The files in this data release (Kellogg and others, 2023) are those referenced in the journal article by Evans and others (2023) entitled “Investigating microbial size classes associated with the transmission of stony coral tissue loss disease (SCTLD).” They contain an amplicon sequence variant (ASV) table and the raw 16S rRNA gene amplicon files from fifty-six 0.22-micrometer (µm) pore size filters, as well as six reagent blanks, three mock communities, and a 0.22-µm-filtered ultraviolet (UV)-treated seawater (FSW) control. The 0.22 µm pore size filters contained tangential flow filtration (TFF) concentrated microbial communities derived from bucket mesocosms consisting of approximately 18 liters (L) of UV-treated FSW in which either healthy or diseased corals (Diploria labyrinthiformis, Pseudodiploria strigosa, Montastraea cavernosa, Colpophyllia natans, and Orbicella faveolata) had been incubated, and a FSW control. Colonies and fragments of apparently healthy corals were collected from Florida reefs or nurseries between April 2018 and September 2020 and transported to the Smithsonian Marine Station (SMS) in Fort Pierce, Florida. These samples were transferred to indoor, temperature-controlled water tables containing FSW and maintained along with other corals as part of the SMS long-term healthy coral stock. Corals exhibiting signs consistent with stony coral tissue loss disease (SCTLD) were collected from the reef immediately prior to each experimental run (Run 1 – October 2019, Run 2 – November 2020, and Run 3 – March 2021), and transported to SMS to be placed in individual mesocosms with weighted air lines. Apparently healthy corals of the same or similar species (i.e., same genus or family) were simultaneously transferred from the SMS healthy stock tanks into identical individual mesocosms. All mesocosms were housed within outdoor water tables containing recirculating freshwater maintained at approximately 28 degrees Celsius (°C) and located under a mesh canopy to allow some sunlight attenuation. Separate “healthy” and “diseased” water tables were maintained to prevent cross-contamination between the different mesocosm types. All corals were incubated within the mesocosms for 2-5 days to enrich the water with microbes. Following the incubation period, the corals were removed from their mesocosms, and the seawater was poured through a sterilized mesh screen (200 µm for Run 1, and 106 µm for Runs 2 and 3). The water was TFF-concentrated, then sequentially passed through a sterile 0.8 µm pore size nitrocellulose filter unit (Runs 2 and 3) and a sterile 0.22 µm pore size nitrocellulose filter unit (all three runs). The 0.22 µm pore size filters were then cut from the units using ethanol-sterilized blades, with portions frozen at -20°C for later processing. Extraction of deoxyribonucleic acid (DNA) from the samples and blanks occurred between February and May 2021 at the U.S. Geological Survey St. Petersburg Coastal and Marine Science Center (USGS SPCMSC) Coral Microbial Ecology Laboratory in St. Petersburg, FL, USA using Qiagen DNeasy PowerBiofilm kits. Library preparation and DNA sequencing were conducted on July 28th, 2021 by the Michigan State University RTSF Genomics Core (East Lansing, MI) using primers 515F: GTGCCAGCMGCCGCGGTAA and 806R: GGACTACHVGGGTWTCTAAT to target the V4 variable region of the 16S ribosomal ribonucleic acid (rRNA) gene on a MiSeq sequencing system with v2 chemistry to obtain paired-end 250-base pair (bp) reads.