Mepacrine, used as a vital stain, accumulates in the lysosomes of most cells. In platelets, however, it does not stain the lysosomes but is a specific marker for dense bodies.Mepacrine, applied as a dilute solution to living cells, stains lysosomes (Allison & Young, 1964). Its localization can be revealed by its yellow-green fluorescence. We show in this paper that this coincides with the cellular distribution of acid phosphatase positive granules in cells such as monocytes. In blood platelets, however, mepacrine has been said to stain the dense bodies (Lorez et al, 1977). Subcellular fractions rich in dense bodies fluoresced brightly but the a-granule fraction including lysosomes did not show fluorescence (Da Prada & Pletscher, 1975). Moreover, large platelets rich in dense bodies (as the platelets in the Bernard-Soulier syndrome) showed many more fluorescent granules than normal platelets (Rendu et al, 1979a, b). So too do the platelets present in idiopathic thrombocytopenic purpura (Boneu et al, 1978). Platelets with reduced numbers of dense bodies found in certain storage pool diseases show reduced numbers of mepacrine staining granules (Lorez et al, 1979). Thus there is strong circumstantial evidence linking dense bodies with mepacrine staining, but the possibility that lysosomes might also stain with mepacrine in platelets was not completely excluded.On the other hand, the numbers of mepacrine positive granules seen in the light microscope showed a unimodal distribution with a mean at approximately 5-5-65 per platelet (Lorez et al, 1977; Rendu et al, 1979a; Boneu et al, 1978); dense bodies observed with electron microscopy in unfixed and unstained whole mounts and identified by their strong intrinsic electron density do not show a unimodal-or even a gaussian-distribution per platelet (Costa et al, 1977). Instead, the number of dense bodies per platelet ranges from none to more than 20 with several indistinct peaks over this range (Costa et al, 1977).In view of the proposed value of mepacrine fluorescence as a marker for dense bodies in purifying subcellular fractions of platelets (Rendu, 1979;Rendu et al, 1979b), we decided to compare directly the same platelets both in the fluorescence microscope and the electron microscope. We have also compared platelets stained with mepacrine and observed in the fluorescence microscope, with the same platelets treated with cytochemical tests for