This series was done on 3 micron alumina sandpaper. On many occasions, the drop of water would grow in size until it resembled Jabba the Hut, at which point the surface tension would break and it would be absorbed by the spinning film of water underneath it. In contrast with the pictures below, the drop was anchored to the hose by what I imagine is surface tension.
This series was done on 300nm alumina sandpaper. The surface was ten times smoother, and now the droplet would break free and slowly travel along the disc before collapsing after a maximum travel of four centimeters. For some reason the drop here had a high enough surface tension to travel on its own on a film of water directly under it! Water and hose conditions & flow were identical, so this could be a function of the surface roughness and/or any particles removed during polishing (I was polishing stainless steel).