The Bearing of Cytological Research on Heredity. 349 



them of certain gens, as shown in fig. 6. A simple and elegant solution of 

 the problem of recombination or of " crossing over " is thus given. 



It should be pointed out that Janssens' actual observations on the 

 chiasmatype still lack adequate confirmation, that the strepsinema has been 

 observed in the longitudinally split single chromosomes of the somatic 

 divisions, and that nearly all cytologists have hitherto believed the twisted 

 threads to untwist again before actual separation takes place. I have not 

 yet been able fully to satisfy myself concerning the facts ; but there is no 

 doubt that in the Amphibia many of the appearances seem to be in favour of 

 Janssens' conclusions. 



The most ingenious part of the explanation relates to the varying strength 

 of linkage. It is obvious that if the twisting be not too close, the likelihood 

 of a chiasma taking place in the interval between any two units (and hence of 

 their separation or dissolution of linkage) increases with the distance between 

 them in the thread. Conversely, the nearer together two units lie the greater 



M h 



(l) 



M 



li 

 I 



X- - ch ----- 



yi-ch. 



li 

 I 



_Ch — 



J 



K i 



...Ay— 



h H 

 i I 



...xh 



k K 



M 

 N 



etc 



(2) 



(3) 



(4) 



(5) 



Fig. 6. — Diagram of the exchange of factors (dissolution of linkage) as explained by the 

 chiasmatype hypothesis. The second chromosome-pair of fig. 5 is here employed as 

 an example. The original condition is assumed to be that shown above (1). The 

 lower figures (2-5) show a few of the many possibilities of exchange or " crossing 

 over " between'the two homologous linkage-groups or chromosomes, H-N and h-n. In 

 each case the point of crossing, fusion, and subsequent splitting is shown at the left, 

 with the result at the light. The position of the chiasma indicated in each case 

 by ch. In (2) and (3) but one chiasma is present, in (4) two, and in (5) three. A 

 very large number of recombinations is possible, even with so small a number of 

 factors as is here represented. 



