The “thermodynamic” hypothesis of protein conformation (1) states that the sequence of amino acids characteristic of a given protein is sufficient to determine its secondary and tertiary structure, for the molecule will assume the conformation thermodynamically most stable. In an analogous vein, the “conformation” hypothesis (2) proposes that the steric and noncovalent bonding properties of poly-cu-amino acid side chains dictate the tendency of these polymers to assume helical, extended, or random coil structures. These hypotheses need little alteration in order to include the possibly significant role of small cofactors and prosthetic groups. Such small molecules may interact with a protein in a specific fashion and thereby influence the relative stability of various folded states. Myoglobin and apomyoglobin offer an attractive system for evaluating the above hypotheses. A comparison of conformation-dependent parameters of the native and apoproteins will reveal what role, if any, is played by the heme moiety in the folding of myoglobin, while a study of reversible conformational changes of apomyoglobin will reflect the adequacy of the amino acid sequence in dictating the structure. The absence of sulfhydryl side chains or disulfide bridges in these proteins is particularly important. The elegant work of Anhnsen on the reversible unfolding of ribonuclease (3), which has placed the “thermodynamic” hypothesis on a secure experimental basis, depends on the oxidation of reduced disulfide bonds. It is thus not capable of distinguishing the role of the amino acid sequence in general from that of half-cystines in particular. The absence of this complicating factor in the myoglobin-apomyoglobin system therefore provides considerable simplification of interpretation. As an indication of conformational changes, we have used optical rotatory dispersion. Recent advances in our understanding of this technique as a measure of a-helical secondary structure (4-11) and recent improvements in instrumentation (12) have increased our confidence in the use and interpretations of ordinary rotatory dispersion. In this paper we report data which indicate (a) that a small loss in helix content, which is recovered on reconstitution, accompanies removal of the heme prosthetic group from myoglobin; (b) that the loss of helical structure of apomyoglobin that occurs on addition of urea can be completely reversed by dialysis against water or buffer; and (c) that apomyoglobin, restored in this manner to its native conformation, is capable of combining with heme to form myoglobin.