An early study which attempted to deduce the enthalpies and entropies of mismatched oligodeoxynucleotide pairs from studies of melting temperatures as determined by absorbance measurements. Unfortunately we discovered that some of the parameters they give are internally inconsistent (containing sign errors) and some are just plain nonsensical (e.g. some implying mismatches that make a strand stronger instead of weaker). We threw out the results from this paper in favor of the results in Werntges et al. '86.