This study presents improved thermodynamic parameters for predicting RNA duplex stability. The parameters were derived from enthalpy and free-energy changes for helix formation in 45 RNA oligonucleotide duplexes. The sequences were chosen to maximize the reliability of secondary structure predictions and minimize experimental errors in ΔG° at 37°C. These parameters predict melting temperatures of most oligonucleotide duplexes within 5°C, which is about as good as can be expected from the nearest-neighbor model. Free-energy changes for helix propagation at dangling ends, terminal mismatches, and internal G-U mismatches, and for helix initiation at hairpin loops, internal loops, or internal bulges are also tabulated. RNA duplex stabilities and secondary structures are often predicted using free-energy parameters from a nearest-neighbor model. However, predictions can sometimes be inconsistent with experimental data. One factor limiting successful predictions is the reliability of parameters, which is constrained by the availability of model oligonucleotides. Recent advances in RNA oligoribonucleotide synthesis allow for the design of oligonucleotides that provide improved parameters. This paper presents thermodynamic parameters derived from data on 45 complementary RNA duplexes. The parameters are able to predict the stabilities of RNA duplexes within the limits of the nearest-neighbor model. The thermodynamic parameters were determined from plots of t_m⁻¹ vs. log(C_T). The parameters were obtained by multiple linear regression on observed oligonucleotide thermodynamic parameters based on the nearest-neighbor model. Parameters derived from t_m⁻¹ vs. log(C_T) plots were used because they are the most reproducible between different laboratories. The parameters were validated by comparing predicted and measured duplex stabilities. The results showed that the parameters predict the stabilities of RNA duplexes with an average deviation of about 1.6°C for melting temperatures, 4% for ΔG°, and 7% for ΔH°. The predictive capability of the parameters is about equal to that expected for the nearest-neighbor model. The parameters were used to predict the secondary structure of RNA. The prediction of RNA secondary structure from sequence is a major application of the parameters. The parameters must be combined with thermodynamic parameters for hairpin loops, internal loops, bulges, mismatches, and other structures. Although parameters for such structures have been tabulated, they are correlated to the helix propagation parameters and should be recalculated with the parameters in this study. The parameters listed in Tables 3–6 are updated free-energy parameters for common RNA structures. These parameters are particularly unreliable and are marked with an asterisk. The parameters were used to predict secondary structures for 142 randomly chosen tRNA sequences, with 82% of the four major stems of the cloverleaf model predicted within 2 base pairs. The parameters in Tables 2–6 predict almost half of the major stems missed by theThis study presents improved thermodynamic parameters for predicting RNA duplex stability. The parameters were derived from enthalpy and free-energy changes for helix formation in 45 RNA oligonucleotide duplexes. The sequences were chosen to maximize the reliability of secondary structure predictions and minimize experimental errors in ΔG° at 37°C. These parameters predict melting temperatures of most oligonucleotide duplexes within 5°C, which is about as good as can be expected from the nearest-neighbor model. Free-energy changes for helix propagation at dangling ends, terminal mismatches, and internal G-U mismatches, and for helix initiation at hairpin loops, internal loops, or internal bulges are also tabulated. RNA duplex stabilities and secondary structures are often predicted using free-energy parameters from a nearest-neighbor model. However, predictions can sometimes be inconsistent with experimental data. One factor limiting successful predictions is the reliability of parameters, which is constrained by the availability of model oligonucleotides. Recent advances in RNA oligoribonucleotide synthesis allow for the design of oligonucleotides that provide improved parameters. This paper presents thermodynamic parameters derived from data on 45 complementary RNA duplexes. The parameters are able to predict the stabilities of RNA duplexes within the limits of the nearest-neighbor model. The thermodynamic parameters were determined from plots of t_m⁻¹ vs. log(C_T). The parameters were obtained by multiple linear regression on observed oligonucleotide thermodynamic parameters based on the nearest-neighbor model. Parameters derived from t_m⁻¹ vs. log(C_T) plots were used because they are the most reproducible between different laboratories. The parameters were validated by comparing predicted and measured duplex stabilities. The results showed that the parameters predict the stabilities of RNA duplexes with an average deviation of about 1.6°C for melting temperatures, 4% for ΔG°, and 7% for ΔH°. The predictive capability of the parameters is about equal to that expected for the nearest-neighbor model. The parameters were used to predict the secondary structure of RNA. The prediction of RNA secondary structure from sequence is a major application of the parameters. The parameters must be combined with thermodynamic parameters for hairpin loops, internal loops, bulges, mismatches, and other structures. Although parameters for such structures have been tabulated, they are correlated to the helix propagation parameters and should be recalculated with the parameters in this study. The parameters listed in Tables 3–6 are updated free-energy parameters for common RNA structures. These parameters are particularly unreliable and are marked with an asterisk. The parameters were used to predict secondary structures for 142 randomly chosen tRNA sequences, with 82% of the four major stems of the cloverleaf model predicted within 2 base pairs. The parameters in Tables 2–6 predict almost half of the major stems missed by the