This paper studies the concept of extracting hover performance from model rotor climb data by extrapolating to the limit of zero climb speed.  A 2-bladed rotor was mounted horizontally and  tested in the 30' x 31' settling chamber of the Ames 7' x 10' #1 wind tunnel. The collective pitch and tunnel speed were varied. CCD video cameras were used to visualize the flowfield illuminated by pulsed white light sheets. Facility recirculation effects were eliminated at all but the lowest rate of climbs. With a steady, non-recirculatory flow, the rotor wake was seen to be fully periodic, with little diffusion and dissipation of the vortices for several rotor cycles. The transition to the far wake occured through a periodic pairing of the tip vortices, followed by their merging into a single diffuse vortex for each rotor cycle. The number of discrete vortex turns in the near wake before the pairing varied with the thrust coefficient and rate of climb. The climb-extrapolation method appears to be a reliable and practical approach to obtaining performance data which are free of recirculation or facility effects.