Changes in protein stability are commonly reported as changes in the melting temperature, Delta T-M, or as changes in unfolding free energy at a particular temperature, Delta Delta G degrees. Using data for 866 mutants from 16 proteins, we examine the relationship between Delta Delta G degrees and AT(M). A linear relationship is observed for each protein. The slopes of the plots of Delta T-M vs Delta Delta G degrees for different proteins scale as N-1, where N is the number of residues in the protein. Thus, a given change in Delta G degrees causes a much larger change in T M for a small protein relative to the effect observed for a large protein. The analysis suggests that reasonable estimates of Delta Delta G degrees for a mutant can be obtained by interpolating measured values of T-M. The relationship between Delta Delta G degrees and Delta T-M has implications for the design and interpretation of high-throughput assays of protein-ligand binding. So-called thermal shift assays rely upon the increase in stability which results from ligand binding to the folded state. Quantitative relationships are derived which show that the observed thermal shift, Delta T-M, scales as N-1. Hence, thermal shift assays are considerably less sensitive for ligand binding to larger proteins.