26 
BULLETIN 1468. U. S. DEPARTMENT OF AGBICrLTUBF. 
would appear to be futile. With only 14 progenies, all closely- 
related, the rate of crossing over ranges from IT to 38. Individual 
plants with significantly lower and higher rates have appeared, and 
there seems little reason to doubt that progenies having still lower 
or higher rates could be obtained. With Mendelian ratios the inter- 
action of a definite number of genes makes possible an expected 
ratio from which departures may be calculated, but with the cross- 
over ratio there is no fixed point of departure. The best that 
can be expected is to determine the rate under constant environ- 
mental conditions and a particular genetic complex. Once deter- 
mined, there is no reason — in maize at least — for assuming the par- 
ticular complex of genetic and environmental factors to be typical. 
c? «? <r 
eo 
Fig. 2. — Inheritance of percentage of crossing over, showing isolation in the fifth 
generation of two strains having widely different mean crossover rates 
Stadler (16), working with three families, has shown that the 
mean rate of crossing over. between C and Wx is not alike in the 
three strains. From a comparison of the differences between ears 
borne on the same plant, he concludes that variations in degree of 
linkage are due at least in part to environmental factors. 
In stability, the rate of crossing over does not approach that of 
the proportions of the Mendelian characters, and there would seem 
to be no justification for assuming a definite fixed rate of crossing 
over between pairs of linked genes. 
The situation with respect to the C-Wx linkage is shown graphi- 
cally for 16 progenies in Figure 5 and the biometrical constants are 
given in Table 5 (p. 12). 
