4. Backcross data involving chromosome 7. Of the three cultures included in the three-point test, the first was grown in the greenhouse in the winter of 1938-39, the second in the garden in the summer of 1939, and the third in the greenhouse in 1939-40.

The marked difference between complementary classes of region 1 and double crossovers are not to be accounted for by differential viability of recessives; for, of the total, in plants constitute 48.4%, v5 plants 48.8%, and gl plants 50.1%. A comparison of frequencies of double recessives with those of corresponding double dominants shows that the one double recessive, in v5, is principally responsible for the differences between complementary classes. The frequency relations of double recessives to corresponding double dominants are as follows:

In view of the approximate equality of V5 and v5 plants in this back-cross progeny, it is hard to account for the deficiency of in v5 plants either on the basis of errors in classifying or a suppressing effect of in upon the expression of v5, like that of R upon j. A further study will be made of this second possibility.

A two-point back-cross gave the following:

The order of these genes is:

in  6  v5  14  gl  Tp

A. C. Fraser