To better understand the history of coral reef ecosystems, we sequenced DNA from Carribean Acropora palmata reef cores dated to ~1000 years old. The project sought to answer two questions: 1) Is it possible to sequence aDNA from coral reefs, and, 2) what does this data tell us about coral ecosystems of the distant past? While we did not find any coral aDNA reads in the cores, we successfully recovered ~100,000 (0.2% overall) Symbiodinacea reads from each sample. To determine the authenticity of these reads we looked for patterns of damage characteristic to ancient DNA, such as rate of C to T substitutions at the 5’ end of mapped reads. Our data matched these expectations and was distinct from the patterns seen in modern day Symbiodinacea reads. By assigning the reads to the species level, we found the predominant symbiont to be Cladicopium spp. This is surprising, as present-day A. palmata associates with Symbiodinium spp. Given the sensitive nature of coral-zooxanthellae symbiosis, this shift may signal key ecological changes. Further, by applying metagenomics approaches, we assigned an additional ~0.2% of the reads to the phylum Rhodophyta. Half of these reads belonged to the order Corallinales (crustose coralline algae) – a key organism promoting coral larvae settlement. Overall, the sequencing of reef matrix aDNA identified ancient organisms at the species level, though lack of read coverage across putative genomes limited further analysis.