One of several possible explanations for the fact that the initial number of ionizing radiation-induced DNA double-strand breaks (dsbs) per cell per Gy appears to be approximately 5-10 times greater than the number of excess fragments produced in prematurely condensed chromosomes (PCCs) is that dsbs in DNA tightly associated with protein, such as might be the case for heterochromatin, are held together and not expressed as discontinuities in the PCC assay. To test this idea the breakage frequencies in chromosome 4, 19 and the heterochromatic and more euchromatic portions of the Y chromosome of non-cycling human fibroblasts were measured and compared over a wide range of doses by inducing PCC immediately after irradiation with Cs-137 gamma-rays. Even for doses up to 400 Gy no evidence was seen that Y heterochromatin or large fragments of it remained unbroken. The only significant deviation from the expected initial breakage frequency per Gy per unit length of chromosome, in the dose range where a genome average for all chromosomes could be obtained, was that observed for the euchromatic portion of the Y chromosome (Y p + cen-->q11.2) where breakage was nearly twice that expected. There was no correlation between breakage per unit length and the (A + T)/(G + C) ratio for these chromosomes or regions.