72 THE CHROMOSOMES 



near the spindle attachment could be * mapped ' 

 genetically ; the others would show complete linkage 

 and no crossing-oVer (Fig. 17). Unfortunately no 

 genetical work has been done on these organisms 

 with strict localization of chiasmata. 



It is a fact which has long been known in genetics 

 that the occurrence of a cross-over at a particular 

 locus prevents crossing- over for a certain distance on 

 either side. This phenomenon is known as inter- 

 ference. Cj^ologically it means that there must be 

 a certain minimum distance between chiasmata. 

 If, as previously suggested, crossing-over results 

 from a torsional strain of some kind it is clear that 

 this strain will be relieved in the adjoining region 

 once crossing-over has taken place. That is to say, 

 a chiasma diminishes the probability of another 

 being formed in its immediate neighbourhood to 

 zero and this ' interference ' falls off until at a certain 

 distance from the first chiasma it is no longer present. 

 In many small bivalents the total length of the 

 chromosome is shorter than the region of complete 

 interference so that only one chiasma is formed. 

 HaldanE ^1 has proved the existence of interference 

 as a cytological phenomenon by a statistical analysis 

 of chiasma-frequencies. 



If one considers two adjacent chiasmata in a 

 bivalent it is clear that the relationship between 

 them may be of three different kinds : (1) the second 

 chiasma may involve the same two chromatids as the 

 first one, (2) it may involve one of those which crossed 

 over in the first chiasma and one that did not, or 

 (3) it may involve the two that did not cross over 

 in the first chiasma. These three types of relation- 

 ship are called reciprocal, diagonal and comple- 

 mentary. ^^^ If the second chiasma arose between 

 the four chromatids at random then the three types 

 would occur in the ratio 1:2:1. The reciprocal 

 and complementary types of relationship cannot, 

 unfortunately, be distinguished from one another 



