The distribution and genetic structure of wildlife populations may be impacted by landscape features and anthropogenic factors. The role of these factors in the genetic relationships among Timber Rattlesnakes (Crotalus horridus) from 21 locations in Pennsylvania, New Jersey, and Virginia was investigated using microsatellite genetic markers. Evidence for null alleles was found in 8 of the 21 populations studied, with frequencies ranging from 0.018 to 0.185. However, the presence of null alleles appeared to have no effect on the interpretation of the patterns of genetic relationships among the populations studied. The analyses showed that the overall extent of genetic differentiation was consistent with isolation by distance; however, a substantial amount of among-groups variation could not be accounted for by geographic distance alone. Bayesian analysis (STRUCTURE), genetic distance analysis, and factorial correspondence analysis (FCA) all indicated that the sampled populations represented three genetic groups, which corresponded closely with physiographic regions. These three groups consisted of an Atlantic Coastal Plain group, an Appalachian Plateau group, and an Appalachian Ridge and Valley group. We recommend that these be considered as separate conservation management units (MUs) for C. horridus in Pennsylvania and New Jersey. Several potential barriers to gene flow were identified, including the Appalachian Mountains, several large rivers, and major roadways. The lowest levels of genetic variation were found in the highly isolated populations from the Atlantic Coastal Plain.