300 THE SECOND-CHROMOSOME GROUP 



reeled to 26.2 according to the probable coincidence of 100, and 

 weighted at 50 per cent of the 62,679 flies (31,340) . The mean position 

 of curved is 27.0 units to the right of black. 



The fifth locus to be considered is that of dachs, which lies to the 

 left of black. The dachs black data give a cross-over value of 17.8 

 corrected to 18.5 according to the probable coincidence of 50 and 

 weighted at about 80 per cent of the 6,725 flies (5,380). The dachs 

 purple value of 19.7 is corrected to 21.1, according to a probable 

 coincidence of 60 and weighted at about 75 per cent of the 1,489 flies 

 (1,150). The locus is thus at 6.2-21.1, or -14.9. The coincidence 

 in the case of dachs vestigial is known to be about 85, so that the value 

 of 29.6 can be corrected accurately to 34.6, corresponding to a locus of 

 — 16.1 and weighted at about 30 per cent of the 5,354 flies (1,605). The 

 mean locus of dachs is thus 17.5 units to the left of black, or at —17.5. 

 With the mapping of the positions of the four genes, black, purple, 

 vestigial, and curved, and also the position of dachs, which lies to the 

 left of black, the skeleton of what may be called the central body of 

 genes is completed. The next step is to tie onto this central group the 

 outlying loci at either end. Of those to the left, streak is the most 

 important locus, and its position is found by combining three sets of 

 data. The streak dachs cross-over value of 12.7, based on 858 flies, 

 needs no correction, since the probable coincidence is under 10 and the 

 correction negligible in amount. The streak black value of 26.0 

 should be corrected to about 28.2, corresponding to a coincidence of 

 60, and wdth the 462 flies weighted at 65 per cent, i. e, 307. The most 

 extensive data is that on streak purple, but the 2,665 flies should be 

 weighted at only about 50 per cent of their number, or at 1,333. The 

 coincidence is probably about 90, so that the value of 33.1 becomes 

 40.4, corresponding to a locus of — 34.2. The mean position of streak 

 is at —31.1. 



The position of star is 15.9 units to the left of streak or at —47.0, 

 according to the star streak cross-over value of 15.9, which needs no 

 correction. The star dachs value of 27.3 is corrected to 28.8, according 

 to the probable coincidence of 30. The locus indicated is 46.3 units 

 to the left of black. The 3,472 flies may be rated at about 80 per cent 

 of their number, or at 2.778. The star black value of 37.9 probably 

 represents a total of 46. o per cent of crossing-over, with a coincidence 

 of 90, and the 16,507 flies may be rated at about 35 per cent of the 

 number, or at 5,775. The mean position of star is thus at —46.5. 



The location of all the right end is dependent on speck, which is 

 itself mainly dependent on curved. The curved speck value is 30.5 

 and the coincidence is known to be very low in this region, probably 

 not over 20, so that the corrected value is about 31.0. Because of this 

 low coincidence a relatively large weighting can be assigned (85 per 

 cent = 8,540). The vestigial speck data furnish 2,054 flies weighted 



