The role of T lymphocytes in the pathology of rheumatoid arthritis (RA) is controversial. To some extent this has resulted from contradictory data, but even where specific points of fact are not in dispute, their interpretation often is.^1–4 Nevertheless, the basic idea of a pathological T cell response in RA receives significant support from the work of several groups who have removed them by thoracic duct drainage, lymphapheresis or total lymphoid irradiation (TLI). Each of these procedures ameliorates RA; in the case of TLI which induces a profound and sustained decrease in CD4+ cells, the effect on synovitis was prolonged⁵.These observations are in line with the general assumption that the strong association of RA with HLA-DR molecules containing a particular conserved region of amino acids implies an important role for T cells in RA⁶, since the physiological role of HLA-DR molecules is to present antigens to T cells. In addition the association with Dw4 and related alleles is strongest for the most severe and persistent forms of disease⁷, arguing that T cells may be important not just in initiation of RA but also in its perpetuation. However, the infiltrating T cells seem to be remarkably inactive and as a population to lack specificity for any particular antigen, leading some to conclude that their role is either passive or irrelevant, while others contend that it is precisely this inactivity which is responsible for the persistence of RA.