To examine the hypothesis that fluid absorption is driven by transepithelial osmotic differences, we measured such differences in proximal tubules perfused under oil and attempted to increase the rate of fluid absorption (Jv) and hence the absorbate-perfusate osmotic differences by lowering the osmolality of the perfusate. We were able to consistently increase Jv in this manner only in proximal straight tubules perfused with a simple perfusate that contained no bicarbonate or amino acids. With the simple perfusate, a small but significant increase in the absorbate-perfusate osmolality difference was seen with increased Jv, which is expected if the volume absorption is driven by a transepithelial osmotic difference. In addition, lowering the perfusate osmolality from 290 to 160 mosmol/kg H2O increased the rate of solute absorption from 79 +/- 7 to 91 +/- 8 posmol . min-1 . mm-1; this increase was partly accounted for by an increase in the rate of absorption of glucose from 6.6 +/- 0.9 to 9.5 +/- 1.1 pmol . min-1 . mm-1. In contrast, with the complete perfusate in proximal straight tubules there was little or no increment in Jv, no change in transepithelial osmolality differences, and a decrease in the rate of solute transport with hypoosmolality from 136 +/- 21 to 87 +/- 22 posmol . min-1 . mm-1. In proximal convoluted tubules, similar results were obtained, but a time-dependent decline of Jv complicated the interpretation of the results in the convoluted tubules. It is hypothesized that the observed changes in solute transport with hypotonic perfusate may be the result of changes in membrane permeability that are subsequent to cell swelling.