Back in the old celestial mechanics forum, I wrote about a simulation in GravitySimulator where I created a moon with twice the mass of Mercury, and set it in orbit around Venus, right at Venus's L1 distance. The moon orbits a few times and then gets left behind. I thought at first that Mercury might have been an escaped moon of Venus, but that didn't happen. What usually happens is that the escaped moon crashes into Venus. The first two or three time I ran the simulation with a 64s time step, the collision occurred within a few hundred years to a few thousand years. The last time I ran it (and it's still running after over a month now) there was no collision. First, the moon settled into a stable pattern with an eccentric interior orbit that precessed slowly. Then a crisis occurred, and the moon got boosted to an exterior orbit, into a new stable attractor, this one with a stable orbit with a perihelion near Venus' orbit, and aphelion at just about exactly 1 AU, and now it's contracted a bit after year 200,000. This leads me to believe that there might be something to Bode's Law. Otherwise, that is a quite striking coincidence. So, my latest thinking is that Venus might quite likely had formed with a massive moon that escaped, and then collided with Venus. This theory would explain both the slow rotation of Venus, as well as the resurfacing event that happened around 800 million years ago. If the resurfacing event was of extravenusian origin, we might be able to find traces of it in Precambrian strata on Earth. Also, a tip when investigating collisions using GS. I used a small timestep so that each object would travel less than the diameter of Venus. But it occurred to me that one could ramp up the time step a bit if one artificially increased the diameters of the bodies in question, and thus cover more time per simulation attempt.