March 31, 1911] 



SCIENCE 



497 



duced, the mai'ginal vein of whose wings was 

 beaded. Bred to his sisters, the beaded condi- 

 tion appeared in one fly in sixty of the next 

 generation. The beaded flies were inbred and 

 produced in the third generation, one beaded 

 winged fly to thirty-five normal. The beaded 

 flies produced in the next generation one 

 beaded to twelve normal wings. The same 

 process continued through many generations 

 has finally produced stock that gives in certain 

 cultures nearly 100 per cent, beaded wings. 



Not only has selection slowly increased the 

 percentage of abnormal wings, but much more 

 extreme forms of the same variation have ap- 

 peared, and an attempt is now being made to 

 fix some of these extreme variations. 



When a beaded male of the stock in its pres- 

 ent condition was bred back to the wild fe- 

 males the beaded condition appeared in one 

 fly in twelve of the first generation. In the 

 second generation from non-beaded flies the 

 proportion was one to twenty. 



In the further course of the experiment 

 eight other modifications have appeared that 

 are undoubtedly mutations; one confined 

 largely to females, some of the others showing 

 strictly alternate inheritance, combined with 

 " sex-limited transmission." These modifica- 

 tions may be described under the following 

 headings. 



Truncated Wings. — The normal wings of 

 Drosophila are longer than the abdomen, ex- 

 tending for about one third their length be- 

 yond the end of the body. The wings have a 

 rather pointed end. In the seventh generation 

 of the beaded wing stock a fly appeared with the 

 end of the wings cut off nearly squarely, and 

 indented at the end or somewhat scalloped. 

 Bred to sister flies there were produced in the 

 next generation twenty-one flies with trun- 

 cated wings to 230 with normal wings. In 

 the nest generation some of these truncated 

 winged flies produced nearly 50 per cent, of 

 flies with truncated wings. It has proved 

 difficult to pass beyond this point, although 

 certain of the later cultures have produced 

 over 90 per cent, of truncated winged flies. 

 This high average has not, however, been 

 maintained in the next generation. There are 



indications, nevertheless, that the inheritance 

 may be raised even to this high standard. 

 Two main points of interest have appeared in 

 the breeding of this modification. First, the 

 condition is confined almost exclusively to the 

 female line. A comparison of many cultures 

 shows that the offspring of truncated parents 

 consist of almost equal numbers of males and 

 females with normal wings, while the trun- 

 cated winged forms are, as stated, confined 

 almost exclusively to the females. The defi- 

 ciency of males in the entire culture is due, 

 therefore, to the absence of males with trun- 

 cated wings. Yet such males appear in small 

 numbers, and especially in cultures in which 

 truncated males are the fathers of the next 

 generation. 



Secondly, by selecting the shortest-winged 

 individuals in each generation the wings have 

 been reduced in some cases until they are not 

 longer than the abdomen, and in a few cases 

 they were even much shorter than the ab- 

 domen, and are not superficially different in 

 this respect from the type to be next described, 

 although in their inheritance they follow an 

 entirely different procedure. The character 

 can not be said to be sex-limited to the female 

 in the conventional sense. When a female 

 with truncated wings was bred to a normal 

 wild male all of the offspring in the first 

 generation had long wings. These inbred 

 produced in the second generation 1,826 

 normal flies to 36 truncated individuals, males 

 and females. When a truncated male was 

 bred to a normal wild female all the flies in 

 the first generation were normal, and in the 

 second generation, when these were inbred, 

 there were produced 1,408 normals to 14 

 truncated males and females. These facts 

 indicate that this condition is not connected 

 with sex-limited inheritance, although other 

 wing modifications, as will be shown, are inti- 

 mately connected with sex transmission. It 

 should be added that normal winged flies of 

 this modified stock throw a high, though 

 variable, percentage of flies with truncated 

 wings. 



Rudimentary Wings. — In the second gen- 

 eration of the beaded winged flies a male ap- 



