Ebola virus (EBOV) is a zoonotic pathogen which causes a fulminant hemorrhagic fever and results in up to 90% fatality. Despite efforts over the past 38 years, a licensed prophylactic or post-exposure option remains unavailable. Several experimental vaccines have already demonstrated protection from lethal EBOV disease in nonhuman primates (NHPs). However, attempts to translate this research from bench to bedside have been hampered, since the immune responses that correlate with protection against EBOV are not well-defined. Without this information, it is not possible to reliably predict the efficacy of vaccines or treatments in humans without lethal challenge. The goal of this thesis is to determine the immune parameters that are predictive of protection against EBOV. Due to the rapid speed of EBOV pathogenesis, it was originally hypothesized that a rapid and robust CD8+ T-cell response must be crucial for survival. Using a previously-characterized adenovirus-vectored Ebola vaccine (Ad5-optGP), transgenic/knockout C57BL/6J mice with ablation for selected immune responses were vaccinated and challenged 28 days later. Surprisingly, while CD8+ T-cell knockout mice survived infection, B-cell knockout mice did not, indicating that the antibody response played a critical role in protection. Humoral and cell-mediated responses were compared between survivor and moribund guinea pigs and NHPs from previous vaccination and post-exposure therapy experiments, either with Ad5-optGP or a vesicular stomatitis virus (VSV)-vectored vaccine (VSVΔG/EBOVGP). Circulating EBOV glycoprotein (GP)-specific IgG antibody levels were the best correlation for protection independent of vaccine platform or timing of exposure (p<0.0001), strongly supporting the role of antibody responses in the control of EBOV infection. To demonstrate that antibodies are also responsible for protection, three previously-characterized monoclonal antibodies recognizing different GP epitopes were combined into a cocktail (ZMAb). Initiation of ZMAb treatment 1 and 2 days post infection (dpi) with EBOV protected 100% and 50% of NHPs, respectively. These results provide a reliable measure for predicting protection from EBOV in three commonly used animal models, and present a strong case for the use of antibodies as an effective post-exposure treatment. This knowledge will ultimately help in the development and validation of a clinical product against EBOV infection.