The Bucaramanga earthquake nest is situated at a depth of about 160 km beneath Colombia at 6.8° N, 73° W, and it produces about eight earthquakes of mb 4.7 or more each year from a source region having dimensions of about 10 km or less. In this study, we evaluate nest properties using reported locations, focal mechanisms, and moment tensors, as well as relocations determined from teleseismic P, pP, and PKP arrivals. We also introduce a novel method for determining the “volume” of a group of hypocenters and use this volume as a statistic for evaluating the accuracy of nest hypocenters. We have relocated 26 nest earthquakes occurring between 1964 and 1986 using the JHD method and carefully reread teleseismic P, pP, and PKPdf arrivals. The relocated earthquakes cluster within a region having a volume of 776 km3 and east-west, north-south, and vertical dimensions of 13 × 18 × 12 km, a smaller region than any found previously from analysis of teleseismic data. In practical terms, this represents an approximate upper bound on the accuracy of relative earthquake locations possible using teleseismic data and standard relocation methods. This compares with a volume of 11 km3 and dimensions of 5.1 × 3.4 × 3.8 km as determined by Schneider et al. (1987) from observations of microearthquakes collected over 16 days with a temporary local network. These dimensions are comparable to those of the rupture dimensions of the largest observed nest earthquakes, as determined from an analysis of their seismic moments (∼5 × 1024 dyne-cm). Furthermore, an evaluation of cumulative moment rate since 1964 indicates that strain rates within the nest volume must be between 10−13/sec and 10−11/sec. The b values (∼2.0) and β values (∼0.9) determined from available magnitudes and moments are remarkably high, suggesting that earthquakes of all observed sizes (3 × 1020 to 5 × 1024 dyne-cm) contribute significantly to the moment release. Focal mechanisms determined from teleseismic observations exhibit a consistent pattern, most having westward-dipping P axes and eastward-dipping T axes; this differs from the more varied pattern reported by Schneider et al. (1987) for microseismic earthquakes.