One of the basic problems of the theory of control in complex systems involves determining the circumstances under which a set of control parameters can effectively manipulate a much larger number of degrees of freedom in the controlled system. Such a control capability obviously requires the presence of additional constraints relating the quantities to be controlled. We explore this general question in the context of a specific situation; namely, in terms of an algorithm recently proposed to compute the three-dimensional structures of folded polypeptides from their amino acid sequences. In this algorithm, a small number of average geometric properties (the control parameters) are utilized to determine a much larger set of inter-residue distances in the folded structure. We derive the specific condition which assures that the algorithm converges pointwise to the correct structure, which in this case turns out to involve relating rates of change of coordinates to increments in the distances consistent with a particular set of values for the control parameters. Some further implications of this kind of result are briefly discussed.