BRISTLE INHERITANCE IN DROSOPHILA 91 



ing through the influence on the food. There can be no question 

 that extra bristles are not influenced by temperature in the 

 way that Miss M. Hoge found applied to extra legs. 



DISCUSSION 



The question in relation to which these experiments have most 

 interest is whether Mendelian units can be modified by selection, 

 and whether selection can accomplish anything more than a 

 sifting and sorting of hereditary elements whose origin is still 

 unknown. The tendency is either to hold that the hereditary 

 elements can never be modified by simply propagating certain 

 ones, or to hold that there is no integrity of such seeming ele- 

 ments or factors, and that almost anything can be accomplished 

 by sufficiently long and painstaking selection. 



Some of the observations will admit both interpretations. 

 According to the selectionist's view, the modification of the extra 

 bristles appearing in a cross with normal, indicates that a factor 

 has been modified. The restricting factor of the normal gamete 

 may have transferred some of its restrictive properties to its 

 allelomorphic mate in the extra germplasm, so that all the chil- 

 dren formed by the union of two so modified exti^a gametes, 

 would have fewer extra bristles than if the gametes had been 

 unmodified. Selection made steady advances at first, showing 

 that the gametes of flies of higher grades must be somewhat 

 different from those of the lower grades, and therefore, the factor 

 for extra bristles must be a variable thing, however truly a Men- 

 delian unit. That the progress of selection does not continue 

 after the sixth generation may only mean that the conditions 

 have been so irregular or unfavorable that the real phenomena 

 of the germplasm were entirelj^ veiled. 



On the other hand, these same facts may be used to form the 

 basis of an interpretation involving one or more smaller, or 

 accessorjT^, restricting factors, which are found in many low 

 bristled flies (some flies may lack all restricting factors, and still 

 be low grade on account of their small size). The more and 



