DNA is the source of the genetic information. Population geneticists and sociobiologists are currently rediscovering this basic fact of life. They are finding, however, that those DNA sequences that contain the least information for the cell bear the most information for them. These sequences in the eukaryotic genome are included in those classes of DNA which do neither harm nor good and which have therefore been termed 'junk DNA', though 'ignorant DNA(') would be a more appropriate name for it.
The exploitation of such DNA regions started with the discovery of mini satellite^(^'^). These are stretches of DNA, 30-50 nucleotides in length, which are tandemly repeated dozens to hundreds of times at many genomic locations. Some of them share a 'core sequence' of similar nucleotide composition, which is the reason why they can crosshybridize with a single probe taken from one of the regiond2). However, there is probably no limit to the number of potential minisatellite variants and new ones are continuously being detected(4). Such minisatellite stretches are hypervariable in length probably because they are subject to unequal crossing over('). Unequal crossover takes place during recombination when homologous chromosomes are aligned and since minisatellite regions are internally repetitious, alignment may occur within any of two repeat units with the result that the recombination products will differ in length. Minisatellite regions can be analysed by a procedure which is technically relatively demanding, but can yield the kind of data that population biologists have always dreamed of. The variability of minisatellite loci is such that closely related individuals will share certain variants, while unrelated individuals of a population will not. This allows one to detect such things as first or second degree relationships, non-mate relationships or multiple paternities, and the degree of inbreeding within populations.
It is not surprising then that many population biologists are eager to apply the technique for analysing these sequences -which was soon dubbed 'genomic fingerprinting' -to their problem. However, in such attempts, frustration may quickly outweigh the enthusiasm. The problems are of both a technical and a conceptual nature. Minisatellite probing involves the