BRIDGES 



fairly weak bands, while a very sharp 

 change in size occurs at the transition 

 from 15F to 16A. 



In the Bar chromosome the condi- 

 tion may be described observationally 

 as the repetition of section 16A, with 

 the exception of the final very faint 

 dotted line. But the whole region of 

 this bulb has undergone changes in the 

 Bar chromosome as follows: the "puff" 

 of the bulbous segment is more pro- 

 nounced and its size is increased; the 

 banding is more discontinuous by 

 being broken into blocks and vesicles, 

 and the regularity of synapsis is dis- 

 turbed by oblique junctions. Thus, in 

 Bar the heavy doublet following the 

 last faint dotted line of 15F is more 

 segmented than normal and more 

 rarely shows its doubleness clearly. 

 This tendency is more pronounced in 

 the heavy broken line of the repeat 

 seriation to the right. All the lines of 

 the repeat seriation to the right differ 

 from the corresponding lines of the 

 initial seriation by being somewhat less 

 intense, more broken, more diffuse and 

 more confused in their synapsis rela- 

 tions. 



In a forked non-Bar stock recently 

 derived from the above forked Bar 

 stock by breeding from the rare Bar- 

 reversions, the banding was found to 

 be precisely identical with that of un- 

 related normals as far as could be ob- 

 served in excellent permanent prepara- 

 tions of well-stretched chromosomes. 



In a forked Bar-double stock, sim- 

 ilarly derived from the same f B stock 

 by breeding from the very rare "Ultra- 

 Bar" type of eye, it was found that 

 the extra section observed in Bar was 

 present still again, giving a thrice-re- 

 peated seriation in direct sequence. 

 The changes differentiating Bar from 

 normal were carried further in Bar- 

 double, as follows: The size and puffi- 

 ness of the bulbous regions was still 

 greater, as well as the blockiness of the 



165 



banding and irregularity and oblique- 

 ness of the synapses. These disturb- 

 ances were greatest in the miudle one 

 of the three seriations. 



These findings enable the Bar "gene" 

 to be reinterpreted as a section of in- 

 serted genes— a duplication. The pro- 

 duction of Bar-double and of Bar- 

 reverted is seen to be the insertion of 

 this extra section twice, or conversely, 

 its total loss— both presumably by a 

 process of unequal crossing-over. That 

 the section of bands should behave as 

 a unit in this process is perhaps ac- 

 counted for by the observation of 

 oblique synapsis, especially frequent 

 in Bar-double, where presumably one 

 entire sequence synapses with another 

 of a different position in the series of 

 three. The oblique synapses were even 

 more frequent in BB/B + , where one 

 series in B+ has a choice of three series 

 in BB, apparently usually synapsing 

 with one or the other end series. 



According to this interpretation the 

 source of the duplication is the ma- 

 terial directly adjacent to the repeat. 

 But whether the point of insertion 

 preceded the heavy doublet of 16A1 

 or the very faint final singlet of 16A5, 

 can not be determined. If Bar is itself a 

 repeat, a reason is thereby provided 

 for its unique behavior of giving rise 

 to Bar-double and Bar-reverted by 

 oblique synapsis. Perhaps half of the 

 Bar-reversions carry the original series 

 and the other half the subsequent re- 

 peat restored to its original position. 



On this interpretation, the "position 

 effect"— the reinforcement of the ac- 

 tion of one Bar gene by another in di- 

 rect sequence next to it— has a visible 

 cytological accompaniment in the in- 

 creased size and puffiness, and the 

 change in the character of the banding 

 of both series in Bar as compared with 

 normal and of all three series in Bar- 

 double as compared to Bar itself. Part 

 of this is presumably due to the 



