Maxwellian velocity distributions in slow time

We extend Maxwellian velocity distributions to long observational timescales in much the same way that short timescale statistical mechanics distributions are averaged to yield normal laboratory timescale thermodynamic distributions. This long timescale view has several novel effects: Fluctuating overall velocities (i.e. wind) thermalizes into an additional component of temperature, while returning a Maxwellian velocity distribution. However, fluctuating temperature results in a new distribution with a Gaussian core but heavy polynomial tails. The power of the polynomial tail is either -3 or -2 depending on whether the precision of the temperature is allowed to extend to +/- in infinity or is required to remain strictly positive. The distribution is also interesting in the way it remains almost exactly Gaussian up to a certain velocity after which it quickly breaks off to become polynomial. The distributions are carefully analyzed mathematically, and physical consequences are drawn.