tables III. and IV., the results were almost identical with those 



of the first tests. 



TABLE III. 



A H e Tr 



PI ru h st p ss 



9 X ru h st p ss e c?. 



Regional Crossover Frequencies. 



TABLE IV. 



PI 



ru h st p ss Tr 



p ss e ro 



9 X p ss e ro 



Crossovers. 



Total Flies. 



In Table V. are results for a test involving the ebony-rough 

 and rough-claret distances. Only those flies which did not show 

 <K-lta hairless could be used in computing crossover values in 

 this experiment. This cross also gave data on the exact location 

 of translocation, as will be discussed later. 



