Upper esophageal manometry is technically problematic. Published normal values are, therefore, few and wide ranging, reflecting catheter and recording-system variables, while the reproducibility of measurements and the influence of food consistency have been little studied. In this investigation, 50 healthy volunteers were studied with (1) a 2.8-mm-diameter six-sensor catheter-mounted transducer assembly and (2) a 3.2 X 7.2-mm sleeve device linked to a computerized recorder with a pressure-sample rate of 32/sec. The study protocol included water, bread, and semisolid swallows. Upper esophageal sphincter (UES) tonic pressures measured with the catheter-mounted assembly were lower and more reproducible than pressures measured with the sleeve system. Compared with water, bread swallows showed greater pharyngeal and sphincter after-contraction pressures, while semisolid swallows had less complete sphincter relaxation. Duration of pharyngoesophageal contractions was greater with bread or semisolid than water. The observations have established normal values for measurements of UES function and, in addition, have shown that (1) catheter variables significantly influence the measurement of upper sphincter tonic pressure, (2) pressures recorded with the catheter-mounted transducer are most reproducible, and (3) pharyngoesophageal motility patterns vary significantly according to the substance swallowed.