Presumably you'd have to adjust the subtraction to correspond at least roughly to the relative bandpasses of the two filters and also to the relative exposures. Basically try to answer the question of how much of the red continuum from cool stars appears in the H-alpha bandpass and subtract that out, so you're left with only the emission. The other suggestion (red only) would attempt to remove the clumpiness in the appearance of the galaxy caused by the emission regions, and show just the light from the stars glowing at the H-alpha wavelength. Obviously this involves subtracting rather little of the total 'red' signal from the narrow slice of H-alpha. What you should get for typical spirals is a rather smooth diaphanous galaxy with the spiral structure smeared out---which will be in stark contrast to the H-alpha-only image. There will be additional features brought out that tells you something about how the star-formation proceeds and where. \Brian -- See message header for info on list archives or unsubscribing, and please send personal replies to the author, not the list.