The folding kinetics of proteins is frequently single-exponential, as basins of folded and unfolded conformations are well separated by a high barrier. However, for relatively short peptides, a two-state character of folding is rather the exception than the rule. In this work, we use a Zwanzig-type model of protein conformational dynamics to study the dependence of folding kinetics on the protein chain length, M. The analysis is focused on the gap in the eigenvalue spectrum of the rate matrix that describes the protein's conformational dynamics. When there is a large gap between the two smallest in magnitude nonzero eigenvalues, the corresponding relaxation times have qualitatively different physical interpretations. The longest of these two times characterizes the interbasin equilibration (i.e., folding), whereas the second time characterizes the intrabasin relaxation. We derive approximate analytical solutions for the two eigenvalues that show how they depend on M. From these solutions, we infer that there is a large gap between the two, and thus, the kinetics is essentially single-exponential when M is large enough such that 2 M+1 is much larger than M 2 . ' INTRODUCTION The kinetics of many long proteins is single-exponential. This implies that folding of such proteins is a two-state process that involves transitions between the folded and unfolded states of the protein, which are separated by a high barrier. However, this is not necessarily the case for short peptides. The goal of the present work is to study how the character of the folding kinetics (i.e., whether the kinetics is single-exponential or not) depends on the protein chain length. We analyzed this question within the framework of a simple model of the protein conformational dynamics that is similar to the model suggested by Zwanzig, Szabo, and Bagchi to study Levinthal's paradox. 1À4