OF MUTANT CHARACTERS. 231 



females were tested either singly or in mass-cultures, and no offspring 

 were produced. On the other hand, the males were entirely fertile. 

 This case was the third to show female sterility. Rudimentary 

 females usually gave no offspring, but occasionally gave a very few 

 which were found to be almost entirely female. An extensive series 

 of outcrosses of rudimentary females (carried out in 1912 by Bridges) 

 produced a total of 623 females and only 9 males. 



The sex-linked character " fused" was likewise found to be sterile 

 as to its females and fertile as to its males, though no such extensive 

 tests were made as in the case of rudimentary and "spineless. " Since 

 these three, several other mutations whose females were completely 

 sterile and three mutations with completely sterile males have been 

 found. All of these mutations with unisexual sterility have been 

 turned over to Miss Clara J. Lynch for further investigation. 



After the failure to maintain "spineless' ' stock by use of the females, 

 the stock was run by mating in each generation "spineless" males by 

 wild-type sisters heterozygous for the gene. 



ARRANGEMENT OF THE FACETS OF MORULA 



At this time it was noticed that there was present in the " spineless " 

 cultures a new type of eye modification, which was called "morula," 

 since in it the facets of the eye had lost their regular pattern and were 

 crowded together irregularly, much like the drupelets of a mulberry. 

 The facets tended to "round up" and become more strongly convex, 

 and to become more circular than hexagonal in outline (plate 10, fig. 3 6). 

 The facets were also irregular in size and color, large ones being usually 

 darker than those smaller. The eye as a whole was somewhat smaller 

 than normal and more convex. The light reflected from the surface 

 was broken up into many twinkling points by the tiny hairs which were 

 found to be pointing in all directions instead of only radially (figs. 3 6 

 and 3 c). This feature is common to several of the later discovered 

 moruloid mutations and is the reason for the name "star" borne by 

 the most useful of them all. 



Most of the morula flies were found to be also "spineless;" several 

 counts established the fact that every "spineless" was at the same 

 time "morula, " but that only about 90 per cent of the "morulas" were 

 "spineless." This fact led to the suspicion that the two characters 

 were effects of the same gene and that while the "morula" was a 

 constant index of the mutation, the "spineless" might be called an 

 "accessory" character. This agreed with the difficulty encountered 

 previously in classifying "spineless" and with the too small proportion 

 separated out. The differences between the numbers of "morula" 

 and "spineless" flies gave a measure of the unreliability of "spineless. " 

 Throughout subsequent experiments attention was paid to the relation 

 of "spineless" and "morula," and since no case was found in which a 



