506 MUTATION AND PLANT BREEDING 



mutagenic action may be a property resident in the breeding sys- 

 tem per se, and not entirely conditioned by such factors as evolu- 

 tionary age, level of adaptation, inherent mutagenic resistance, 

 tolerance of aneuploidy, level of ploidy, nuclear volume, etc., which 

 have been considered previously and in more detail by MacKey 

 and Sparrow. 



An important limitation to the expression of genetic varia- 

 tion is the effective size of the population which can be studied. 

 In conventional breeding programs population size may be restricted 

 by incompatibility and sterility barriers and by technical difficul- 

 ties in obtaining Fj seeds in sufficient quantities. In mutation breed- 

 ing the effective population size is also limited, not by the number 

 of seeds which can be irradiated, but by the number and kinds 

 of plants which survive after initial injury, induced sterility, and 

 diplontic selection have taken their collective toll. Nevertheless it 

 appears that the judicious use of appropriate mutagens, the manip- 

 ulation of environmental conditions and the elaboration of screen- 

 ing techniques which have been discussed by Gaul, Nilan and Kon- 

 zak, Caldecott, and others, provide a greater opportunity for accu- 

 mulating variation per unit population size than do conventional 

 breeding methods. It may well be, but has yet to be demonstrated, 

 that the amount of useful variation is proportional to the total 

 amount of variation present. 



Many of the conventional plant breeder's efforts are applied 

 to selection in polygenic systems. These sometimes appear to be 

 genetically incomprehensible and statistically exhausting, but they 

 have contributed considerably to long-term plant improvement. The 

 breeder cannot always expect to find in his material economically 

 valuable characters which behave as Mendel thought they should. 

 But when he finds opportunities to obtain disease resistance and 

 other specific qualities through simple gene transference, he accepts 

 them gratefully. Some of the most spectacular advances have been 

 made through the selection of qualitative characters, and some 

 crops like sorghum, as Quinby has shown, are particularly well- 

 endowed with such opportunities. In general, though, it is prob- 

 ably fair to say that spectacular advances through the incorporation 

 of qualitative characters have always been accompanied by the 

 long-term selection of appropriate polygenic complexes which have 



