Diagnosing Sézary syndrome (SS) on clinicopathological grounds alone is far from straightforward, particularly in the early stages of the disease. Atypical lymphocytes may be seen in the peripheral blood of patients with reactive forms of erythroderma, so additional criteria are needed to establish the diagnosis of a T-cell leukemia/lymphoma. A wide variety of confirmatory tests have been proposed in the literature, but there has been no systematic attempt to compare the specificity and sensitivity of these different methods. Recent data indicate that T-cell receptor (TCR) gene analysis is the most useful test currently available and that methods based on polymerase chain reaction are more sensitive than Southern blot analysis. We propose that the diagnostic criteria for SS should include erythroderma, atypical circulating mononuclear cells, and evidence of a clonal T-cell population in the peripheral blood. Clonality can be established with certainty by cytogenetic or TCR gene analysis, but only the latter is sufficiently sensitive to be of value in routine diagnosis. Immunophenotypic data showing an expanded CD4⁺/CD7^– population, an elevated CD4/CD8 ratio, or restricted V beta expression are not specific to T-cell malignancy and should not be used as a sole diagnostic criteria in SS. Entry criteria for future clinical studies will need to be more rigorous if meaningful comparisons are to be made between different treatment options. (J Am Acad Dermatol 1999;41:254-9.)