The survival of DNA, the most informative biological molecule, for periods of at least several thousand years in bone was demonstrated more than four years ago. However, difficulties with authenticating ancient DNA have made diagenetic studies problematic. It is therefore essential that these problems be overcome before the question of DNA survival can be addressed. Here we describe our work with a range of Holocene skeletal material from domestic animals and humans, and discuss how we go about authenticating the results. In the first instance, results should be reproducible between different laboratories, in order to eliminate the possibility of laboratory-specific contamination. The main risk is then contamination of the material prior to sampling for DNA. It is therefore important to have criteria whereby the authenticity of the results can be evaluated. These include amplification with species-specific primers to target DNA from domestic animals, sex determination in humans, and phylogenetic position in both. Since animals can be tested for the presence of contaminating human DNA, work with animals can be used to control and assess methodology. Our initial studies in this area suggest that more than 50% of skeletal remains from the past two thousand years are likely to contain amplifiable endogenous DNA, but that in the case of human material great care is needed to distinguish this from contamination introduced before the samples reach the laboratory.