抄録
Since a reprint of the paper concerning the "Ein-Korn-Ramsch" method put foreword by FREISLEBEN and LElN (1943) could be secured, their method is discussed in relation to the "one-plant-one-grain method" which is one of the new methods proposed previously by the author. The rigorous proofs are mathematically given for their method. Comparing their method with the author's, the procedure of taking one seed from each of the X1-plants is identical, but in other respects it is quite different ; and it is possible to conclude that the theoretical arguments and conculusion, the mathematical approach and the utility for application to radiation breeding of the author's method are much better than those of FREISLEBEN and LElN's method. It is very common to face two or more different kinds of desirable mutants in the practice of radiation breeding. Then, the following three cases are concerned in an X2-population : 1) detecting at least one mutant belonging to the j-th kind of mutants, 2) detecting at least one mutant belonging to any one of all of the different kinds of mutants, and 3) obtaining at least one mutant per kind for the different kinds of mutants. In all of these cases, the "one-plant-two-grain method", the "one-plant-three -grain method" and especially the "one-plant-one-grain method" will frequently be much better, because of the smaller or smallest total of X2-plants, and/or X1- and X2-plants in the aggregate. These results will be useful for breeders apart from the discussion for the "Ein-Korn-Ramsch" method
