We have compared two statistical methods of estimating the time to most recent common ancestor (TMRCA) from a sample of DNA sequences, which have been proposed by Templeton (1993) and Bandelt et al. (1995). Monte-Carlo simulations were used for generating DNA sequence data. Different evolutionary scenarios were simulated and the estimation procedures were evaluated. We have found that for both methods (i) the estimates are insensitive to demographic parameters and (ii) the standard deviations of the estimates are too high for these methods to be reliably used in practice.