Selective dissolution of aragonitic grains is emerging as a volumetrically significant process that affects a broad range of modern carbonate settings. This study explores mechanisms and implications of aragonite loss in Challenger Mound, a giant cold-water coral (Lophelia pertusa) mound of Pleistocene age, which lies on the continental slope off south-west Ireland. A comprehensive sampling scheme allowed the integration of petrographic data with geochemical analyses of sediment and pore water. The mound remains virtually unlithified and consists of stacked, fining-upward cycles of silty coral floatstone–rudstone and bafflestone grading into wackestone. Whereas calcitic grains appear unaltered, aragonitic grains are corroded and fragmented. Aragonite dissolution is attributed to organic matter oxidation at/near the sediment–water interface and, at greater depths, to the initial stages of bacterially mediated sulphate reduction, when alkalinity production is outpaced by the generation of H+. Pore water profiles indicate that undersaturated waters are diffusing towards the mound interior from two centres of sulphate reduction: one located in the upper 10 m of the sediment column and a second that lies below an erosional unconformity which marks the base of the mound. Continued aragonite dissolution is expected to gradually lower the diagenetic potential of the Challenger Mound and delay lithification until deep burial, when solution-compaction processes come into play. Despite a fundamental role in predestining the final taphonomic and textural characteristics of Challenger Mound, the processes described here are expected to leave little trace in the geological record due to a lack of cementation and calcitization. Assuming that similar processes have been active throughout the Phanerozoic, results imply that the understanding of diagenetic processes in carbonate systems may be incomplete.