It is argued on the basis of MNDO-PSDCI calculations that, at the wavelength of electronic dichroism studies used to infer structure, the transition dipole moment of the retinyl Schiff＇s base chromophore of bacteriorhodopsin makes an angle of 10.5 +/- 3.5 degrees with respect to the long axis of the chromophore. The magnitude and direction of this off-axis angle helps to reconcile the differences between the deuterium NMR and optical determinations of the angular orientation of the chromophore in bacteriorhodopsin. Reconciliation of the NMR and optical dichroism structures is only possible, however, when the chromophore is oriented so that the imine proton points toward the extracellular surface. After correction, the published electronic dichroism data predict a chromophore angle of Ohm = 61 +/- 7 degrees, where Ohm is defined as the angle of a line connecting chromophore atoms C-5 and C-15 relative to the membrane normal. The corresponding NMR results are Ohm = 53.7 +/- 4.1 degrees. The combined data with equal weighting yield Ohm = 57 +/- 8 degrees. This estimate is smaller than Ohm = 70.6 +/- 3.2 degrees derived from recent diffraction studies, and the possible origins of these differences are discussed.