Additional restrictions on exposure to ethylene oxide, an effective and widely-used sterilant, have recently been proposed (EPA, 1978). Substantial amounts of evidence indicate that ethylene oxide is mutagenic. The data from sub-mammalian systems, including demonstrations of point and chromosomal mutations in bacterial and eukaryotic systems, and from a limited number of studies in mammalian systems are reviewed. The accumulated information, and its context of current guidelines for mutagen testing, are considered in terms of extrapolation to evaluation of potential risk to exposed humans. The available evidence does not appear sufficient to condemn ethylene oxide as a human mutagen at currently acceptable exposure levels.