Subtle differences in Fe-C-O geometry between the recombinant (pH 9, space group P6) and native (pH 7, space group P21) crystal forms of carbon monoxide sperm whale myoglobin are magnified by using n-butyl isocyanide as a probe of distal pocket stereochemistry. Crystal structures of methyl, ethyl, n-propyl and n-butyl isocyanide bound to recombinant wild-type myoglobin were determined and then compared with the corresponding native myoglobin isocyanide structures determined by Ken Johnson (Ph.D. Dissertation, Rice University, 1993). The wild-type structures show all the ligands pointing inward and the distal histidine in the "down" conformation, whereas the native myoglobin n-propyl and n-butyl isocyanide structures show the ligand side chain pointing outward causing upward movement of the distal histidine. In order to examine the causes of these differences, the structures of Mb n-butyl isocyanide in native P21 crystals and recombinant V68F P6 crystals were determined at both pH 7 and 9. The results show that pH and not crystal packing is the key determinant of ligand orientation. Neutral or low pH favors rotation of the distal histidine and outward movement of the side chain of bound n-butyl isocyanide. Four mutant Mb structures were determined to examine the roles of the distal histidine and valine residues in governing ligand orientation. The results demonstrate that His64 (E7) sterically restricts outward movement of the isocyanide side chain since its replacement with alanine (H64A) or an increase in its mobility (F46V) causes the appearance of the "out" ligand conformation. Steric crowding in the protein interior by the V68F mutation is not sufficient to cause the ligand to point outward at pH 9.0. Since the rate and equilibrium constants for myoglobin n-butyl isocyanide show only a small pH dependence, the "in" and "out" conformations appear to have similar free energies. This suggests that ligand movement into the protein interior and outward through the distal histidine gate have similar probabilities which is what has been observed in laser photolysis experiments with oxymyoglobin (Scott, E. E. and Gibson, Q. H. (1997) Biochemistry 36, 11909--11917).