Rob Matson's calculation needs a bit of modification. He suggests the satellite will reflect the Sun directly to an observer on Earth sixty times per revolution. That turns out not to be the case; the correct number is less than one tenth of that - nearer five times than sixty. I understand where the error comes from; I made the same error myself. If the equator of the ball is studded with mirrors it will take sixty to go 'round. If the observer and the Sun are in the equatorial plane then there will indeed be sixty flashes per revolution. An observer half a degree outside the equatorial plane, however, won't see any flashes. Let's redo the calculation a bit. One thousand half-degree diameter beams would cover 200 square degrees*. All of space covers 41,253 square degrees**, so only about half a percent of the surface of the Earth is likely going to see a sun glint while observing the illuminated satellite. Thus the average magnitude of a satellite spinning with sufficient speed to have a flash period less than the flicker time would have an apparent magnitude almost six magnitudes down from the glint maximum. I didn't check that out, but it looks about right. In any event this object will be about mag five on time average, but as a much brighter glinter I expect it will be naked eye visible, if sporadic. Lageos, by the way, would not be bright except to an observer approaching it for the direction of the Sun. It is not covered with mirrors; those are retroreflectors. Leigh * 0.064 steradians ** 4 pi steradians