The problem of protein folding-unfolding is age-old and not fully solved yet. An amino acid sequence alone is insufficient to specify completely the local configuration, because nonspecific long-range forces tend to impart to the protein a different structure from what would result from covalent bonds only. The adsorption of polar molecules on proteins involves an energy of the same order as the one which stabilizes the native configuration, so that the solvation resulting from the adsorption is expected to modify progressively the protein configuration. The surface-reconstruction model proposed by Landsberg for the Elovich adsorption kinetics is modified to describe the adsorption equilibrium on reconstructable surfaces. In this model adsorption produces a progressive exposure of the protein skeleton to the atmosphere. The model can be worked to account for denaturation and the formation of the molten globule.