In our laboratory we have about a hundred progeny lines of rice plants descended from the plants which were exposed to atomic radiation in Nagasaki. These lines have been classified into some categories, genic mutation lines, chromosome structural changed lines and so on, according to their genetic behavior or the characteristics of their abnormal derivatives. However, many kind of variant types have occurred irregularly in the process of pedigree culture of them even now. The author planned to clarify such complicated genetic behavior of the progenies of atomic bombed rice plants. The present paper deals mainly with the results of the cytogenetical studies, First of all, the data for kind of variants, their morphological and genetical characteristics and others in the progeny lines which demonstrated before 1958 by Dr. T. Nagamatsu were arranged. As a result of this arrangement of data, it was shown that the progeny lines including chromosome structural changes ceuld be classified into three groups as follows. 1) Semi-sterile lines, which descended from the semi-sterile mother plant and segregated into normal and semi-sterile plants in the ratio of 1:1. 2) High sterile lines, which descended from the high sterile mother plant and segregated into normal, semi-sterile and high sterile plants in ratio of about 1:2:1. These sterility and the segregation ratio are caused by the reciprocal translocation. 3) Primary and/or tertiary trisomic lines derived from their sterile lines. And moreover, it was supposed that there were not a few chromosome structural changes, which is latent in these lines as homozygous conditions and can be detected only by test crossing to chromosomal standard plant. By the test crossing between chromosome standard plant and fertile plants descended from progenies of atomic bombed rice plants, the plants having chromosome structural changes were detected in 60 out of 79 lines used. According to the observation of seed fertility (s), pollen fertility (p) and the chromosome behavior at meiosis in the PMCs of the F_1 hybrids, these chromosomal changes were categorized into A (s≑p≑50%, a quadrivalent was observed), B (s≑p≑50%, a quadrivalent was not observed), C (s≑p≑50%, chromosome observation was not yet made), D (s<p, a quadrivalent was observed infrequently), E (s>p, a quadrivalent was not observed) and compound types of them. It was indicated that each of the types, A, B, C and D, was the homozygote for one reciprocal translocation and that the compound types such as double A (s≑p were lower than 25％, two quadrivalents were observed) and A+D (s<p＝30％, a quadrivalent was observed but two quadrivalents were observed infrequently) were the homozygotes for two reciprocal translocations. Therefore, the semi-sterile plants which segregated in semi－sterile lines and in high sterile lines were heterozygous plants for A, B, C and D types and the high sterile plants which segregated in high sterile lines were heterozygous for the compound types of chromosome structural changes. These facts were also confirmed from their segregation patterns of fertility in the F_2 generations. Moreover, it was observed in several cases that the H hybrids between standard plant and fertile plants descended from semi-sterile lines showed the high sterile plants of compound types in addition to semi-sterile plants. It demonstrated the being of latent structural changes in sterile lines. For the identification of the reciprocal translocations isolated from progenies lines, the F_1 hybrids were compared with each other in some characteristics, with special reference to the chromosome configuration and frequency of quadrivalent at meiosis. And these translocation lines were also tested as to homology of translocated chromosomes by reciprocal crosses among them. As a result, it was confirmed in some cases that the homologous changes existed among the lines even which derived from different original lines. For example, 6 kinds of homologous changes were observed in category A, and moreover, in category D, one homologous change was identified among 10 lines which derived from the 5 original lines. Nevertheless, about 20 different reciprocal translocations were isolated. Then, they were analyzed for translocated chromosomes by means of reciprocal translocation analysis. Occasionally, the high sterile plants, which may be the heterozygote for two of chromosome structural changes, have occurred in sterile lines in the process of the maintenances and the renewals of progeny lines of atomic bombed rice plants. The cause of that was infered from the mode of occurrence, genetic behavior of them and moreover the outcrossing ratio of semi-sterile plants as follows: as a result of outcrossing, 1) the chromosome structural changes which had been latent as homozygous condition in semi-sterile lines were appeared as heterozygous condition in high sterile plants, or 2) some chromosome structural changes were introduced from other semi-sterile lines. Most of the progeny lines of atomic bombed rice plants have been maintained up to date through such high sterile generations, once or more, in process of line renewals. This fact was probably supposed the cause of the existance of the homozygous chromosome structural changes among the lines which derived from different original lines and, chromosome structural changes among the lines which derived from different original lines and, consequently, there were fewer changes than that expected in progeny lines of atomic bombed rice plants at the beginning of this study The D types, in which a quadrivalent was not observed practically, had rather higher pollen fertility (60-70%) than seed fertility, so it was suggested that the male gametes with a duplicate－deficient of translocated chromosomes may be viable in a certain case. On the E types, which had been categorized by lower pollen fertility than seed one, a bridge or a lagging chromosome was observed at the first anaphase in the PMCs and thus it suggested the presence of an inversion. Trisomic lines which have descended from the progenies of atomic bombed rice plants were not treated directly. However, some trisomics derived from reciprocal translocation heterozygotes, in which translocated chromosome had already identified, were examined for several generations as to the chromosome constitution, chromosome behavior at meiosis in the PMCs, the segregation mode of trisomic phenotypes in their progenies and so on. As a result, a general tendency was observed as follows: in early generations the chromosome constitution of trisomics which derived from translocation heterozygous plants were homo type II (primary trisomic interchange homozygote and tertiary trisomic) or hetero type (primary trisomic interchange heterozygote and tertiary trisomic interchange heterozygote) and the segregation of semi-sterile plants were observed in the hetero types; and with the generations, these chromosome constitution of trisomics became to the homo type I (primary trisomic and tertiary trisomic interchange homozygote) and finally fixed in this type which does not segregate the semi-sterile plants. In this process, the segregations of various morphological characteristics in primary or tertiary trisomics were observed according to the kind of extra chromosomes and their size of reciprocal translocations involved. Then, the endeavors to analyze the relation between the extra chromosome and the morphologica1 characteristics were made and the morphological characteristics of 11 sorts of primary trisomics were demonstrated. On the basis of those knowledge described above, the complicated occurrence of variant types and their genetic behavior in the process of pedigree culture of these progenies of atomic bombed rice plants more than 10 years period were discussed.