Objective: To explore the relationship between single nuclear polymorphisms (SNPs) in ectodysplasin A receptor (EDAR) and EDAR-associated death domain (EDARADD) genes and non-syndromic tooth agenesis. Methods: Ten putative SNPs in EDAR and EDARADD were selected, and a case-control study was conducted in 112 subjects with non-syndromic tooth agenesis and 112 normal control subjects. DNA was obtained from peripheral blood samples. Genotyping was performed by Sanger sequencing. Results: Three SNPs (rs3749098, rs3749099, and rs10432616) in EDAR exhibited significant differences in the alleles and/or genotype frequencies between the case group (individuals with non-syndromic tooth agenesis) and control group (normal individuals). The T allele was identified in the SNP rs3749098 in 99.1% of the case group and in 96.0% of the control group (P = 0.0326). Regarding the SNP rs3749099, the C allele was identified in 99.1% of the case group and in 96.0% of the control group (P = 0.0326). Regarding the SNP rs10432616, the C allele was identified in 97.8% of the case group and in 100.0% of the control group (P = 0.0245). Conclusion: Our results suggested that SNPs in EDAR could be a pathogenic factor for non-syndromic tooth agenesis. Furthermore, EDAR can be regarded as a marker gene for the risk of tooth agenesis. Keywords: case-control study, ectodysplasin A receptor, non-syndromic tooth agenesis, single nucleotide polymorphism