It would appear that it has become almost common practice to regard arachidonic acid (AA) as the sole precursor of eicosanoids. The fact that both dihomogamma-linolenic acid (DGLA) and eicosapentaenoic acid (EPA) give rise to distinct families of eicosanoids is commonly almost completely ignored. Elevated tissue levels of AA eicosanoids have been found in and have been implicated in the etiology of a number of diseases. Drugs which selectively block AA mobilization or its eicosanoid metabolism have therefore been developed for therapeutic use in these conditions. The fact that such drugs will also simultaneously block the eicosanoid metabolism from DGLA as well as from EPA is also commonly ignored. It is suggested that the profoundly adverse side-effects displayed by some of these drugs, resulting in some instances in their withdrawal from use, could be the direct result of their concomitant action of interfering with the eicosanoid metabolism of DGLA and EPA. It is further suggested that, before the interactions between the eicosanoids derived from AA and those derived from DGLA and EPA are understood, the use of drugs for the manipulation of AA eicosanoid metabolism in isolation, could be hazardous. This implies that all such drugs currently in use are to be regarded as experimental and provisionally toxic in terms of their effects on the whole system of eicosanoid metabolism. Thus even drugs which have been passed by the FDA and similar Drug Control Councils require total re-evaluation especially in view of the fact that the non-steroidal anti-inflammatory drugs are often prescribed for chronic conditions which require therapy for several years.(ABSTRACT TRUNCATED AT 250 WORDS)