1. A study is described of the relationships which exist between disaccharide hydrolysis and glucose transport in the small intestine of Rana pipiens and Bufo vulgaris. The experiments were undertaken on the intestine perfused in vitro through the vascular system and with fluid circulating through the intestinal lumen. For this system it was found that, with [U-(14)C]glucose in the intestinal lumen, the apparent specific activity of the glucose appearing in the vascular effluent was not significantly different from that in the lumen.2. Changes of ionic composition of vascular and luminal fluids, and the presence of phloridzin or strophanthin, had little effect upon the maltase activity in situ in R. pipiens, although this activity was somewhat reduced when the sodium of the intestinal lumen was replaced by lithium. In contrast, in all cases a marked reduction was found in the rate of glucose translocation in the vascular effluent.3. With Tris substituted for the luminal sodium, there was evidence of a competitive inhibition of the maltase activity in situ by the buffer cation. At the same time, the rate of glucose translocation into the vascular effluent was but little affected and there was an apparent increase in the efficiency with which the cellular systems responsible for the translocation were able to capture the glucose liberated.4. It was found that competition for transepithelial translocation occurred between the glucose initially present in the intestinal lumen, and glucose derived from either maltose or trehalose. There was no evidence for competition for hydrolysis between maltose and trehalose, yet the glucose units derived from these two disaccharides competed with each other for translocation.5. The significance is discussed of the finding that it is possible to dissociate the processes of disaccharide hydrolysis from those underlying the translocation of hexose units into the vascular effluent. It is suggested that monosaccharide units released by the hydrolysis of disaccharide molecules have access to a pool of glucose which is equally accessible to glucose free in the intestinal lumen. It is also suggested that the rate of transport of glucose from the pool into the vascular effluent (i.e. glucose translocation) is determined by the concentration of glucose within the pool. From consideration of the properties of a model operating upon these principles, it is possible to predict the relative contribution of disaccharides and monosaccharides in the intestinal lumen to the glucose appearing in the vascular effluent. The experimentally determined contributions of the two sources are very similar to those predicted from the model. The implications of possible sites for the postulated pool are discussed.