214 MUTATION AND PLANT BREEDING 



compared with genes controlling so-called qualitative characters, 

 there are no convincing reasons to assume that polymeric genes are a 

 different sort of genes so far as their nature is concerned. A mutation 

 step in one of these multiple genes may affect a quantitative char- 

 acter in a measurable way. For instance, increasing yielding poten- 

 tial of, say, 5 per cent in a top variety of barley, wheat, or oats is 

 usually considered to be a desired goal of breeding. It seems possi- 

 ble that such progress can be obtained by mutation of only one 

 or a few polymeric genes without a pronounced morphological 

 effect. However, the experimental evidence for such a speculation 

 is as yet meagre, except for the extensive evidence with peanuts 

 from the careful investigations of Gregory (39, 40, 41, 42). This 

 author found a striking increase in the genetic variance for yield 

 in progenies of normal-appearing M 2 plants which were selected 

 at random. He also succeeded in the selection of mutants with hioher 

 yielding capacity. 



In addition, Cooper and Gregory (15) presented evidence 

 of small mutations exerting a quantitative effect on leaf-spot resist- 

 ance. Starting with progenies of normal-appearing M 4 plants in 

 rice, Oka, et al. (81) found a considerable increase in the genetic 

 variance for plant height and heading date. For both these char- 

 acters, the authors succeeded in selecting plus and minus mutants. 

 In barley, a study of induced variability for quantitative char- 

 acters has recently been presented by Moes (73). He describes posi- 

 tive and negative alterations of a great number of characters, such 

 as number of tillers per plant, number of seeds per spike, kernel 

 size, lodging-resistance, leaf size, straw height, etc. Also at our 

 laboratory in Koln-Vogelsang, following a program to develop 

 selection methods for small mutations in barley, we succeeded in 

 obtaining a greater number of small variants in spring barley and 

 winter barley, and some of the results will be presented later in 

 this paper. 



Despite these studies, the field of small mutations is largely 

 unexplored as compared with that of large mutations. More infor- 

 mation is needed about their features and their frequency for an 

 evaluation of their significance in plant breeding. A priori it may 

 be expected, that the more genes that are involved in a character, 

 the higher the probability of obtaining an alteration by a mutation 



