Since the technique of PCR was developed in 1985 (Saiki et al). many new forms of applications have been described in the literature. Of particular interest are PCR approaches, allowing amplification of unknown sequences. In this review these approaches are generally termed "Random PCR"s. While "conventional PCR" in the form that was first described by Saiki et al. is utilized for the amplification and subsequent detection of specific DNA sequences, which are precisely characterized in length and sequence, Random PCR is either used for universal amplification of prevailing DNA or for amplification of unknown intervening sequences which are not generally defined in length or sequence. We depict criteria for discrimination between conventional PCR and Random PCR. Furthermore, we have compiled a classification system for Random PCR approaches based on differentiation by primer structure (degenerate--non-degenerate). According to this classification system a general overview of published Random PCR approaches is given. Future aspects of application of Random PCR are mentioned and commented. Own investigations, combining Random PCR with the technique of chromosome microdissection and thus allowing characterization of unknown chromosomes or chromosome fragments via FISH, are presented. Results are briefly summarized.