A theory of molecular absorption from the small intestine is outlined on the basis of macrotransport analysis. Certain features of the transport process that have not been previously considered in the literature are quantitatively described. These include complex interrelationships between lumen and membrane diffusion, convection, degradation and absorption mechanisms - and overall axial convection, dispersion, degradation and absorption rates. Therapeutic molecules are assumed to be introduced in the form of a bolus into the duodenum. They subsequently convect and diffuse through the duodenum,jejunum, and ileum. Absorption:into the systemic circulation across the epithelial barrier, as well as possible degradation or aggregation in the lumen or at the apical epithelial membrane, contribute to the disappearance of the therapeutic as the bolus travels through the lumen in an oral to caudal direction. Space- and time-varying lumen concentrations are predicted, as are time-varying systemic concentrations following introduction of the bolus. The inputs to the model are primarily anatomical or physicochemical characteristics that are either known or can be measured for a given therapeutic and animal model. A detailed parametric study is made, elucidating the individual roles of permeability and degradation rates. This leads to a simple paradigm for determining the two unknowns of the model (the membrane permeability and degradation rate constant) from systemic absorption data; it is shown that the membrane permeability constant can alternatively be estimated by independent in vitro measurements. Comparisons with published experimental systemic concentrations are made for molecules ranging from small lipophilic substances, such as ibuprofen, to polypeptides, such as calcitonin,and proteins, such as insulin. The deduced epithelial permeability values show reasonable agreement with values determined using alveolar epithelia and Caco-2 cell monolayers. By contrast, the membrane permeability values deduced from a simplified model of absorption from the small intestine show relatively poor agreement with experimental values. The model may be useful as numerical simulation tool for predicting (estimates of) oral dose-response relationships in animals and humans given relatively limited in vivo data.