132 MUTATION AND PLANT BREEDING 



tendency of chemical mutagens to produce minute rather than 

 large rearrangements may be due to some spreading effects of the 

 chemical damage. 



Of particular interest is the ability of a few — and possibly of 

 more — chemical mutagens to produce high frequencies of small 

 duplications. Many duplications were found in Vicia chromosomes 

 that had been treated with nitrogen mustard (38), and formalde- 

 hyde food produced many duplications in Drosophila (81). 



While there is good evidence that a common mechanism is 

 responsible for the production of gene mutations, chromosome 

 breaks, and minute rearrangements by chemicals, crossing-over 

 seems to be induced in some different way. Formaldehyde food pro- 

 duces crossing-over but not mutations in spermatogonia (83, 86). For 

 a given frequency of mutations, mustard gas produces considerably 

 fewer cross-overs than does formaldehyde (85). The relative fre- 

 quencies of lethals and induced cross-overs were markedly changed 

 when mustard gas was given to male pupae rather than imagines 

 (10). These findings agree with observations on plants which show 

 that there is no correlation between the ability of a chemical to 

 produce translocations and its effect on chiasma frequency (55). 

 Further, chemical treatment of plants can produce translocations 

 in pachytene when crossing-over is completed (54). 



In barley, gene mutations are scored as segregants for visible 

 abnormalities, mainly chlorophyll defects, in F 2 , while chromo- 

 somal changes are recognized by their effect on the fertility of the F^ 

 Judged by these criteria, chemical mutagens differ widely in their 

 relative abilities to produce intergenic and intragenic changes (30). 

 Curiously enough, the extremes are formed by two purines — 8- 

 ethoxycaffeine, which produces almost exclusively chromosome 

 changes, and nebularine (purine-9-riboside) which produces almost 

 exclusively mutations. Unfortuntately, it is not possible to deter- 

 mine how many of the segregating chlorophyll mutations are caused 

 by small deficiencies or rearrangements. Even so, the striking dif- 

 ference between the effects of these two substances, and similar 

 differences between other chemicals, lend support to the model of 

 a chromosome in which the genes are connected by links of non- 

 genic material. 



