No. 614] 



THE BLUE AN DAL USIAN 



107 



lusian <$ and a white Plymouth Eock % showed only rod- 

 shaped granules. Feathers from a blue-gray individual, 

 whose dam was a blue-splashed Andalusian and whose 

 sire was a crossbred, the offspring of a Houdan £ X 

 single-combed white Leghorn ? cross, showed only round 

 granules. 



If, as Goldschmidt assumes, his factors mQ and Mq are 

 so closely linked that they never separate, and behave 

 only as a single pair of factors, it is simpler to assume 

 that there is but one pair of factors. As already pointed 

 out, however, the discontinuity in the gradations from 

 blue-splashed to black is such as to lead one strongly to 

 suspect that two pairs of factors are at work. This dis- 

 continuity is greatly emphasized in the case of the blue 

 offspring from the white Wyandotte X blue-splashed An- 

 dalusian cross. It is perhaps not impossible that a single 

 pair of factors should bring about the result found in 

 Andalusians, but it is so unusual as to make the assump- 

 tion of two pairs of factors reasonable. 



If this assumption is correct it must be further as- 

 sumed, as Goldschmidt implies but does not state, that 

 the black and splashed races each contribute a dominant 

 and a recessive factor, and that in the blues we have the 

 expression of both dominants, namely, the extension of 

 pigment to all feathers, furnished by the black (or, in the 

 Wyandotte cross noted above, by the white) parent, and 

 the restriction of the pigment in the feather structure in 

 such a way that the effect is bluish-gray, furnished by the 

 blue-splashed parent. It is of interest in this connection 

 to note that the blue condition produced by the restriction 

 of the pigment in the barbule cells is recessive in pigeons 

 (Cole, 1914, p. 325), while in Andalusians, on the above 

 assumption, it is dominant. 



While exact data concerning the breeding behavior of 

 blue Andalusians are exceedingly meager, the experience of 

 breeders generally seems to be in accord with such data 

 as there are, and with the interpretation offered by Bate- 

 son and his associates. In order to account for the fail- 



