Surface markers have been extensively used for the identification and fractionation of different sets of human lymphocytes. Although they have proven useful for the identification of different cell subpopulations and for delineating various stages of T cell maturation or activation, there are a number of limitations in their application that should be carefully considered. In most instances, markers which are reported as defining functional subpopulations are not directly involved in the functional capability to be defined. In addition, a precise correlation between surface markers and cell function is difficult if only a small percentage of the cells in a given subset is involved in the functional activity measured. It is also clear that cell activation and differentiation, or mechanisms of cell fractionation, may affect the expression of some surface markers. Moreover, the assumption that functional activities ascribed to cell subsets or to surface markers in vitro are also operating in vivo may be incorrect. Human T cell surface markers can be grouped in various (as well as overlapping) categories, namely those recognizing the total T cell population, those present on T cell subsets and those expressed only on T cells at a given stage of maturation or as a consequence of cell activation. A more precise correlation between the presence of surface markers and the expression of functional activities can be performed by more complex experimental approaches such as the combined use of various markers and the analysis of surface phenotype of T cell clones with defined functional properties. While the study of T cell clones has clearly pointed out the inadequacy of most markers commonly used for identifying functional subsets of T cells, it provides a valuable tool for identifying new markers that may indeed be selective for a given T cell function.