LINKAGE 



A to a as well as from a to A), recombinations 

 after correction should be fewer than before. 

 True crossover value will not be greater than 

 that indicated by ratio thus obtained, but it 

 may be considerably less. The value of the 

 method lies in the fact that it gives a maximum 

 which is lower than may be calculated from the 

 frequencies directly. 



If the percentage of single recombinations 

 is known for each of two regions, the probabil- 

 ity of two recombinations occurring simultane- 

 ously, one in each region, is the product of 

 the percentage of singles in each region. The- 

 oretically, the percentage of double recombina- 

 tions observed should equal the product of the 

 percentages of singles but actually, in most 

 cases, the observed double percentage is lower 

 than that expected because of interference. 

 Stevens (1936) has derived the following formu- 

 la for measuring interference in terms of coin- 

 cidence using the method of Maximum Likelihood: 



c = wn/(w + x)(w + y) 



where c equals coincidence, w equals number of 

 double recombinations, n equals total number in 

 cross, x equals region (1) recombinations, and 

 y equals region (2) recombinations. This for- 

 mula makes possible the estimation of coinci- 

 dence directly from the raw data. A coincidence 

 value of one (1) indicates no interference; a 

 value of zero (0) indicates absolute interfer- 

 ence. 



There is no significant change in recombina- 

 tion percentage because of aging of the female. 

 vCoincidence values also show no significant 



119 



