We have inverted the teleseismic P waveforms recorded by stations of the Global Digital Seismograph Network for the 8 July 1986 North Palm Springs, California, the 28 October 1983 Borah Peak, Idaho, and the 19 September 1985 Michoacan, Mexico, earthquakes to recover the distribution of slip on each of the faults using a point-by-point inversion method with smoothing and positivity constraints. In the inversion procedure, a fault plane with fixed strike and dip is placed in the region of the earthquake hypocenter and divided into a large number of subfaults. Rupture is assumed to propagate at a constant velocity away from the hypocenter, and synthetic ground motions for pure strike-slip and dip-slip dislocations are calculated at the teleseismic stations for each subfault. The observed seismograms are then inverted to obtain the distribution of strike-slip and dip-slip displacement for the earthquake.