188 



Bulletin No. 155.— 1913. 



regarded as homozygous for /. If the W. L. Bantam in question 

 possessed the constitution C^Bhf^Ii, one would expect in Fi equal 

 numbers of dark and light birds; and each sort would be equally 

 divided between the sexes. Of the dark birds, half should be barred 

 and half non-barred. The actual and the expected results can be 

 represented as follow^s: 



It is clear that, on this interpretation, there is close correspondence 

 between the actual and expected in Davenport's first instance, second 

 series {I. c, p. 37), in which the cT" parent was the W. L. Bantam 



Instance 2. — In this case, Davenport's results are explainable on 

 the hypothesis made for Instance 1. If the constitution of the 

 W. L. Bantam were CiBhf^Ii, we should have the following ratios: 



This explanation is also in accord with Davenport's results in 

 mating the same W. L. Bantam d^ with W. L. 99 (Davenport's 

 Nos. 127 and 128). 



Instance 3. — The results in Fi are in full accord with the present 

 hypothesis. Apparently the W. L. cf was homozygous for /. 



