The rabbit was bled to activate the erythropoiesis, and the serum was stocked for culture. After two days, non-adherent bone marrow cells of the same rabbit were havested, and cultured consequently with auto-plasma and auto-serum, which had been prepared, alpha-medium, erythropoietin (Ep) and spleen extract derived from rats sublethally irradiated 8 days before. The FB-16-24-TC (Limbro Co.) was used for culturing bessel by the addition of round cover-glass on the bottom of each well. The preparations were made by the routine Giemsa and hemoglobin stain methods on 3rd, 5th and 9th day of culture. The erythroid and non-erythroid colonies were calcurated under the microscope with x200 or higher magnifications. Small erythroid colonies consisted of 8 approximately 16 cells were scattered on the 3rd day, and undifferentiated or granulocytic colonies also frequently appeared on the specimen. On the 5th day, medium size colonies consisted of 32 approximately 64 cells were observed. The most effective irradiation for active extract was 700R. On the other hand, if the culture medium did not contain the spleen extract, the number of colonies decreased. Other results showed that this extract have hardly any enhancement on colony formation without Ep in the medium. Therefore, it is probable that the extract may have contained the potentiators for the erythroid, as well as for the myeloid progenitors. Whether these spleen factors are the same or not as that reported by Knospe et al. or Morley et al. is unknown. The study is in progress.