Hi Say I have a Gaussian velocity distribution for a collection of particles. Then I make a change to the setup in a way that the velocity distribution changes. This change is such that a part (not all) of the distribution gets bunched into one particular velocity class lower than the mean of the Gaussian. Does this constitute cooling of the atoms as well as slowing? I would say no myself, as "cooling" is a collective parameter of all the atoms, not just some of them. Best, Niles.