The location of the subsurface top of the chimney formed by the collapse of the cavity resulting from an underground nuclear explosion is examined at five sites at the Nevada Test Site. The chimneys were investigated by drilling, coring, geophysical logging (density, gamma-ray, caliper), and seismic velocity surveys. The identification of the top of the chimney can be complicated by chimney termination in friable volcanic rock of relatively high porosity. The presence of an apical void in three of the five cases is confirmed as the chimney horizon by coincidence with anomalies observed in coring, caliper and gamma-ray logging (two cases), seismic velocity, and drilling. In the two cases where an apical void is not present, several of these techniques yield anomalies at identical horizons, however, the exact depth of chimney penetration is subject to some degree of uncertainty. This is due chiefly to the extent to which core recovery and seismic velocity may be affected by perturbations in the tuff above the chimney due to the explosion and collapse. The data suggest, however, that the depth uncertainty may be only of the order of 10 m if several indicators are available. Of all indicators, core recovery and seismic velocity indicate anomalous horizons in every case. Because radiation products associated with the explosion are contained within the immediate vicinity of the cavity, gamma-ray logs are generally not diagnostic of chimney penetration. In no case is the denisty log indicative of the presence of the chimney. ?? 1993.