432 T. H. MORGAN AND C. B. BRIDGES 



For the sake of simplicity, in this paper we have restricted our 

 studies to four eye color factors, although we have confirmatory 

 data on important points from experiments with five other pri- 

 mary eye colors. The four eye colors used here have been described 

 and figured in former papers, but a short comparative description 

 may assist in following the experiments. 



The eye of the wild fly is a deep red, dense, and with a dark 

 metallic fleck which changes its position according to the angle 

 of light and the viewpoint of the observer. 



The white eye (sex-linked) looks as' though it were due to a 

 white pigment. A transparent fleck indicates the facet turned 

 directly to the observer. 



The eosin eye color (sex-linked) has been already described. 

 We may add that the color is slightly cloudy in appearance 

 except at the margin, where a rim of greater translucency gives 

 the impression that the pigment has retreated from the periphery 

 to the deeper portion of the eye. The fleck is yellowish. 



The vermilion eye (sex-linked) is a very bright, clean, true 

 vermilion, not transparent, and with the fleck very inconspic- 

 uous. This eye was formerly called 'coraF or 'bright red.' 



The -pink eye color (third chromosome group) is a transparent,' 

 deep, pink, which uniformly fills the eye, and shows no fleck. 



The various 'permutants' or double and triple recessives pos- 

 sible with these primary eye mutations will be described as we 

 have occasion to use them. Since, however, white is invariable 

 in color it is impossible to distinguish between simple whites, 

 white vermilions, white pinks, and white vermilion pinks. In 

 objective classifications these will all be referred to as 'white.' 



A recent paper by Morgan- gives our reasons for the use here 

 of a new system of nomenclature. We shall use as the symbol 

 for the factor which differentiates any mutations from the nor- 

 mal, the initial letter of the descriptive name of the mutation. 

 Since we now have over one hundred primary mutations we 

 shall often have to use two letters to avoid confusion. For 

 example, we use for the factor concerned in the recessive muta- 



^ Factors and unit characters in Mendelian heredity. Am. Nat., January 1913. 



