A few years ago, http://home.earthlink.net/~mellowtigger/atmosphere.html [Broken] considering what gaseous molecules a planetary body could expect to retain or lose based on the mass of the planet. At the time, I found a claim that any molecule whose average velocity (at ground level) was one-sixth of the planet's escape velocity would likely escape the planet's atmosphere over geologic timescales. Can anybody point me to a reference that explains the calculation behind this 6X rule of thumb? My reliance on "I found it on the web, so it must be true" verification just doesn't sit very well with me, especially since I intend to duplicate the calculations now for Mars. I'd like to be a little more thorough this time.