The problems associated with lethal gas gangrene infections that arise in connection with medical treatment are described and discussed, based on our own observations and on cases reported in the literature. Special emphasis is placed on questions relating to causality and responsibility. The difficulties encountered in the evaluation of such "iatrogenic" gas gangrene infections are due to the fact that clostridia occur ubiquitously, i.e. that they also reside in the human organism, for instance on the skin. For this reason a positive bacteriological diagnosis of gas gangrene does not necessarily mean that a clinically relevant infection and/or disease state is present at the same time. The implications for morphological practice arising from this situation are discussed, especially with regard to the significance of the intravital, agonal or postmortem bacteriological diagnosis of gas gangrene. Questions concerning the obligation to report such cases when encountered during forensic autopsies are also addressed.