2l8 T. H. MORGAN. 



Summary .--The FZ generation was inte mediate n number of 

 tail feathers, feathers on legs, and color, although in regard to 

 the last point the bi ds were practically 1 ke the white parent, 

 a few feathers alone placing hem 'n the intermediate class. 

 The crest \vu> dominant. The F 2 generation showed segregation 

 practically complete for the feathers on the legs, it showed segre- 

 gation of white color to some extent, the recessive type appearing 

 only to a very limited degree. The number of tail feathers 

 showed a return to the normal, but the number of individuals is 

 too small to lay much emphasis on the point. Yet it is significant 

 that in all other directions the F 2 birds sufficed to show the in- 

 fluence of both parents. 



CROSS BETWKKN A TURBIT AND A STARLING. 



The FI hybrids of those birds were given to me by Mr. E. B. 

 Southwick, who kept both parent stocks. The turbits (Fig. 4) 

 were not very high-bred strains, but showed the main character- 

 istics of their class. Their color is red or blue. The head is 

 uniformly colored and has a cres : the- wing bars are dark; along 

 the middle line of the breast there is a series of reversed feathers 

 (Fig. 4). The starling (Fig. 5) was a black bird with a white top 

 to the head and white wing bars. On the breast there is a large 

 light crescent with a, metallic tinge. There are no reversed 

 feathers on the breast. 



The two hybrids (Fi) were red birds (Fig. 6). The crest was 

 well developed. The top of the head was white (like that of the 

 starling), but in one case not so sharply delimited. The wing 

 bars were dark. 



From these two hybrids eight, Fj birds were reared. Five 

 have crest and two lack the crest, five are red and three blue, 

 all but one have darker bars on wing. T is one has a dark bar 

 on most of the feather^ of the wings but a fe\v feathers ha\ e \\ hii< 

 bars. 



Summitry. The most interesting result of this combination is 

 the failure of the reversed feathers on the breast to reappear in 

 the second generation. The result i> like that of the number of 

 tail feathers in the fantail-s\vallo\\ cross, but here also the 

 numbers are too small to make any conclusions possible, however 



