To resolve discrepancies between models used to calculate myocardial capillary solute permeability we measured tissue-fluid pressure (Pi), perfusion pressure (Pp), and heart weight continuously after a step change in the NaCl or inulin concentration of crystalloid solutions perfusing isolated rabbit hearts. Linear regression analysis of the mean values for near steady-state Pi plotted against Pp showed significantly different (P less than 0.01) slopes of 0.77 and 0.40, respectively, for the inner and outer halves of the left ventricular free wall. Tissue compliance (Tcm) was measured as the fractional change in whole heart weight divided by the concurrent change in Pi. For NaCl and inulin osmotic transients Tcm values were 0.005 +/- 0.001 (SE) and 0.014 +/- 0.004 ml X cmH2O-1 X g tissue-1, respectively. A model for heart weight change based on the Starling hypothesis predicts significantly different tissue compliance (Tcc) values of 0.031 +/- 0.004 and 0.040 +/- 0.005 ml X cmH2O-1 X g tissue-1 for the same transients, respectively. This model shows myocardial capillary permeability for inulin and NaCl are overestimated by 25% if tissue compliance is not included in the analysis. The discrepancy between Tcm and Tcc is discussed.