Helix-coil theory has proven to be extremely successful in describing the stability of cc-helices. However, less is known about the applicability of the nucleation-elongation model to the kinetics of a-helix formation. A recent nanosecond infrared temperature-jump study in a helix-forming peptide has revealed unprecedented kinetic complexity (Huang, C. Y.; Getahun, Z.; Zhu, Y. J.; Klemke, J. W.; DeGrado, W. F.; Gai, F. Proc. Natl. Acad. Sci. U.S.A. 2002, 99, 2788-2793). In this investigation, the apparent kinetic relaxation varies depending on the location of the spectroscopic probe on the molecule and on the magnitude of the perturbation (T-jump size). In an effort to rationalize these results, we have developed a detailed kinetic model of alpha-helix formation. The model is based on a simple nucleation-elongation description of a-helix formation and incorporates sequence dependence factors from a previous equilibrium implementation of helix-coil theory (AGADIR; Munoz, V.; Serrano, L. Nat. Struct. Biol. 1994, 1, 399-409). Combining the model and an elementary description of the peptide bond amide I spectrum, we successfully simulate the experiments of Huang et al. Analysis of the simulations reveals that the dependence of the kinetic relaxation on the location of the spectroscopic probe is a consequence of helix fraying at the ends. End fraying is a general property of the helix-coil transition, which, in this case, is further modulated by strong capping effects. The decrease in apparent relaxation time the larger the perturbation is also an indirect consequence of different distributions of helical lengths at each initial condition. Our results demonstrate that nucleation-elongation is a valid mechanistic description of the kinetic process of alpha-helix formation. Furthermore, from the comparison between the experiments of Huang et al. and our calculations we can conclude that the cost in nucleating a helix is small and that the 150-300 ns relaxation observed in many T-jump experiments corresponds, indeed, to reequilibration between the coil and helix ensembles. The success of our kinetic model of a-helix formation opens the possibility of making specific predictions that can be tested experimentally.