This work began 1995 with the realization that the susceptibility of step-scan measurements to noise is largely due to very low spectral multiplexing frequencies. The low modulation frequencies of step-scan operation are a consequence of the small average mirror velocity. This work represents the opposite extreme - scanning an interferometer mirror very rapidly to produce interference signals nearly 1000 times higher in frequency than those used with pyroelectric detectors, and nearly 1,000,000 times higher than those used in step-scan measurements. The target of this work is to scan a 4 cm^-1 spectrum every 1 ms. It appears that we will eventually be able to at least double this rate.