Little is known at present about the saccharide components of lipofuscin (age pigment) and ceroid pigments in situ. The purpose of this study was, therefore, to study in detail the lectin reactivities of lipofuscin in neurons and cardiac myocytes of old humans and rats. In addition, those of diverse ceroid pigments found in human aortic atheromas, in the livers of choline-deficient rats, in the uteri of vitamin E-deficient rats and in the crushed epididymal fat pad of rats, are included. Cryostat and deparaffinized sections from all these tissues were either extracted with a solvent mixture of chloroform-methanol-water (10:10:3, v/v) and incubated with 7 different biotinylated lectins or left untreated. Delipidation was done in order to study whether it was possible to discriminate between the saccharide moieties of glycolipids and glycoproteins of lipofuscin and ceroid pigments in situ. Other similarly treated sections were used to study the autofluorescence, sudanophilia, acid-fastness and reactivity to PAS. The frequency and intensity of lectin binding and standard histochemical properties of all the pigments were evaluated semi-quantitatively and blind. The results indicated that mannose was in general the most consistently detected sugar residue in lipofuscin granules of humans and rats, and that this pigment may also contain acetylglucosamine, acetylgalactosamine, sialic acid, galactose and fucose. However, notable differences were found not only in the lipofuscin saccharide components of different cell types of humans and rats, but also in those in the same type of cells in both species. Although mannose was not detected in the hepatic ceroid of choline-deficient rats, this saccharide moiety was almost always present in the other ceroid pigments. Each of the ceroids also contained other types of saccharides although the frequency of the latter varied between different ceroid pigments. While lipofuscin and each of the ceroid pigments showed somewhat different lectin binding patterns, the variability in the frequency of reactivity to lectins suggests that these patterns may not be permanent but transient. In this sense, it appears that lectin histochemistry may not allow these pigments to be differentiated. Furthermore, the extractive procedures used in this study did not enable us to determine whether the saccharides detected in the pigments in situ corresponded to glycolipids or glycoproteins.