Among species, populations, and individuals, there exists a tremendous amount of variation in how hosts respond to, and are thus protected from parasites. Such variation inevitably affects host-parasite dynamics and ultimately how parasites will move through and evolve in communities. A likely factor in the diversity of immune responses seen in nature are the costs associated with activation of the immune system upon exposure to parasites. Costs can manifest in many ways, including changes in resource usage or metabolism, self-damage from inflammatory reactions, lost opportunities (e.g., foraging reproduction), and often as tradeoffs with other physiological processes. However, we do not yet fully understand the factors that influence costs of immune activation across ecological scales, nor the relationship between immune costs and protection from parasites, despite the common assumption that greater costs equates to better protection. For my dissertation, I have investigated large-scale drivers of immune costs, specifically whether life history and/or body mass influence costs of immune activation (Chapter 1), whether magnitude of parasite exposure affects immune activation costs at the population level (Chapter 2), and the relationship between costs of immune activation and benefit in terms of parasite protection (Chapter 3). Across taxa, immune costs are likely to be affected by host life history traits such as longevity and reproductive scheduling such that long-lived and slow to mature species (i.e., slow-paced) should experience lower costs than animals that die comparatively early (i.e., fast-paced). By reducing immune costs, slow-paced species could reduce the accumulation of damage associated with repeated activations of the immune response that could reduce successful reproduction over a long life. Likewise, body mass should also be an important determinant of immune costs as physical size should affect the extent to which hosts are exposed to parasites as well as their ability to combat infection. For the first chapter, I used meta-analysis to determine whether life history traits (lifespan and time to maturity) and/or body mass affected functional costs of immune activation (e.g., changes in performance, food intake, growth, mass, reproductive effort/success, survival) across taxa. The results of this study showed that, in general, animals incur costs of immune activation and that costs are influenced by life history and body mass such that species that are relatively long-lived experience greater costs of immune activation than species that are relatively short-lived. We also found that small species experienced relatively greater costs than large species. Such patterns may arise because long-lived species may have been selected to endure high costs of immune activation to develop more robust adaptive responses in order to extend lifespan. In addition, small animals may experience greater costs than large animals because of a higher cell turnover rate potentially resulting in greater self- damage (via oxidative damage). While costs of immune activation can be broadly predicted by lifespan and body mass, smaller- scale factors, such as magnitude of exposure to parasites are also likely important drivers of immune costs within populations. Indeed, if costs increase linearly with magnitude of exposure, then at a certain level of parasite burden, costs may become too great for the host and selection may favor hosts that tolerate, rather than eliminate infection. In the second chapter, I investigated how exposure to multiple concentrations of Salmonella lipopolysaccharide (LPS), a immunogenic component of Gram negative bacteria cell walls, affected immune costs in brown anole lizards (Anolis sagrei). To quantify costs, I examined allocation of leucine, a critical amino acid to immune tissue (liver) and reproductive tissue (gonads). I predicted that immune costs would increase with exposure, however, because females often suffer decrements in immune function during reproduction, I expected their immune costs to be less pronounced than males. I also hypothesized that leucine allocation to reproductive tissue would decrease with increasing magnitude of LPS exposure because immune activation often results as tradeoffs as competing physiological systems vie for limited resources. I found that costs of immune activation increased with magnitude of LPS exposure, and that costs differed between the sexes as males allocated more leucine to the liver at high LPS doses, but did not shift allocation from gonads to livers. Females, however, tended to bias resources to livers and away from gonads with increasing LPS doses. Interestingly, I also found that cost of immune activation increased linearly with magnitude of exposure, indicating that this species may tolerate high levels of infection with Salmonella bacteria as the cost of resisting infection is likely to become unmanageable. Such a finding suggests that at the population level, brown anoles may be particularly important contributors to the spread of Salmonella within populations, or even communities. Although costs of immune activation may drive a host’s response to infection by limiting the magnitude of its response when costs of a strong response are too great, we cannot predict how immune costs may be driving parasite prevalence and selection on hosts unless we understand the relationship between costs of immune activation and protection. While it is commonly assumed that high costs of immune activation are indicative of a stronger immune response and ultimately better protection for the host, the relationship between immune costs and benefits remains unstudied. In the third chapter, I examined host- and parasite-mediated costs of parasite exposure and the relationship between costs and benefits of immunity by exposing brown anole lizards to live or killed malaria parasites and following the course of infection over 7 weeks. I measured costs by quantifying glucose oxidation during the 12-36 hours following exposure to determine whether and how cost of initial exposure related to protection from malaria parasites. As glucose is the primary fuel source for malaria parasites, we predicted that hosts that oxidized little glucose would also experience lower parasite burdens. We found that lizards infected with killed parasites oxidized less glucose than control-infected animals, however lizards infected with live parasites did not differ from either the control or killed parasite group, indicating the possibility of a parasite-mediated mechanism to slow host-driven glucose sequestration. Importantly, we also found that the cost of exposure appeared to come with a benefit: lizards infected with live parasites that experienced the lowest glucose oxidation also had lower parasite burdens over the 7 weeks following exposure. The results from this study show that brown anoles that experience greater costs of malaria exposure also experience greater protection, contributing to our understanding of how individual-level processes drive disease dynamics within communities. Altogether, my dissertation has identified that broad characteristics such as life history and body mass can be helpful in predicting costs of immune activation and has additionally demonstrated the importance of smaller-scale processes such as the relationship between immune costs and magnitude of exposure at the population level and the relationship between immune costs and benefits at the individual level. These studies have expanded our understanding of the factors that drive variation in immunity at multiple levels and give the field of ecoimmunology a more holistic picture of the species, population and individual-level processes that affect host-parasite interactions within communities.