Based on a hypothesis that post-mortem cellular (chiefly nuclear) changes in the white blood cells could reliably be correlated with the time interval since death, (ti), serial observations were made on the counts (total, differential) and light-microscopically observable 'degenerations' of white blood cells obtained from 30 non-refrigerated cadavers (experimental group) and similar cells obtained from 200 hospital patients (control group). While neutrophils degenerated rapidly, lymphocytes did so slowly; the eosinophils and monocytes degenerated at rates between these extremes. In cadaveric blood total counts of identifiable leucocytes on average dropped to zero by 84 hours, identifiable eosinophils and monocytes were first to 'disappear' (by 60 hours), followed by neutrophils (by 66 hours), and finally lymphocytes: identifiable lymphocytes disappeared completely at or around 84 hours from the time of death. This 'differential degeneration' was surprising but useful. Based on the use of all four characteristics--total and differential white cell counts, differential degeneration and morphology of cells--a method for a reasonably exact estimation of ti is presented. The method is appropriate for ti up to 84 hrs (3 1/2 days). Zero white cell counts (total, differential) and bizarre morphology (unidentifiable white blood cells) indicate a ti > 84 hrs. Avenues for further research are indicated.