496 University of California Publications in Zoologij [Voi,. i:> 



virginity of the females was assured by using only those whereof the 

 time of hatching was known and removing them from the bottles 

 immediately, Ix'fore the wings had uncurled or the pigment had formed 

 to any extent. 



Besides the extra-bristled flies, other variations were found, inci- 

 dentally, one club-winged mutant, one individual with only one wing 

 and with head and thorax bent much to one side. There wxre also 

 seven beaded-winged mutants, but as they were exceedingly non-viable 

 and sterile no offspring were obtained. 



Abdominal variations in banding were found and were carried 

 through three generations with a verification of Morgan's theory that 

 the inheritance is dependent on a factor of environment, namely moist 

 food. After the character makes its appearance it may be kept or 

 masked by the character of the food. It is not induced to appear by 

 moist food, but its presence and inheritance is made manifest by the 

 continuance of moist food. 



An inherited difference in pigmentation- was also observed. It 

 occurred in the shape of a shield on the thorax between the bristles, 

 finally increasing into a trident with the ends running forward to 

 the head. There was also a marked irregular line of pigmentation 

 extending transversely in a jagged line between the anterior bristles. 

 Lack of time prevented extensive study of this variation. 



All the flies were carefully examined throughout for other possible 

 differences, but with the above-mentioned exceptions the individuals 

 were uniform in all their characters. There were, therefore, no facts 

 on which it might be possible to base an assumption that the various 

 ratios obtained were explainable through linkage with other characters. 



Five strains, X, Y, Z, A, and B, were used, and the bristles of 450Q 

 flies were examined. 



I wish to express my sincere gratitude to Professor R. Ruggles 

 Gates, Avho criticized and made suggestions on the work from time to 

 time. 



Crossing Experiments 



As has been said, the normal number of the bristles is four, form- 

 ing a rectangular pattern on the dorsal surface of the thorax. The 

 exact positions of the extra bristles are not inherited, but they occur 

 in random positions on the thorax. However, more often than not, 

 they are in line with the normal bristles. Following are some of the 

 arrangements : 



