mac key: induced mutation in crop improvement 351 



of the recent finding of a genetic control of the disomic behavior of 

 hexaploid wheat chromosomes (86, 87, 95, 96), it seems just as possible 

 that one mutational step would be enough to improve fertility as a 

 successive accumulation of many small changes of the chromosome 

 structure. 



All through the above presentation, the importance of an inter- 

 fusion between mutation experiments and conventional breeding 

 methods is stressed. Mutation is the first step in a creative process 

 followed by efforts to bring it in progressive interaction with the back- 

 ground genotype. This dependence on other breeding methods, how- 

 ever, can also be reversed. Mutations may open possibilities to apply 

 new approaches in breeding a specific crop. Thus, Lewis (66, 67) and 

 Pandey (83) were able to increase the mutation rate of sell- 

 incompatibility alleles in Oenothera organensis and Trifolium 

 species, respectively, by the use of X-irradiation. Similar phenome- 

 non can also be produced as primary and thus nonheritable effects of 

 radiation (9). It may also be of interest to mention that gamete 

 irradiation is a possibility to overcome certain interspecific barriers 

 (14, 84, 85, 108, 109). Primary irradiation is also known to induce 

 intersexuality in hemp (79). 



The ability of radiation to disrupt a biochemical sequence and 

 to induce a heritable shift in the same metabolic pathway is also 

 demonstrated by the interesting transfer from apomixis to sexuality in 

 Poa pratensis (32, 53, 54) and Potentilla species (Asker, unpublished). 

 Apomixis is evidently dependent on a rather complex gene balance 

 where many different genetic events may interfere and induce a high- 

 er or lower degree of sexuality. Due to the high tolerance to chromo- 

 some disturbances in the multiploid Poa, even losses of whole chro- 

 mosomes were ultimately found to result in sexuality. The artificial 

 induction of such a shift enables the breeder to utilize very strictly 

 apomictic clones as parental components in crosses or merely to 

 explore the variability inherent to an extreme heterozygote. In both 

 cases he may select further among more or less sexual segregates or 

 isolate fixed, apomictic types, where the balance is restored (Figure 1). 



The above review is an attempt to understand mutation experi- 

 ments and their possibilities in plant breeding. Just as the induced 

 mutation can be evaluated only in relation to its background geno- 

 type, mutation breeding can be evaluated only as an integral part 



