We report the results of an empirical study of 256 paternity cases referred to 7 different German laboratories for DNA fingerprinting with oligonucleotide probe (CAC)5/(GTG)5. All parameters characteristic of such multilocus DNA fingerprints were found to differ significantly between the contributing centres. Despite these differences, clear-cut decisions between paternity and non-paternity could be made in all but one case. Furthermore, we found no systematic deviation of the gel-phenotype distribution among trios from random expectation as derived from commonly adopted analytical models. Thus, we conclude that oligonucleotide DNA fingerprinting is a robust and reliable means for the resolution of paternity cases.