166 



Linkage and Crossing Over 



What is the result of applying this method to Brink's data? 

 Let us calculate the deviation and standard error. 



Rg A rg a Rg a rg A 

 160 142 103 115 



3 02.218 

 520 



= V 126.61 = 11.25 



42 

 11.25 



= 3.73 



When we test this ratio with the hypothesis of independent 

 assortment, we find that the deviation is 3.73 times the standard 

 error. Since any deviation greater than twice the standard error 

 makes it improbable that the observed ratio is an example of 

 the expected, we are forced to conclude that there is no inde- 

 pendent assortment here, which is in accord with our results 

 from the x^ method. It would also be possible to use this method 

 with F2 data instead of testcross data, but the calculation of 

 the standard error becomes more complicated. As the testcross 

 is much more widely used, the method from F2 data need not 

 be given. 



Different Crossover Values 



It has been shown that the percentage of recombinations be- 

 tween two linked genes is the same whether the original cross 

 was in the coupling or in the repulsion phase. From the crosses 

 cited in Drosophila, maize, the evening primrose, and the sweet 

 pea, we see that the percentage of crossovers is not the same for 

 all pairs of linked genes. If we test several genes located on the 

 third chromosome of maize, we find that the crossover percent- 

 age between the genes for anthocyanin-1 and ragged leaves is 

 42; that between anthocyanin-1 and crinkly leaf-1 is as high as 

 48 ; but the percentage of recombinations between anthocyanin-1 

 and nana-1 is only 28. Therefore, we find that the percentage 

 of crossing over may be very different for different pairs of genes. 

 However, for any two given linked genes, the crossover percent- 

 age is always constant, provided that the plants or animals are 

 grown under the same conditions. 



In maize, the crossover percentage between a and cr will 

 always be 48 and, for a and na, it will always be 28 no matter 

 how often the experiment is carried out, provided the condi- 

 tions are the same in each experiment. The crossover per- 



