408 MUTATION AND PLANT BREEDING 



attempts were made to select individuals from them for progeny 

 evaluation? Selection of plants from a heterogeneous population with- 

 in the limits of sampling that are imposed at best represent only a 

 limited portion of the potential population range. It would seem to 

 me that the probability of excluding the most unproductive end of 

 the population distribution would be greatly enhanced if selection 

 pressure for survival of individuals were allowed to operate on such 

 mutation-derived variants. 



During the progress of this session of the symposium major 

 emphasis has been directed to the use of mutagenic agents to produce 

 new variants in self-pollinated plants. The literature does not record 

 many irradiation experiments with the large group of economic 

 plants in which cross pollination is the rule. Obviously, it would be 

 difficult to distinguish between natural and induced mutations in 

 populations that predominantly are heterozygous as a consequence 

 of random cross pollination. The real question, however, is whether 

 or not there is need to produce new variability within a crop in which 

 abundant natural variability already exists. It is a common belief of 

 plant breeders that we have sampled and evaluated only a small por- 

 tion of the total gene supply in such species and hence adding to this 

 reserve, for the most part, is not likely to be an important breeding 

 procedure. It should not be inferred, however, that among established 

 inbred lines (as in maize) all desired characters have been fixed by 

 inbreeding and selection. The use of mutagenic agents to produce 

 desired variants among existing inbred lines is in reality a problem 

 comparable to their use in naturally self-pollinated crops. 



In summary, to this discussant, one should view the usefulness of 

 induced mutations not as a sole means to attain a desired objective, 

 but rather as a supplement to our present well-established procedures 

 that have adequately demonstrated their value in plant breeding. 

 This is not a "negative" point of view. If a desired character cannot 

 be found within the existing pool of germ plasm in a species or in its 

 related species, there is ample justification to create new variability 

 through induced mutations even though our present techniques are 

 comparatively crude in respect to gene selectivity. Finally, the use of 

 mutation breeding techniques rarely can be expected to immediately 



