In a recent Letter, Murata et al. [K. Murata, Y. Sugita, Y. Okamoto, Chem. Phys. Lett. 385 (2004) 1] claim that replica-exchange umbrella sampling is superior to conventional umbrella sampling, both in terms of sampling efficiency and accuracy, when used to compute the free energy of stacking in DNA dimers. This is in contrast to our previous results for the same system. Since Murata et al. use our reaction coordinate, a direct comparison is possible, and in this comment we argue that the discrepancies are more likely to be caused by the different parameter sets than by the sampling methods. (C) 2004 Elsevier B.V. All rights reserved.