The study of apoptosis in relation to various human disease states, particularly HIV infection, has seen a tremendous increase in activity. In this article, values obtained by seven different assays, designed to quantify apoptosis and applicable to the study of HIV infection, are compared in two cell systems: (1) stimulus-induced apoptosis in Jurkat cells treated with anti-Fas antibody and (2) spontaneous apoptosis in PBMCs isolated from HIV-infected children. The methods used included measurement of cells with subdiploid DNA content, labeling of DNA strand breaks by the TUNEL reaction, annexin V surface labeling for the detection of exposed phosphatidylserine, cytoplasmic antigen labeling with the apoptosis-specific antibody Apo 2.7, detection of changes in flow cytometric light-scattering properties, trypan blue dye exclusion by light microscopy, and detection of changes in cellular chromatin by fluorescence microscopy. These methods produced well-correlated values in the Jurkat system, whereas the same set of methods produced more discrepant values in the PBMC analyses, especially in those patients with low CD4 counts. Specifically, our results showed that the trypan blue test was unacceptable for quantification of apoptosis during HIV infection, whereas TUNEL, of all the methods tested, showed excellent overall correlation in both cell systems, was highly specific, and matched microscopic observation of the cells. Although many of the methods were suited to the study of a homogeneous cell line, caution must be exercised when examining cell death in a heterogeneous cell mixture from an HIV-infected individual.