The postmortem stability of brain RNA was measured in rat and human samples for up to 48 hr. Whole rat brains, cooled at a rate approximating that of human brains awaiting autopsy, were collected at intervals from 0 to 48 hr after death and frozen. These samples were compared with freshly obtained, unfrozen rat brains. Two potentially independent characteristics of these RNA populations were measured: the recovery or yield of RNA/gram tissue (quantity) and the integrity or extent of degradation (quality). Total RNA yields were similar after all postmortem delays. Hybridization of [32P]-labeled cDNA probes to nitrocellulose filter blots of electrophoretically separated rat brain RNA failed to reveal degradation of the specific rat brain mRNAs during the postmortem period. Similarly, in vitro translation of these same rat total RNA samples produced high molecular weight translation products with no differences between long and short postmortem times. Human cerebral cortex RNA was prepared by the same methods as those used for rat brain from a neurosurgical sample and four other donors with postmortem intervals from 7 to 36 hr. Typically, the yield of total RNA from human brain was 40-50% of the yield from rat brain. When analyzed by RNA gel blot hybridization studies, as for rat brain RNA, human cortical RNA appeared slightly degraded. However, the degree of apparent RNA degradation was not related to the postmortem interval. In vitro translation products of human cortical RNA revealed high molecular weight peptides at all postmortem intervals, but slightly less [35S]incorporation into these bands was found at the longer postmortem intervals relative to the shorter times. Together, these results demonstrate an extensive stability of brain RNA that invites aggressive use of molecular genetic techniques for the study of human neurodegenerative diseases.