When a compound that is removed from the body by metabolism produces toxicity in extrahepatic organs directly, rather than via active metabolites, induction or inhibition of the drug-metabolizing enzymes simply will decrease or enhance, respectively, the toxic effects of the compound. On the other hand, the effects of chemicals whose toxicity depends on their activation by metabolism may be modified in a complex way by pretreatment with inducers or inhibitors of the enzymes; it may, therefore, be impossible to predict the effect of pretreatment on the metabolism and toxicity of a given compound. The major sources of such complexity are that (a) inducers and inhibitors can have multiple effects on pathways of drug toxification or detoxification, both in the liver and in extrahepatic tissues, (b) active metabolites can be formed both in the liver and at extrahepatic sites, and they may not be sufficiently stable for transport from one site to another, and (c) regardless of the effect of pretreatment on pathways of extrahepatic metabolism, the accompanying effects on hepatic metabolism may determine the extrahepatic distribution and site of metabolism (in vivo) of a protoxin or its active metabolite(s). A review of studies on pulmonary toxicity produced by three agents--monocrotaline, bromobenzene, and 4-ipomeanol--illustrates several of these problems, and also shows the value of using inducers and inhibitors in the experimental analysis of extrahepatic toxicity produced by reactive metabolites.