132 PHYSICAL BASIS OF HEREDITY 



dence for his view demonstrate just the opposite, when 

 complications due to crossover variations are excluded, 

 by using only data in which three or more loci are recorded 

 simultaneously. In his attempt to explain the all-import- 

 ant fact of rarity of double crossovers, Castle is obliged 

 to assume that there is a difference in frequency of cross- 

 ing over in different planes (directions). This assump- 

 tion can be shown to be inconsistent with the primary 

 assumption that he accepts, viz., that crossing over is 

 proportional to the distance between genes. 



