Certain enzymes and other proteins, traditionally thought to be exclusively cytosolic, have very often been found to be associated with isolated plasma membranes. These findings raise fundamental questions about concepts of membrane structure and the methods used to isolate membranes. It is conceivable that the interaction of cytosolic proteins is central to the overall structure and function of the cell in vivo. However, it is necessary to consider the specificity of the various interactions and evidence for their occurrence under conditions which can exist within the cell.The interaction of cytoplasmic proteins with the erythrocyte membrane has enjoyed renewed interest [ 11. Although earlier work considered the problem of the interaction of hemoglobin with the membrane [2-171, more recent studies have shown that certain glycolytic enzymes also bind to specific sites [7, and that a class of peripherally associated, so-called cytoskeletal, proteins exist [23-281. The interaction of hemoglobin and the glycolytic enzymes has been shown to be largely electrostatic in nature and the discussion of the physiologic significance of these interactions is accelerating.One might at first disregard the electrostatic association of hemoglobin and the glycolytic enzymes as simply an isolation artifact. However, closer examination has shown a surprising degree of specificity in binding even under the nonphysiologic conditions necessary to study binding stoichiometry. The site of high affinity binding has been shown to be band 3 protein, the integral membrane protein involved in C1-/ HC03-exchange [l]. In this paper, I shall attempt to critically summarize the later Abbreviations used: G3PD, glyceraldehyde-3P dehydrogenase; EPR, electron paramagnetic resonance; PFK, phosphofructokinase; DIDS, 4.4 -bis (isothiocyano)-