1304 BIOLOGICAL EFFECTS OF RADIATION 



A small amount of nonhomologous pairing is evident in most of the 

 interchanges. In some there is a large amount (Burnham 15, 16; 

 McClintock, 32). It has not yet been ascertained whether these cases 

 of extensive nonhomologous pairing are due to some other alteration 

 in the chromosome at or near the point of interchange. 



The extensive series of interchanges listed by Anderson (2) from 

 irradiation are distributed among the 10 chromosomes approximately 

 in proportion to the relative lengths of the different chromosomes. Not 

 enough data have been reported to show the relative frequency of 

 interchanges in different portions of a chromosome. 



NATURE OF THE ALTERATIONS PRODUCED BY IRRADIATION 



All of the chromosome alterations produced by irradiation, except 

 terminal deficiencies, involve an interchange of parts between non- 

 homologous regions. Terminal deficiencies imply a break in the chromo- 

 some without the broken ends becoming reattached. 



There are no data available on the relative frequency of inversions, 

 terminal deficiencies, internal deficiencies, ring chromosomes, and 

 interchanges. Any simple relationship between the frequencies of 

 different alterations would go far toward indicating the nature of the 

 mechanism involved. Perhaps such data are not to be obtained because 

 of the differences in the means of detection. The absence of "simple 

 translocations" in maize seems significant. So far as studied, all altera- 

 tions involving two chromosomes are reciprocal. All inversions found 

 are internal. No attachments to or by the normal ends of chromosomes 

 have been found. These same normal ends are conspicuously prone 

 to be stuck to other chromosomes in prophase preparations. This 

 stickiness is probably not merely an artefact produced by the reagents 

 used in fixing and staining. The stickiness disappears in the later, more 

 condensed stages. Interchanges occurring near the ends of chromosomes 

 are difficult to detect and to study, except some of those in the satellite 

 of chromosome 6. Of 18 interchanges known for chromosome 6, three 

 are in the satellite itself and do not involve more than a maximum of 

 three chromomeres. Finally, two of the three deficiencies found for ygz 

 (Creighton, 21) were shown by the terminal knob to be internal. This 

 throws suspicion on the validity of terminal deficiencies. 



Crossing over in Drosophila has been shown to take place in the double 

 strand stage and only two of the four strands cross over. Rhoades 

 (42, 43) has shown that the same applies to maize. Since interchanges 

 between nonhomologous chromosomes resemble crossing over in other 

 ways, it seems not unreasonable to place it also in a double-strand 

 stage when the conditions are most nearly those surrounding crossing 

 over. We may likewise expect only two of the four strands to become 

 interchanged. 



