Thiocyanate (SCN -) contamination threatens aquatic ecosystems and pollutes vital fresh water supplies. SCNdegrading microbial consortia are commercially deployed for remediation, but the impact of organic amendments on selection within SCNdegrading microbial communities has not been investigated. Here, we tested whether specific strains capable of degrading SCNcould be reproducibly selected for based on SCNloading and the presence or absence of added organic carbon. Complex microbial communities derived from those used to treat SCNcontaminated water were exposed to systematically increased input SCN concentrations in molasses-amended and -unamended reactors and in reactors switched to unamended conditions after establishing the active SCNdegrading consortium. Five experiments were conducted over 790 days and genomeresolved metagenomics was used to resolve community composition at the strain level. A single Thiobacillus strain proliferated in all reactors at high loadings. Despite the presence of many Rhizobiales strains, a single Afipia variant dominated the molasses-free reactor at moderately high loadings. This strain is predicted to breakdown SCNusing a novel thiocyanate dehydrogenase, oxidize resulting reduced sulfur, degrade product cyanate (OCN − ) to ammonia and CO2 via cyanase, and fix CO2 via the Calvin-Benson-Bassham cycle. Removal of molasses from input feed solutions reproducibly led to dominance of this strain. Neither this Afipia strain nor the thiobacilli have the capacity to produce cobalamin, a function detected in low abundance community members. Although sustained by autotrophy, reactors without molasses did not stably degrade SCNat high loading rates, perhaps due to loss of biofilm-associated niche diversity. Overall, convergence in environmental conditions led to convergence in the strain composition, although reactor history also impacted the trajectory of community compositional change.