This study was a comprehensive evaluation of population reconstruction techniques. Population reconstruction techniques are population estimation methods that calculate a minimum population size based on age-specific harvest data (Downing 1980, Roseberry and Woolf 1991). Population reconstruction techniques share the following characteristics: 1) utilization of catch-at-age data and 2) backward addition of cohorts to estimate a minimum population size. I developed a questionnaire to survey the biologists participating in this survey to determine the most common reconstruction technique used to estimate population sizes of exploited white-tailed deer (Odocoileus virginianus) and black bear (Ursus americanus). Downing reconstruction (Downing 1980) was the most commonly used reconstruction technique among biologists participating in this study. Based on a comprehensive literature review and discussions with state biologists, I decided to evaluate virtual reconstruction (Roseberry and Woolf 1991) and develop a new reconstruction technique: Reverse Order reconstruction. I developed a quantitative population model in Microsoft Visual Basic 6.0 to evaluate the ability of the 3 reconstruction techniques to estimate population sizes given a variety of conditions. I evaluated the effects of stochasticity on reconstruction population estimates by incorporating different levels of environmental stochasticity (i.e. process error) and measurement error in the generated or "known" population. I also evaluated the effects of collapsing age classes and aging biases on population estimates. In all conditions, Downing and virtual reconstruction were underestimates of the actual population size. Reverse Order reconstruction more closely estimated the actual population size, but is also more data-intensive than the other 2 methods. Measurement error introduces more uncertainty in the reconstructed population estimates than does process error. The population simulation model proved that Downing and virtual reconstruction are consistently underestimates and the percent underestimation is due to lack of inclusion of a natural mortality rates in population estimation. I used the results of the questionnaire to characterize the harvest datasets of the states participating in this study. From these results, I chose two harvest datasets to further analyze: a white-tailed deer harvest dataset from North Carolina and a black bear harvest dataset from Pennsylvania. I analyzed these datasets with Downing and virtual reconstruction. I also applied the quantitative population model to these datasets to evaluate the effect of increasing levels of measurement error on the variance of the population estimates. I found that Downing and virtual reconstruction estimated the population sizes very closely to one another, within 5%, for both datasets, and the reconstructed estimates closely tracked the actual harvest numbers. I also found that increasing levels of measurement error increased the variance associated with reconstructed population estimates and may decrease the ability of these techniques to accurately capture population trends. / Master of Science