The proton affinities of the 20 common amino acids have been computed at the G3MP2 level using structures derived from broad conformational searches at a variety of levels including G3MP2. In some cases, the conformational surveys identified more stable species than had been used in previous studies of proton affinities, though the differences in energy are sometimes rather small. The present values are likely the most reliable measure of amino acid proton affinities in the gas phase. An analysis of differences between these values and those obtained experimentally via the kinetic method indicates that the extraction of proton affinities from kinetic method data can potentially lead to large errors linked to the estimation of relative protonation entropies. . As noted by these authors, there is substantial variation in the experimental values reported for these species. The origin of this variability is most likely related to the difficulties of completing thermodynamic measurements in the gas phase on materials with low volatility. As a result, many of the experimental values are from kinetic method measurements and are subject to errors associated with the kinetic method approximation (as well as cumulative errors associated with the development of ladder-type scales). Therefore, in these systems, high-level computational work offers an attractive means of resolving differences in the various experimental measurements.While working on a computational study of the conformational preferences of neutral, gaseous amino acids, we realized that it would be possible to improve on the accuracy of the proton affinities reported by Paizs and coworkers. First, their computational strategy relied on the B3LYP approach to identify global minima and, often, they identified conformations for the neutral amino acids that did not match with those reported in previous computational and experimental studies (apparently in some cases this was the result of the search strategy and in others the computational method). For example, Paizs and coworkers concluded that glycine prefers a structure with an internal hydrogen bond between the carboxylic acid and amine groups. Highlevel ab initio, other B3LYP data [2], and experimental data suggest a structure with a syn carboxyl group and a weak interaction between the amine hydrogens and the carbonyl [3][4][5]. While the energetic impact of this problem is generally minor, we feel it is important to report values based on the most appropriate conformations that can be obtained within a computational method. For the protonated amino acids, this problem is less pronounced because the hydrogen bonding interactions are much stronger; most computational methods identify similar structures as the global minimum. Second, values reported by Paizs and coworkers are at 0 K. We have added a correction for the thermal energy of a proton, which amounts to about 1.5 kcal/mol at 298 K, as well as for the neutral and protonated amino acid. Although these corrections cancel to some extent and while the ge...