94 PHYSICAL BASIS OF HEREDITY 



is not in evidence unless a larger number of genes than 

 two is studied in the same experiment. These questions 

 will be discussed in Chapter IX. 



For the purpose of clearer exposition I spoke of link- 

 age, in the preceding chapter, as though the term should 

 be limited to cases where all the genes of a group hold 

 together, and have used the tenn crossing over to mean 

 the breaking of the group in one or more pairs. As a 

 matter of fact, it is not desirable to emphasize this sharp 

 distinction. There is, however, a real distinction that lies 

 behind the phenomena. In the male of DrosopMla there 

 is no crossing over at all between homologous chromosome 

 groups, while in the female there is crossing over between 

 the pairs of chromosomes. The cases of the male and 

 female are, therefore, on a different footing here. 



We speak of pairs of characters as being loosely linked, 

 meaning that crossing over of genes frequently takes 

 place, and as strongly linked when crossing over is very 

 infrequent. We have seen that yellow and white break 

 apart only once in 100 times. If characters (or the genes) 

 were still more closely linked, they might break only once 

 in a thousand times, and if closer still once in many 

 thousand times, in which case they would appear to be com- 

 pletely linked for all practical purposes. Such a grada- 

 tion, however, does not appear to be the case, but the lower 

 limit of crossing over seems to be well within the range of 

 human capacity to detect. This means probably that 

 there is a limiting value for crossing over, and if this 

 can be established it may give us the lower limit of size 

 of the gene (in terms of chromosome length), or at least 

 it may allow us to form some idea as to how many genes 

 are present in the hereditary material. 



In this same connection it has been suggested that 

 when more than two allelomorphs occur, we may be deal- 

 ing only with close or even absolute linkage. For instance, 

 suppose in a white-eyed race of flies a mutation should 

 take place in a gene so closely tied up in some way to the 



m 



