In order to assess the subcellular distribution of estrogen-binding components in their native state, plasma membrane and other cell fractions were prepared by quantitative methods from uterine cells in the absence of [ 3H]-estradiol-17β (E 2β). Cells isolated from uteri of ovariectomized rats were disrupted with minimum homogenization in buffered isotonic sucrose with CaCl 2 and fractionated using isotonic media throughout. Activities of succinate dehydrogenase and acid phosphatase were concentrated in the mitochondria:lysosome(M + L) fraction. Glucose-6-phosphatase occurred predominantly in the microsome-rich (P) fraction. Alkaline phosphatase and 5'-nucleotidase were found principally in P and crude nuclear (N) fractions, with N also retaining 96% of DNA. Determinations of specific [ 3H]-E 2β binding to cell fractions at equivalent protein levels were conducted by equilibration for 2 h at 4°C. Binding-sites for E 2β were present in N > P > M + L > 105,000 g supernates (S). However, by using alternative homogenization procedures known to elicit lysis, fragmentation, and stripping of cell structures, E 2β binding-sites as well as 5'-nucleotidase, a plasma membrane marker enzyme, both occurred predominantly in S. Using the more conservative cell disruption and fractionation procedures, plasma membrane subfractions (F2) with densities of 1.13–1.16 were partially purified, principally as smooth membrane vesicles and large “ghosts”, by further centrifugation of N in a discontinuous sucrose density gradient. Activity of 5'-nucleotidase in F2 was enriched to 12 times that of the homogenate. Specific binding-sites for E 2β were concentrated in F2 to 23 times homogenate levels. Such binding of E 2β in F2 was saturable, with an association constant of 4.3 × 10 10 M −1. At saturation, E 2β receptors in F2 correspond to 2.0 pmol/mg membrane protein. Ligand specificity of [ 3H]-E 2β binding to F2 was established by negligible competition by 200-fold molar excess of unlabelled estradiol-17α, cortisol, testosterone, or progesterone, whereas E 2β and diethylstilbestrol were effective inhibitors. Specific binding of E 2β to F2 at 4°C was blocked by prior exposure of membranes to trypsin or to 60°C, but remained essentially undiminished by extraction of membranes with either severely hypotonic or high-salt buffers. In controlled experiments, it was found that only 6–9% of [ 3H]-E 2β-labelled cytosol components became associated with F2 during 2 h incubation at 4°C; the bulk of such adsorbed material was readily extracted by high-salt buffers. Thus, enrichment of F2 in E 2β-binding activity cannot be attributed to gross entrapment or adsorption of cytosolic components. These data indicate that high-affinity membrane binding-sites with specificity for E 2β must be further considered in investigations of the uterine cell recognition of and response to the hormone.