Smokers are at high risk for 2 bacterially driven oral diseases: peri-implant mucositis and peri-implantitis. Therefore, the purpose of this investigation was to use a deep-sequencing approach to identify the effect of smoking on the peri-implant microbiome in states of health and disease. Peri-implant biofilm samples were collected from 80 partially edentulous subjects with peri-implant health, peri-implant mucositis, and peri-implantitis. Bacterial DNA was isolated and 16S ribsomal RNA gene libraries sequenced using 454-pyrosequencing targeting the V1 to V3 and V7 to V9 regions. In total, 790,692 classifiable sequences were compared against the HOMD database for bacterial identification. Community-level comparisons were carried out using UniFrac and nonparametric tests. Microbial signatures of health in smokers exhibited lower diversity compared to nonsmokers, with significant enrichment for disease-associated species. Shifts from health to mucositis were accompanied by loss of several health-associated species, leading to a further decrease in diversity. Peri-implantitis did not differ significantly from mucositis in species richness or evenness. In nonsmokers, by contrast, the shift from health to mucositis resembled primary ecological succession, with acquisition of several species without replacement of pioneer organisms, thereby creating a significant increase in diversity. Again, few differences were detected between peri-implantitis and mucositis. Thus, our data suggest that smoking shapes the peri-implant microbiomes even in states of clinical health, by supporting a pathogen-rich community. In both smokers and nonsmokers, peri-implant mucositis appears to be a pivotal event in disease progression, creating high-at-risk-for-harm communities. However, ecological succession follows distinctly divergent pathways in smokers and nonsmokers, indicating a need for personalized therapeutics for control and prevention of disease in these 2 cohorts.