mac key: induced mutation in crop improvement 345 



technique by which a most rare mutant can easily be detected in a 

 large population. 



Screening for disease resistance by means of artificial infection 

 with known races (58, 59) or extracted toxin (116) allows enormous 

 amounts of individuals to be handled. Rapid chemical selection 

 methods like those elaborated for detection of sweet lupines (97) or 

 coumarin-free sweet clover (88, 90) will also greatly improve efficiency 

 in mutation breeding. The same is true for micro-methods for quality 

 analyses of individual plants like the new Svalof method for oil 

 determination which has radically changed the possibilities in breed- 

 ing oil crops for higher yield of fat (82, 1 12, 1 13). 



Properties like frost, heat, drought, and sprouting resistance, 

 where artificial selection methods are elaborated (57, 65), have prefer- 

 ence in mutation breeding. An interesting work along these lines is 

 reported by Zacharias (117). By simply germinating X 2 material of 

 soybeans at low temperatures, he was able to select mutants with 

 the ability to develop and start growth earlier, which is of greatest 

 importance for extending soybean cultivation to cool climates. In 

 crops with distinct response to day length or vernalization, mass 

 screening may also be used at low cost. By spring sowing of X-material 

 of Skandia III winter wheat, off-types were easily selected which were 

 able to shoot and ripen before the end of the vegetation period (69). 

 This technique is now used by Konzak (personal communication) in 

 large-scale breeding projects for better spring wheats. 



In all situations where aim and detection technique are precise, 

 selection can very well have set in already in Mo, the first segregating 

 generation. If the inventory is laborious or otherwise expensive, selec- 

 tion in M 2 will even be preferable, since it will imply the highest 

 chance to find the desired mutation among the smallest number of 

 plants (103). If the screening is easily done and the treatment given 

 to seeds or seedlings, it may, however, be advisable to select first in 

 M.,. where every mutation is no longer represented by single plants 

 but rather by a whole group of plants. By testing only a part of each 

 M 2 progeny, screening can also work efficiently even if the tested 

 material has to be destroyed during the analysis. All M 2 lines proved 

 to include the desired off-type can in this way be picked out and 

 propagated. In most situations, where the desired mutations are 

 difficult to detect due to imperfect selection methods or too vague a 



