The previous contention by several groups that double peaks that arise in NMR spectra from NH and a-CH protons of poly L-alanines are caused by the coexistence of helical and random-coil structures in solution is disputed.Numerous reports have appeared showing that under certain circumstances many polypeptides exhibit NMR spectra with double peaks from NH and a-CH protons (2-8). The doubling of peaks was attributed initially by Ferretti to the coexistence of helical and random-coil structures in solution, in spite of the fact that dynamic measurements indicate that the average rate of the helix to coil transition (0.1-10 Asec) is extremely rapid on the NMR time scale (9-12). Such rates should lead to a single averaged peak. Explanations for the doubled peaks have been offered based on polydispersity (13), slow nucleation of the transition (14), and protonation of the peptide bonds (15). Scheraga and his coworkers (16) speculated that the a-helix peak should not be observable because of dipolar broadening. They suggested that double peaks arise from specific, slowly exchanging solvations of random-coil species in mixed solvents containing acids such as trifluoroacetic acid (CF3COOH). Recently, Bradbury and his associates (17) assigned the double peaks to the a-helix and random coils on the basis of comparisons of the areas for a-CH with the side-chain peaks. The Bradbury group concludes that rapid exchange between a-helix and random coil takes place, with the double peaks arising from polydispersity. NMR spectra of poly(L-alanine) were analyzed by Jardetzky (3), Bradbury (18), and Ferretti (3) and their associates. The second two groups found double peaks for the a-CH at 220 MHz. One peak at 4.5 ppm was attributed to the a-helix, while the other peak at 4.7 ppm was assigned to the random-coil conformation of the polypeptide chain.In this paper, we present results indicating that the double peaks seen for poly(L-alanine) may not arise from the helix to coil transition, and that the above explanations may not apply to the observed phenomena. We believe the lowfield peak arises from the a-CH of oligomeric peptides. The high-field peak represents a single averaged peak of helix and random coil. We performed measurements using a 220-MHz Varian spectrometer and very low concentrations of L-alaninederived peptides, necessitating time-averaging to improve the signal-to-noise ratios in solvents composed of deuterochloro- form (CDC13) and CF3COOH. We varied the molecular weight of the polymers, sample concentrations, solvent compositions, and temperature at which the experiments were performed. Although the spectra do not exhibit good signal-to-noise ratios (even after time-averaging), we were able to obtain clear and interpretable results.Double peakst at about 4.6 and 4.3 ppm are observed for poly(L-alanine)i-[average degree of polymerization (DP)37], as is shown in Fig. 1