LINKAGE 389 



crossing over as shown in the diagram. Most of the F z flies show. the 

 same combinations as their grandparents, which indicates that no crossing 

 over has occurred in the majority of the gamete-forming cells. Since 

 there are many more F 2 individuals with YWL or GRM (second column) 

 than with YRL or GWM (third column), it is believed that the distance 

 between the genes W and M (and between R and L) in the original chromo- 

 some must be greater than that between Y and W (and between G and R), 

 so that there is more chance for crossing over to occur in the lower part 

 than in the upper. Thus if two linked characters are often separated 

 (i.e., have their linkage broken) their genes are thought to lie relatively 

 far apart in the chromosome, whereas linked characters separated only 

 rarely are supposed to have their genes located very near each other. 

 Recombinations involving a double or even more complex crossing over 

 (fourth column) would be expected very rarely. In this way the genes 

 for the various characters have been assigned to their loci in the chromo- 

 somes on the basis of the frequencies in which the various new combi- 

 nations in FZ appear. 



In Fig. 153 is shown the "map" of the chromosomes of Drosophila 

 as determined by Morgan and his associates, each factor being placed 

 a certain number of units of distance from the end. Many other known 

 genes are not shown in the diagram. As the unit of distance is taken 

 that space separating two factors whose linkage is broken (i.e., between 

 which crossing over occurs) in 1 per cent of the cases. Thus crossing 

 over occurs between "yellow" and "bifid" in 7.3 per cent of the observed 

 cases; the factors are therefore placed 7.3 units apart (first chromosome). 

 Furthermore, if some linkage relations are known, it is possible to calcu- 

 late certain other linkages in advance. For example, if it were known 

 that crossing over occurred between "sepia" and "pink" (third chromo- 

 some) in 20.2 per cent of the flies, and also that "sepia" and "kidney" 

 showed a 34.7 per cent crossing over, the prediction that "pink" and 

 "kidney" would show a 14.5 per cent crossing over (not including the 

 modifying effects of double crossing over, should this occur) would be 

 borne out by experimental results. Such an agreement of the results of 

 new crosses with predictions made on the basis of known linkages has 

 occurred over and over again in the experiments of Morgan and his 

 students. The chiasmatype hypothesis as thus elaborated obviously 

 fits the observed facts remarkably well. 



Interference. Another piece of evidence brought forward in support of 

 the hypothesis that the factors have a linear arrangement within the 

 chromosome is the phenomenon of interference, which has been elucidated 

 by Sturtevant (1913), Weinstein (1918), and particularly by Muller. If 

 the factors or genes are arranged in a series as supposed on the above 

 hypothesis, it would be expected that when crossing over occurs at a given 

 point in a pair of chromosomes, the regions immediately on either side of 



