Transport studies of glucose analogs [3H] 3-O-methyl-D-glucose (mD-glu) and L-[14C]glucose (L-glu) across the isolated retinal pigment epithelium (RPE) of the bullfrog was undertaken to determine whether the glucose transport mechanism was dependent upon the postulated ion-transport scheme and/or whether glucose transport is insulin-mediated. In addition, metabolic inhibitors were tested to explore the energy requirements of glucose transport across the RPE. Flux studies of mD-glu and L-glu performed with mounted RPE tissues, with short circuit current (SCC) and potential difference (PD) monitored via automatic voltage clamp apparatus, indicate that transport is clearly stereospecific with D-glucose being transported at least 13 times faster than L-glucose. The system was found to be saturable with a Km of about 24 mM glucose and Vmax of 1400 nmol cm-2 hr-1. Unidirectional Michaelis-Menten constants indicate that the RPE glucose carrier is accessible for transport from either the choroid or retinal side and a bidirectional facilitated diffusion mechanism is suggested. Insulin had no effect on either ion transport (SCC) or glucose transport (passive or facilitated). Both aerobic and anaerobic energy inhibitors decreased ion transport to less than 25% of control, but had little effect, if any, on glucose transport across the isolated RPE. Sodium iodoacetate decreased ion transport by 90% of control, but a much slower decrease in facilitated glucose transport of 22% of control suggests that carrier energy requirements, if any, are not direct or immediate. Osmotic studies performed with sucrose and glucose suggest that elevations in osmolarity increase passive glucose movement and decrease facilitated glucose-transport rates. Glucose was found to be much more detrimental to glucose transport than sucrose, suggesting that at high concentrations molecular glucose decreases facilitated transport and increases passive glucose movement by a mechanism other than can be accounted for by osmotic considerations. A model for RPE glucose transport, consistent with current data, is proposed which translocates D-glucose, via an alternating conformational change of the glucose carrier. This carrier does not require a direct supply of metabolic energy, nor a functioning ion-transport mechanism. At a given moment, a single binding site for D-glucose is postulated to be available on either side of the RPE membrane for glucose translocation, although binding site affinity for glucose could differ on each side.