Measurements of the slope of the travel-time curve have been made on a group of 167 events in the distance range 27°-90° along an azimuth of 300°-320° from the LASA array in Montana. A first-order correction of +0.05 seconds/degree has been applied to the raw data to make them consistent with the travel-time residuals observed from the LONGSHOT nuclear test in the Aleutian Islands. The agreement between these two sets of data is sufficiently good that other station corrections at the array are assumed to be negligible. The resulting graph of dT/dΔ against Δ is shown to be consistent with other traveltime and amplitude observations. Using an assumed crust and upper mantle structure, a velocity-depth model for the mantle has been found which agrees, within the scatter of the data, with both the dT/dΔ observations and the LONGSHOT residuals. This model shows three anomalous regions within the mantle, at depths of about 800, 1300, and 2000 km, where the velocity changes very slowly with depth. In each case the possibility that the velocity decreases with depth exists, but is not proven by the present data.