Abstract:The puff-like structure (PLS) is an intranuclear structure often observed in cells of higher plants. It is a spherical and fibrillar structure with a configuration similar to the puff of polytene chromosomes. A number of studies upon the PLS have been done and different hypotheses regarding the nature of the PLS remain. The authors describe the number, volume and distribution of the PLSs in the interphase nuclei of Allium cepa L. based on electron microscopical (EM) observation of serial sections and 3-D reconstruction of complete series for 6 nuclei. The results are as follows. 1. The number of PLSs differed from the diploid number (2n= 16) of A. cepa. In S phase nuclei the number of PLSs is 4 to 10 and in G2 phase nuclei the number is 4 to 9. 2. Volume variation occurred among the individual PLSs and most of the structures measured from 0. 1 to 0.4μm3. The total volume of PLSs in individual G2 phase nuclei varied considerably and in some G2 phase nuclei the total volume of PLSs is much smaller than that in some S phase nuclei. 3. The PLSS were clustered in certain area of the nucleus. The authors previously reported that: (1) Radioactivity was found in PLSs after they were pulse-labelled with tritiated uridine; the PLS contained cytochemically distinct regions in some of which RNP was predominantly present; (2) A brushlike structure consisting of an axial fibre and lateral fibrils was observed in the PLS; after treatment of a-amanitin, the brush-like structure disappeared and some of the PLSs were destroyed ultrastructurally. These results indicate that the PLS is a transcriptional structure of chromatin instead of an inclusion of nucleolar transcripts as suggested by some authors. Some other authors proposed that the PLS corresponds to the centromere in interphase nuclei. The centromere is a special region of a chromosome. It is uncertain at the present time how the Ultrastructure of a centromere changes during cell cycle and we know even less about how a centromere functions in interphase. According to the hypothesis mentioned above, the total volume of PLSs in individual G2 phase nuclei should be consistent with each other in the rough and the total volume of PLSs in a G2 phase nucleus should not be remarkably smaller than that in a S phase nucleus. Our results in the present study, however, are opposite to this hypothesis and we believe that the hypothesis should be further tested.