XII THE CHROMOSOME THEORY 137 



parallel between the behaviour of the chromosomes 

 and the segregation of factors led to the suggestion 

 that the chromosomes might themselves be the 

 factors. It was not long, however, before the view 

 was seen to be too crude. As the work went on, it 

 became apparent that the number of factors exhibit- 

 ing Mendelian segregation in a species may be con- 

 siderably greater than the number of chromosomes. 

 On the chromosome theory, therefore, some, if not 

 all, of the chromosomes must carry more than one 

 factor. This introduces a difficulty. Let us suppose 

 that in the imaginary case we have just been dis- 

 cussing, the long maternal chromosome carries two 

 factors A and C, while the short one carries B. 

 Since the chromosomes must be supposed to preserve 

 their individuality, and of this there is independent 

 evidence from other sources, it is clear that A and 

 C must always remain completely linked together. 

 If we find that they can be separated, we must either 

 give up the chromosome theory altogether, or find 

 some acceptable explanation. 



It is to meet this difficulty that the American 

 workers have devised a very ingenious hypothesis. 

 In Drosophila there is evidence for over 100 different 

 factors, while the number of chromosomes in the 

 gamete is only 4 (cf. Fig. 30, p. 104). Large 

 numbers of experiments, involving the breeding of 

 hundreds of thousands of flies, have shown that the 

 numerous factors already demonstrated in Drosophila 

 fall into four groups. The members of any one of 

 these groups exhibit towards one another the phe- 

 nomena of coupling and i-epulsion, or, as it is 

 generally termed in America, linkagre. Between 



