The structures of a family of peptides that contain variants of the major alpha-helix of barnase (residues 6-18) have been analyzed in solution by circular dichroism (CD) and H-1 NMR at various concentrations of the helix-inducing cosolvent trifluoroethanol (TFE). The very low equilibrium constant (similar to 10(-2)) for the formation of helix in water, K-H2O, was estimated from titration of the helical ellipticity signal at 222 nm with [TFE] using the equation, K-TFE = K-H2O exp((m/RT) [TFE]/[H2O]), where K-TFE is the equilibrium constant for the formation of helix in TFE/H2O and m is characteristic for each peptide but appears to be proportional to the lengths of related helices. NMR studies show that the peptide is mainly random coil in water, but that the helix is induced cooperatively by TFE and extends from residues 6 to 18 for wild-type peptide in 35% TFE. The mutant Peptide Tyr-17-->Ala, however, has a helical region extending only for residues 9-15. Truncation of the helix upon mutation is also detected in the TFE titration procedure, which finds correspondingly lowered m-values upon mutation. This is is also supported by measurements of the pH dependence of K-H2O, which is caused by the ionization of the C-cap residue, His-18, whose pK(a) is raised by the interaction of the protonated form with the helix dipole. Whereas there is an apparent charge/dipole interaction energy of 1.1 kcal mol(-1) in the wild-type peptide, similar to that measured in the native protein, this drops dramatically upon mutations that disrupt the C-terminus of the helix. Mutation of Tyr-17-->Ala lowers K-H2O only slightly, as do the other helix-destabilizing mutations. The combined results show that the helix-weakening effects of mutations act here primarily by shortening the length of the helix, with smaller effects on the equilibrium constants between helix and coil (K-H2O). Such effects would not be detected by simple measurements on the percentage helicity of peptides measured by the amplitude of the CD signal at 222 nm. TFE appears to act by perturbing a preexisting equilibrium between unfolded and fully helical peptides since NMR studies at TFE concentrations spanning those for the helix-coil transition suggest a predominance of one folded state. This is to be expected since m increases with increasing helical length, and so TFE selectively stabilizes the longest helix. Data acquired in the presence of TFE give information on the longest helices, but there is evidence that shorter ones may exist in the absence of TFE.