The rules for T-cell-mediated control of viruses that infect via the respiratory mucosae show both common themes and differences, depending on the nature of the pathogen. Virus-specific CD8(+) cytotoxic T lymphocytes (CTLs) are the key effecters of virus clearance in mice infected with both negative strand RNA viruses (influenza and Sendai) and a DNA virus, the murine gamma-herpesvirus-68 (MHV-68). Recently completed experiments establish that these activated CD8(+) T cells indeed operate primarily via contact-dependent lysis. Perforin-mediated cytotoxicity seems to be the preferred mode, though a Fas-based mechanism can apparently serve as an alternative mechanism. Immune CD4(+) T cells functioning in the absence of the CD8(+) subset cannot eliminate MHV-68 from lung epithelial cells, are somewhat less efficient than the CD8(+) CTLs at clearing the RNA viruses, and are generally ineffectual in mice that lack B lymphocytes. Though cytokine secretion by CD4(+) and CD8(+) T cells in the virus-infected lung may promote both T-cell extravasation and macrophage activation, such processes are not alone sufficient to deal consistently with any of these infections. However, CD4(+) T help is mandatory for an effective B-cell response, and can operate to promote the clonal expansion of virus-specific CD8(+) T cells in the lymph nodes and spleen. Furthermore, a concurrent CD4(+) T-cell response seems to be essential for maintaining continued CD8(+) T-cell surveillance and effector capacity through the persistent, latent phase of MHV-68 infection in B cells. Thus, the evidence to date supports a very traditional view: CD8(+) T cells function mainly as killers and the CD4(+) T cells as helpers in these respiratory virus infections.