mac key: induced mutation in crop improvement 347 



ly propagated and where a mutation is able to change a highly valu- 

 able autogamous genotype in a positive direction. In all other cases, 

 the new mutant will be utilized through recombination, and com- 

 bining ability will be about as difficult to see "from outside" as it is 

 between entries in a world collection. This limitation of mutation 

 breeding as an independent method is often overlooked, or perhaps 

 it is better to say that there has so far been a tendency to restrict, 

 mutation breeding to such specific uses where it is sufficient by itself. 

 It is true that one of the great promises in mutation breeding is the 

 possibility to add a single characteristic to a delicate system of genie 

 balance in which recombination may cause a breakdown. It is, how- 

 ever, just as true that a mutation will seldom be induced in its own 

 optimal genie environment. The extensive work at Svalof on stiff- 

 strawed erectoides mutants in barley may be a good example in this 

 discussion. 



We now know that it was a fairly gross simplification, when 

 Gustafsson and MacKey in 1948 (43) stated that "strength of straw 

 can be produced at will" in barley radiation experiments. The prom- 

 ised stiffness of straw was no mistake, but from hitherto 166 erectoides 

 mutations analysed extremely few have given a direct practical result. 

 Only one, the Pallas barley, has been released (7). The overwhelming 

 majority was induced in inferior germ plasm or had pleiotropic 

 by-effects which made them either lower in yield, more sensitive to 

 drought, more specialized in soil and nutritional demands, more 

 liable to stay in boot, etc. It is also important to observe that, the 

 erectoides factor in Pallas shows a different pleiotropic pattern, both 

 in relation to straw and head, if transmitted from the original Bonus 

 genotype. Thus, it interferes generally not as well with the brittle 

 straw of Carlsberg II but even better with the elastic straw of Rika 

 (Hagberg, personal communication). 



The erectoides mutations, as well as other categories of induced 

 mutations, have shown us that it is often much more efficient to 

 transfer a successful mutation than to try to induce it anew in another 

 genie background. It is further just as likely to make progress by 

 trying to transfer interesting mutations into new genotypes with the 

 hope for improved combining ability than to start searching in new 

 M 2 material. The recent introduction of growth chambers to speed 

 up the backcross technique has added further arguments along this 



