Model studies have been advanced to suggest both that a siphon does and does not support cerebral blood flow in an upright position. If a siphon is established with the head raised, it would mean that internal jugular pressure reflects right atrium pressure minus the hydrostatic difference from the brain. This study measured spinal fluid pressure in the upright position, the pressure and the ultrasound-determined size of the internal jugular vein in the supine and sitting positions, and the internal jugular venous pressure during seated exercise. When the head was elevated approximately 25 cm above the level of the heart, internal jugular venous pressure decreased from 9.5 (SD 2.8) to 0.2 (SD 1.0) mmHg [n = 15; values are means (SD); P < 0.01]. Similarly, central venous pressure decreased from 6.2 (SD 1.8) to 0.6 (SD 2.6) mmHg (P < 0.05). No apparent lumen was detected in any of the 31 left or right internal veins imaged at 40 degrees head-up tilt, and submaximal (n = 7) and maximal exercise (n = 4) did not significantly affect internal jugular venous pressure. While seven subjects were sitting up, spinal fluid pressure at the lumbar level was 26 (SD 4) mmHg corresponding to 0.1 (SD 4.1) mmHg at the base of the brain. These results demonstrate that both for venous outflow from the brain and for spinal fluid, the prevailing pressure approaches zero at the base of the brain when humans are upright, which negates that a siphon supports cerebral blood flow.