Traditionally, autoimmune pathogeneses have been attributed to CD4+ T lymphocytes, as in multiple sclerosis (MS), rheumatoid arthritis, type 1 diabetes mellitus, and/or to B lymphocytes, as in myasthenia gravis and systemic lupus erythematosus. That is because their primary genetic associations are mostly with certain human leukocyte antigen class II alleles, whose gene products present antigens to CD4+ T cells. Because few autoimmune diseases show stronger associations with major histocompatibility complex class I alleles (ankylosing spondylitis, Behçet's disease, and psoriasis), CD8+ T cells, which interact with major histocompatibility complex class I molecules, have been largely ignored in autoimmunity research. However, a variety of findings has recently revived interest in this population, particularly in MS. First, it shows associations with major histocompatibility complex class I alleles. Second, its lesions show a predominance of CD8+ T cells. Third, these represent effectors that can directly damage central nervous system target cells. Furthermore, several clinical trials of monoclonal antibodies specifically against CD4+ T cells, or the polarizing cytokines on which they depend, have failed to show any therapeutic benefit in MS, unlike broader-spectrum antibodies that deplete all T cells. Here, we review the evidence that CD8+ T cells play a role in MS pathogenesis. Ann Neurol 2009;66:132–141