"Underground nuclear detonations at the Nevada Test Site (NTS) provide a unique opportunity to study the behavior of radionuclides in a field environment. The 1965 Cambric event, a small detonation in the saturated zone of the Valley Fill alluvial aquifer at NTS, has been studied extensively to improve the understanding of groundwater flow and radionuclide migration. The 1974 field investigation of the Cambric event included: (1) a re-entry boring into the original detonation cavity and (2) the installation of a pumping well 91 meters away from the cavity to draw mobile radionuclides such as tritium and krypton-85. This satellite well, sampled regularly over 16 years, produced some unexpected results. Because of the supposed ideality of tritium and krypton as tracers, their arrival was expected to be concurrent; however, krypton-85 arrival was delayed compared to tritium. More important is the fact that, while over 90% of the decay-corrected tritium from the original blast was drawn out through the well, less than half of the krypton was recovered. This work presents a hypothesis for understanding these unexpected results based on the site hydrogeology and underground detonation phenomenology. Carbon dioxide evolved from caliche during the detonation appears to have caused the upward migration of krypton-85 and other gaseous radionuclides through the collapsing chimney and into the unsaturated zone above. In the process, krypton became distributed throughout the cavity and chimney, while tritium remained immediately around the cavity. This post-detonation movement of krypton-85 provides one explanation, then, of the unexpected elution curve results. To demonstrate the validity of this hypothesis, a variety of numerical simulations were performed using the finite difference code TRACR3D. Calibrating two unknown hydraulic conductivity parameters on the tritium field data, it was shown that krypton-85 may indeed have been distributed throughout the chimney as suggested by the hypothesis."--Abstract.