In the case of irradiation of seeds of self-pollinated and disomic plant materials, the frequency of mutated plants appearing in the M2 generation divided by the Mendelian ratio is an index, as proposed by Gaul (1960), that enables the direct estimation of the mutation rate, representing the ratio of the number of mutated cells to the number of survived cells after treatment of mutagen. In the case of chronic irradiation of growing plants throughout the entire life cycle from germination to maturity, on the other hand, the mutations induced at the stage preceding the differentiation of the male and female organs appear at the M2 generation, the but mutations induced after this critical time are observed in the M2 plants in the heterozygous condition and are detected only at the M3 generation. Therefore, the mutation rate can not be correctly estimated only on the basis of the frequency of mutants either in the M2 or M3 generation. A new method for the estimation of the mutation rate by combining the data of the frequency of mutants and the M2 and M3 generation, which is comparable to Gaul's method for seed irradiation, is proposed. Application of this method to the data of acute and chronic gamma-ray irradiation of growing barley revealed that a much higher mutation rate was observed after irradiation at a higher daily exposure rate, when the plants were treated with the same accumulated dose.