100 T. H. MORGAN AND ELETH CATTELL 



it is found in the same experiments that other factors not showing 

 sex-linked inheritance do not show associative inheritance with 

 those factors that are sex-hnked. It is here that the chromo- 

 somal hypothesis seems to give an insight into the nature of the 

 difference in the two cases. 



In. those counts where crossing-over in color is expected there 

 were 61 such cases in a total of 12115; a gametic ratio of 1 to 

 198. The crossing-over is such a rare occurrence that although 

 the total number of individuals is large, it is by no means large 

 enough to make the ratio significant. 



On the other hand, where the gametic ratio for crossing is low, 

 as in the wing-characters, the numbers suffice to make the values 

 significant, and it will be observed how closely these approximate 

 1 to 2. Moreover, the crossings balance quite well the counter- 

 crossings, which is expected on the theory of chromosomal inter- 

 changes. 



The same balance is expected for the color ratio but here we 

 find, as the following table shows, that 18 crossings were in one 

 direction and 43 in the other: 



RB Wb 



4 1 



in 260 



43 18 12115 



But these numbers are, as stated, too small to be significant. It 

 should be noted that no actual exchanges between the eggs them- 



