[en] In recent years, both passive- and active-neutron nondestructive assay (NDA) systems have been used to measure the uranium and plutonium content in 200-ell drums. Because of the heterogeneity of the wastes, representative sampling is not possible and NDA methods are preferred over destructive analysis. Active-neutron assay systems are used to measure the fissile isotopes such as ²³⁵U, ²³Pu, and ²⁴¹Pu; the isotopic ratios are used to infer the total plutonium content and thus the specific disintegration rate. The active systems include 14-MeV-neutron (DT) generators with delayed-neutron counting, (D,T) generators with the differential die-away technique, and ²⁵²Cf delayed-neutron shufflers. Passive assay systems (for example, segmented gamma-ray scanners)5 have used gamma-ray sessions, while others (for example, passive drum counters) used passive-neutron signals. We have developed a new passive-neutron measurement technique to improve the accuracy and sensitivity of the NDA of plutonium scrap and waste. This new 200-ell-drum assay system combines the classical NDA method of counting passive-neutron totals and coincidences from plutonium with the new features of ''add-a-source'' (AS) and multiplicity counting to improve the accuracy of matrix corrections and statistical techniques that improve the low-level detectability limits. This paper describes the improvements we have made in passive-neutron assay systems and compares the accuracies and detectability limits of passive- and active-neutron assay systems