Abstract: Terrestrial salamanders are often used as indicators of forest ecosystem health.  However, they are often monitored using count data. Recent efforts have tried to improve abundance estimates by using mark-recapture methods. We hypothesized that inclusion of age structure within true encounter probability of models better fits mark-recapture data from terrestrial Plethodontid salamanders. In addition, we analyzed the effectiveness of two common search methods, diurnal cover object searches and nocturnal activity searches, with regards to maximizing encounter probability. Finally, we hypothesized that including age class and search method in the models would alter abundance estimates. We used Program MARK to analyze three years of Plethodon cinereus mark-recapture data using Pollock’s robust design taken from a 144 meter2 plot. Most top models, including the best model in each year including search method; there was no support for models that did not include search method. Initial encounter probabilities during night searches were consistently greater than during diurnal cover object searches. However, only a few top models included age class. In two of the three years we found no difference between adult and juvenile encounter probabilities. We found that abundance estimates did not drastically differ between the model with the lowest AICc value (which in each year included search method and/or age class) and the best model that did not account for age class or search method. We did find evidence for yearly variation in juvenile abundances.  For these reasons we recommend that sampling of terrestrial salamanders include nocturnal activity searches and be conducted over multiple years.