Data on the construction characteristics and the composition of influent and effluent at 13 underground, limestone-filled drains in Pennsylvania and Maryland are reported to evaluate the design and performance of limestone drains for the attenuation of acidity and dissolved metals in acidic mine drainage. On the basis of the initial mass of limestone, dimensions of the drains, and average flow rates, the initial porosity and average detention time for each drain were computed. Calculated porosity ranged from 0.12 to 0.50 with corresponding detention times at average flow from 1.3 to 33 h. The effectiveness of treatment was dependent on influent chemistry, detention time, and limestone purity. At two sites where influent contained elevated dissolved Al (>5 mg/liter), drain performance declined rapidly; elsewhere the drains consistently produced near-neutral effluent, even when influent contained small concentrations of dissolved Fe^+ (<5 mg/liter). Rates of limestone dissolution computed on the basis of average long-term Ca ion flux normalized by initial mass and purity of limestone at each of the drains ranged from 0.008 to 0.079 year^-1. Data for alkalinity concentration and flux during 11-day closed-container tests using an initial mass of 4kg crushed limestone and a solution volume of 2.3 liter yielded dissolution rate constants that were comparable to these long-term field rates. An analytical method is proposed using closed-container test data to evaluate long-term performance (longevity) or to estimate the mass of limestone needed for a limestone treatment. This method condisers flow rate, influent alkalinity, steady-state alkalinity of effluent, and desired effluent alkalinity or detention time at a future time(s) and aplies first-order rate laws for limestone dissolution (continuous) and production of alkalinity (bounded).