340 CALVIN B. BRIDGES 



cross of eosin by wild. That is, none of the F2 females, all of 

 which were not-eosin, showed a trace of dilution, and likewise all 

 of the not-eosin males were perfectly normal in color. About 

 one-quarter of the eosin males were modified in eye color and 

 showed the cream color of the grandfather. It is evident that 

 the assumption was correct that the character cream is a double 

 recessive, the product of the action of a recessive autosomal gene 

 added to the effect of the sex-linked gene eosin. However, this 

 gene does not by itself produce any visible effect, as is proved by 

 the fact that a quarter of the wild-type flies must have been 

 homozygous for it (not eosin creams*) and yet all were of normal 

 undiluted color. The case of cream a was the first in Drosophila 

 for which such a relationship was shown, and the class of genes 

 typified by cream a has been called specific modifiers. Such 

 specific modifiers have since been found to be very frequent and 

 underlay many of the difficult early cases which were amenable 

 to selection. 



The cream a gene is probably autosomal rather than sex-linked, 

 as shown by the fewness of the creams; if the cream gene were 

 sex-linked, then, because of the linkage between eosin and cream, 

 from a quarter to nearly all of the eosin F2 males should have been 

 modified to cream, the frequency depending on whether the locus 

 of the cream gene were far from or close to that of eosin. 



^ It is our custom to name mutants on the basis of the change they produce 

 from the type of the wild fly, but since in the case of cream the stock homozygous 

 for the gene is indistinguishable from the wild stock, the name was given from the 

 most striking characteristic, namely, the power to dilute eosin to 'cream.' This 

 mutant was not called cream a (symbol Cra) until after the appearance of a second 

 cream whose gene was found to be in the second chromosome. This new cream 

 was then called cream II (the Roman II designating the second chromosome) and 

 the first cream was renamed cream a (the a, b denoting simply relative order of 

 discovery). For convenience in the further discussion of all these creams, the 

 term 'cream' when referring to the character will indicate the double mutant form, 

 eosin cream (W^w^ CrCr), and in those infrequent cases in which we refer to flies 

 homozygous for the cream gene, but not for eosin, the absence of the eosin char- 

 acter will be denoted by 'not-eosin' (WW CrCr or Ww« CrCr). 



