A linear array of eight Caltech portable broad-band seismograph trailers was set out from NTS to near Phoenix, Arizona, for the pre-announced underground nuclear test, OSCURO, on September 21, 1972. Travel-time and amplitude information were used to find an average crustal model by calculating synthetic seismograms using the Cagniard-de Hoop method. Rayleigh waves from other nuclear events at NTS, as recorded at the Tucson WWSSN station, were examined as a control for determining the structure of the top half of the crust. Group-velocity curves were found and synthetic Rayleigh waves calculated for Tucson and Kingman (LRSM). The formations and characteristics of Pn, a reflected Pn, and the Pg phases are examined. Pg is demonstrated to be composed of the primary P reflection from the mantle and contains multiple arrivals of P—SV conversions. Comparisons of synthetic and observed seismograms indicate a crustal thickness of 30 km with a Poisson's ratio of 0.23. The crust-mantle transition appears to be sharp, jumping from 6.1 to 7.9 km/sec. The amplitude behavior of Pn shows little evidence of any lid structure.