22 THE THEORY OF THE GENE 



near the middle of the series (between vermilion and 

 garnet), the result would be that shown in Fig. 16. Two 

 complete halves have interchanged. 



In other cases, crossing-over may take place near one 

 end (for example, between echinus and cross-veinless). 

 The result is like that shown in Fig. 17. Only the short 

 ends of the two series have interchanged. The same 

 kind of process occurs whenever an interchange takes 

 place. Whole series of genes are interchanged, although 

 as a rule the interchange is noticed only between the 

 genes on each side of the crossing-over. 



When simultaneous crossing-over occurs at two levels 

 at the same time (Fig. 18) very many genes are also in- 

 volved. For example, in the series just given one cross- 

 ing-over is supposed to take place between cut and tan, 

 and another crossing-over between garnet and forked. 

 All the genes in the middle of the two series have been 

 interchanged. This would pass unobserved were there no 

 mutant genes in the region to indicate the fact that two 

 crossings-over had taken place, since the two ends of both 

 series remain the same as before. 



The Linear Order of the Genes. 



It is self-evident that if two pairs of genes should be 

 near together, the chance that crossing-over occurs be- 

 tween them is smaller than if they are further apart. If 

 two other genes are still further apart the chance of cross- 

 ing-over is correspondingly increased. We may utilize 

 these relations to obtain information as to the ' ' distance ' ' 

 at which any tw^o pairs of elements lie with respect to 

 each other. With this information w^e can construct charts 

 of the series of elements in each of the linkage groups. 

 This has been done for all the linkage groups of Droso- 

 phila. Such a chart (Fig. 19) gives the result as far as 

 carried out. 



