Recordings from a low magnification (V = 4), intermediate period (T ≈ 5 sec) seismograph at Mt. Hamilton, within 35 km of the San Andreas fault, show about 8 sec of P-wave energy and the first half-cycle or so of the initial S wave before going off-scale. The times, polarities, and overall amplitudes are consistent with Bolt's conclusion (1968) that the main shock began closer to San Francisco than Olema, as originally proposed by Reid (1910). This and the distribution of surface slip imply a bilateral rupture, but the seismic moment for the segment to the northwest was on the order of 2.5 times greater than for the southeasterly segment. Although Mt. Hamilton is only 35 km from the rupture surface, the most massive faulting apparently took place at least 75 km away. The recording at Mt. Hamilton returned to scale after about 60 sec; the duration of energy with periods close to 5 sec was comparable to that from more recent strong-motion recordings.
Theoretical modeling using both body and surface waves showed that the surface waves dominated the motion at Mt. Hamilton. The modeling also emphasized the sensitivity of the ground motion to directivity effects and rupture velocity. The general characteristics of the data (polarity, amplitude, period content, and duration) were matched reasonably well by a simple dislocation model using fault lengths, depths, and offsets determined from independent data.