No. 614] 



THE BLUE ANDALUSIAN 



Finally, blue-splashed cross-over eR er mated with or- 

 dinary blue would give an expectation of 1 black to 1 blue 

 to 2 blue-splashed instead of the ordinary expectation of 

 equal numbers of blues and blue-splashed, viz. : 

 eR er = blue-splashed cross-over X Er eR = ordinary 

 blue; 



F x eREr = h\ue, 



eR eR = blue- splashed, 

 erEr = black, 

 er eR = blue-splashed. 



The possible matings not indicated in the foregoing are 

 those which would produce the same phenotypic ratios as 

 if ordinary individuals (i. e., non-cross-overs) of the same 

 appearance as the cross-overs were used. Such matings 

 are naturally of no value for analysis. 



If it should later be shown that crossing-over does occur 

 as suggested above and there are two pairs of factors con- 

 cerned, there is the possibility of occasionally securing 

 ER gametes. This in turn would seem to make possible 

 the blue Andalusian breeder's long-time dream of pro- 

 ducing blues that 4 'breed true." With the appearance of 

 the double recessive gamete er another race of Andalu- 

 sian would apparently become possible, which, if the 

 factors assumed in this paper are correct, should be white 

 splashed with black instead of with blue. 



The second possible interpretation of the facts so far 

 established is that my postulated factors R and E occupy 

 identical loci on homologous chromosomes, neither being 

 recessive to the other in its phenotypic expression. For 

 the present at least any evidence that this is the correct 

 interpretation will be largely negative and come from con- 

 tinued failure to find cross-over individuals with regard 

 to R and E. If these cross-overs should not be found it 

 might at first appear that the interpretation of the case 

 of the blue Andalusian is in all probability exactly what 

 has been suggested from the first, namely, that blue is 

 a heterozygote intermediate between the parental types. 



