174 L- J- COLE AND W. A. LIPPINCOTT. 



of male plumage with respect to shape and structure of feathers, 

 the soma was still heterozygous for the barring factor and con- 

 sequently the white bars in the "cock" feathers remained narrow 

 as in the normal female. Davenport (1912, p. 17) classified 

 sex-dimorphic characters into two classes: " (a) characters whose 

 development is controlled primarily by determiners located in 

 the sex-chromosomes and, (b) characters whose development is 

 specially influenced or modified, probably by secretions of the 

 sex-glands." In "Kansas 1050" we have both these classes 

 represented in the same feather. 



In the hackle and saddle feathers of the male the bars are 

 less straight and regular than in the corresponding feathers of 

 the female or in other parts of the plumage of either sex. In 

 general there is a lagging of the middle portion of the bar, at the 

 shaft, producing a more or less V-shaped bar instead of one 

 straight across. This is illustrated m Figs. 9 and 10, and Figs. 

 13 and 14 show that the corresponding feathers on "Kansas 

 1050" had the same character. Thus it would seem that shape 

 of bar, unlike width, is a secondary sexual character rather than 

 sex-linked. 



But while our case conforms to the interpretation given above, 

 Pearl and Surface (1910, p. 53) state that in the cross between 

 Barred Plymouth Rock and Cornish Indian Game "there is the 

 same sort of sexual dimorphism in color pattern to be observed 

 in the hybrids as in the pure Barred Plymouth Rocks," and 

 specifically that "the male hybrids have narrower dark bars 

 on the feathers especially in hackle and saddle feathers than do 

 the females." These cross-bred males must of course have been 

 heterozygous for barring, so that on Spillman's hypothesis they 

 should have had the same width of bars as the hens. The fact 

 that the crossbred males were darker than pure Barred Rock 

 males does not alter this point, so that the matter cannot be 

 considered as settled. It is of course possible that there were 

 genetic factors carried by the non-barred breed (Cornish Indian 

 Game) which might have had some modifying effect on the 

 barring of the crossbred offspring. 



"Kansas 1050" was killed and autopsied on July 28, 1917. 

 Her behavior subsequent to the operation had not been notice- 



