262 



THE SECOND-CHROMOSOME GROUP 



CHROMOSOME CARRYING STAR. 



To test the relation of star to the third chromosome, a star male 

 was out-crossed to the double-recessive peach sooty (peach is an allelo- 

 morph of pink, and sooty an allelomorph of ebony) . In FI the flies were, 

 as expected, half stars and half wild-type (culture 1730, table 111). 



Some of the FI star females were back-crossed by peach sooty males 

 (table 111). With two loci as far apart as peach and sooty were known 

 to be, there was no need to run a male back-cross test, since the female 

 test must readily reveal linkage to one or to the other of these two 

 loci if the tested gene is in the third chromosome. As a matter of fact, 

 there was complete independence of star and peach (52.6 per cent of 

 recombination) and also of star and sooty (50.4 per cent of recombi- 

 nation). Peach and sooty gave 27.3 per cent of crossing-over, which is 

 a trifle higher than the usual value. 



LOCUS OF STAR. 



Since the locus of star was proved not to be in the third chromosome, 

 the chances were about 50 to 1 that its locus was in the second chro- 

 mosome. This probability was so great that an extensive experiment 

 was planned and started without the relation to the second chromosome 

 having been previously tested. This experiment was the quadruple 



TABLE 112. PI, star 9 X purple curved speck cf. 



back-cross of star and purple curved speck, which was to serve several 

 purposes. In the first place, it was to give an accurate measure of the 

 amount of crossing-over between curved and speck, which was very 

 important, since up to that time only a relatively small amount of 

 data was available on this value whereby the locus of speck and with 

 it the entire right end of the chromosome was mapped in relation to the 

 rest; in the second place, it was to establish the locus of star, which, 

 as was then realized, might prove to be the most useful of all the second- 

 chromosome characters. These linkage values were both to be con- 

 trolled and linked up by means of the accurately mapped loci purple 

 and curved. The third purpose was to test more thoroughly the extent 

 and nature of the change of crossing-over with age in different broods 

 and in different regions of the second chromosome, but more especially 

 the relation between this change and the change in the amount of 

 coincidence (see Bridges, 1915). 



