LINKAGE 395 



process as originally proposed. 1 The splitting of chromatic threads is, 

 moreover, a process already known to occur in all other mitoses. But 

 interpretations involving the actual fusion of the conjugating threads 

 have been adversely criticized with much effect by Gre"goire (1910), and 

 only further research on the cell can lead us to an adequate evaluation of 

 the above outlined suggestion. 



Other Theories of Linkage. The chiasmatype hypothesis is not the 

 only one which has been advanced to account for the phenomenon of 

 linkage. Most prominent among other attempts to solve this problem is 

 that of the English geneticists, especially Bateson, Punnett, and Trow, 

 who have advanced what is known as the Reduplication Hypothesis. In- 

 stead of accounting for the new factor combinations manifested by a cer- 

 tain percentage of the gametes on the basis of an interchange of factors in 

 the chromosomes at the time of reduction, these investigators seek to 

 explain them by postulating a series of differential divisions in the earlier 

 cells of the germ cell lineage, whereby Mendelian factors, not carried by 

 chromosomes, are segregated in such a manner that the observed types of 

 gametes are produced. Furthermore, the various factors are supposed to 

 be segregated successively, and at such stages in the cell lineage that the 

 proliferation or reduplication of the cells with new combinations of factors 

 shall account for the ratios in which the new types appear. Although 

 the differentiation of the somatic and early germ cells is accompanied by 

 visible differences in the constitution of their cytoplasm (see p. 406), 

 there is at hand no cytological evidence for such a segregation of heredi- 

 tary units as is thought to occur by the proponents of the Reduplication 

 Hypothesis. So long as this is the case, discussion of the hypothesis, 

 together with the subhypotheses formulated to meet certain serious objec- 

 tions, hardly belongs to cytology. One fact, however, pointed out by 

 Morgan (1919a) as very significant in this connection, is found in the 

 results of some experiments by Plough (1917). Plough investigated the 

 effects of temperature on the frequency of linkage breaking (crossing over) 

 in Drosophila. Not only did he find that temperature does affect the 

 amount of crossing over, but the effect was clearly produced at the time 

 of maturation and not earlier. This evidence is directly opposed to the 

 view that the new factor combinations are formed during cell-divisions 

 some time prior to reduction. 



Another effort to account for the results of crossing over without 

 resorting to the chiasmatype process is represented in a suggestion 

 (Goldschmidt 19176) to the effect that the two factors of an allelomorphic 

 pair are held to their places in the two homologous chromosomes by a 

 pair of variable forces, which allow them to exchange places in a certain 



1 A form of this interpretation of synapsis has suggested itself also to Dr. 

 C. W. Metz and Dr. E. G. Anderson, who inform the author that such a hypothesis 

 will in all probability conform satisfactorily to the data of genetics. 



