Several studies are discussed which investigated the usefulness of the disappearance rate of macromolecules from the peritoneal cavity for estimating convective fluid loss from the peritoneal cavity into the peritoneal lymphatic system. It is shown that dextrans are removed from the peritoneal cavity by a size-independent process at a mean rate of 1.37 +/- 0.15 ml/min, whereas the clearance from blood to dialysate of dextrans is size-dependent. The fluid removal rate estimated by the difference in bidirectional transport of inulin (1.79 +/- 0.38 ml/min; p < 0.0005) was of the same order of magnitude as has been found using the removal rate of macromolecules from the peritoneal cavity. Also, the role of local accumulation of macromolecules was studied during continuous administration of dextrans. No differences were found in the dextran disappearance rate before and after saturation of the peritoneal interstitium with dextran (1.1 +/- 0.6 vs. 1.0 +/- 0.4 ml/min). During a study using a hypoosmotic solution we calculated a net transcapillary backfiltration of fluid, whereas the dextran removal rate was in the same order of magnitude as found using commercially available dialysate. In our opinion, the disappearance rate of macromolecules is an estimate of convective fluid loss from the peritoneal cavity into the peritoneal lymphatic system.