MANNER OF PRODUCTION OF MUTATIONS 495 



wings, found that the spontaneously arisen ones tended to be very small, 

 of not more than 13 of the "bands" of Bridges' (1938) standard map. 

 Those resulting from the X irradiation of spermatozoa had a much larger 

 range of size, up to about 40 bands, but the larger they were, the lower 

 was their frequency, exactly contrary to the expectation for random posi- 

 tions of breakage. These induced deletions were interpreted as probably 

 being composed of one group of very small dimensions, like the spontane- 

 ous ones, in which the two breaks were produced by a single hit (or track), 

 and a second group, occupying a larger range of size, which was limited 

 only by the reduced viability of large deficiencies. The deletions of this 

 second type were supposed to have had their two breaks produced by 

 independent hits (or tracks). 



It might, however, have been postulated that the reason that the spon- 

 taneous deletions occupied a more limited size range than the others was 

 because most of them had arisen in interphase nuclei, in which the chro- 

 mosomes are less coiled and in which there is therefore a greater physical 

 distance between points a given distance apart along the chromosome, 

 whereas the induced deletions had arisen from breaks induced in the 

 tightly coiled chromosomes of spermatozoa and from unions occurring in 

 chromosomes which were still somewhat more coiled than in ordinary 

 interphase. As for the question why even those deletions which had 

 been produced by radiation showed a frequency w^hich declined with size, 

 the reason (in addition to that of the lower viability of larger deletions) 

 might have been that ends derived from more distant breaks are less 

 likely to come into contact with one another. It is true, as opposed to 

 this, that for breaks much farther apart, represented by gross inversions, 

 the frequencies of different sizes had been found to correspond to a ran- 

 dom distribution of breaks (Bauer, Demerec, and Kaufmann, 1938). 

 However, these inversions must have involved chromosome loopings on 

 a far larger scale, and so they would have been much more likely to be 

 free of such small-scale distance limitations. Therefore there are serious 

 difficulties in the way of using these results for drawing conclusions con- 

 cerning the mechanism of production of the breaks that take part in small 

 deletions. 



It happens, moreover, that the Notch region of the X chromosome 

 shows certain peculiarities which suggest that it may contain a duplica- 

 tion of loci and may include material of a more or less heterochromatic 

 nature (Prokofyeva-Belgovskaya, 1939). It may therefore be subject to 

 forces of attraction or of stress which, as in typical heterochromatic 

 regions, cause union of nearby (not necessarily extremely nearby) points 

 to be especially likely to occur. At any rate, recent work (J. I. Valencia 

 and Muller, 1949) on the sizes of deletions induced in several other (per- 

 haps more typical) euchromatic regions by X irradiation of Drosophila 

 spermatozoa has shown, for deletions of more than three of the bands of 



