CROSSING OVER AND CHROMOSOMES 117 



were present in each stamen in the expected proportions. 

 These and other difficulties make it improbable that link- 

 age can be the result of this kind of reduplication. 



Bateson and Punnett formulated their hypothesis at 

 first for only two pairs of linked factors. When it was 

 shown that three pairs of factors could show linkage, 

 Bateson and Punnett assumed that all three pairs of fac- 

 tors might segregate at the same time (or in three suc- 

 cessive divisions), the observed ratios being due, as 

 before, to unequal division rates later. Trow has sug- 

 gested that in such cases the segregation and reduplica- 

 tion for the third pair of factors might not occur until 

 that for the first two pairs was completed. This view 

 seemed to meet certain inadequacies of the former hypoth- 

 esis, but meets with certain difficulties on its own account. 

 One of the most obvious of these objections is, as Sturte- 

 vant has pointed out, that the number of cell divisions, 

 necessary to produce some of the higher ratios that are 

 known, would produce a mass of cells thousands of times 

 larger than the animal itself. 



