Influenza A viruses pose a significant threat to public health. Seasonal circulation of influenza viruses causes significant morbidity and mortality each year, despite wide-spread seasonal vaccination. Influenza viruses continually acquire substitutions in the viral membrane protein hemagglutinin (HA), the primary target of human immunity, thus allowing viruses to escape vaccine-induced immunity through a process called antigenic drift. Additionally, pandemic influenza strains occasionally enter the human population that we do not have vaccines against. To address this problem, universal influenza vaccines that target the conserved HA stalk domain are currently being developed. In this dissertation we explore the protective capacity of antibodies that target the HA stalk domain and the durability of this domain as a vaccine target. We used human serological cohorts to demonstrate for the first time that HA stalk antibodies are associated with protection against influenza infection and severe disease. In addition, we used a humanized mouse model to demonstrate the in vivo protection of HA stalk antibodies against severe influenza disease. We next determined that the HA stalk domain is acquiring amino acid substitutions and assessed the antigenic effect of these substitutions using a human serological cohort. We demonstrate that these substitutions do not reduce binding of HA stalk antibodies in polyclonal human sera, suggesting this domain has not experienced antigenic drift during the past decade of seasonal circulation. Overall, the work presented here identifies a potential correlate of protection that can be used for HA stalk-based universal influenza vaccine trials and establishes the HA stalk domain as a durable vaccine target.