1. The aims of this study were to determine the clinical feasibility of continuous, non-invasive Finapres recordings as a replacement for intrabrachial pressure during a 30 min head-up tilt, and the reliability of continuous cardiac output computation by pulse contour analysis from the finger arterial versus the brachial waveform. 2. In eight healthy subjects a 30 min 70 degrees passive head-up tilt was performed. Finger arterial (FINAP) and intrabrachial (IAP) pressures were measured simultaneously. Beat-to-beat changes in stroke volume were computed using a pulse contour algorithm. 3. Accuracy (the group-averaged FINAP-IAP difference) and precision (the SD of the difference) of Finapres measurements were 4 and 9 mmHg for systolic blood pressure, -5 and 9 mmHg for mean blood pressure and -5 and 9mmHg for diastolic blood pressure. 4. The time course of the FINAP-IAP differences during head-up tilt showed a linear trend (P < 0.001 for all pressure levels). Averaged for the group, the difference increased 7 mmHg for mean blood pressure. The difference in stroke volume computed from FINAP and IAP was 0.3 +/- 5% (mean +/- SD), and independent of the duration of the tilt (P > 0.05). This difference did not change at low blood pressure levels (0.5 +/- 6%). 5. The qualitative performance of the Finapres allows it to be used in the clinical setting as a monitor of sudden changes in blood pressure induced by a 30 min head-up tilt. Relative changes in stroke volume, as obtained by pulse contour analysis of the finger arterial waveform, closely follow intrabrachial values during long-duration head-up tilt and associated arterial hypotension.