Six concentrations of sucrose and six concentrations of sodium saccharin were presented as stimuli to 30 Ss. The concentrations were equally spaced logarithmically, covering a 32: I range for each solute. The 12 aqueous solutions were presented at four levels of viscosity, ranging approximately from I to 10,000 centipoises, with sodium carboxymethylcellulose (cellulose gum) as the thickening agent. Magnitude estimates indicate that perceived sweetness decreases with viscosity. In log-log coordinates the Viscosity-sweetness relation is roughly linear with a negative slope, suggesting that sweetness of both sodium saccharin and sucrose varies as a power function of viscosity, with an exponent between -.20 and -.25.The viscosity of a substance is one of the many variables that appear to affect taste, although few studies have been made of it. Rather, viscosity has been viewed as a "disturbing secondary criterion" in taste studies (Lewis, 1948), and a deus ex machina inserted in the discussion of results by various Es, Crocker (1945) speculated that the viscosity of a solution thickened with gum tragacanth, flaxseed, or other mucilaginous material interferes with the diffusion of sapid substances to the receptor. Several results from taste threshold studies suggest that this interference occurs in the detection task. In one such study, Mackey and Valassi (1956) showed that sucrose, sodium chloride, caffeine, and tartaric acid were easiest to detect in a liquid solution, intermediate in a foam base, and most difficult in a gel medium. In a later experiment, Mackey (1958) found that when methylcellulose, a common thickener, was added to aqueous solutions of caffeine, quinine, and saccharin, the threshold rose, presumably because of the increased viscosity. Stone and Oliver (1966) also reported that for a number of thickening agents the detection threshold for sucrose was higher than when sucrose was dissolved in water alone.Stone and Oliver (1966) also invcstigated the relation between suprathreshoJd taste intensity and viscosity. They showed that sucrose solutions which had been thickened with various types of natural and artificial gums were rated higher in sweetness than sucrose solutions in pure water. In a similar study, 'Current address: Department of Psychology. Stanford University, Stanford, California 94305.t Requests for reprints should be mailed to Howard R. Moskowitz, PRL Division, U.S. Army Na tick Laboratories, Natick, Massachusetts 01760. Pangborn (1963) found that in water solutions fructose was rated sweeter than sucrose (on a weight basis), whereas in viscous pear nectar the order was reversed. On the other hand, Skramlik (1926) demonstrated that the intensity of taste was consistently lower in paraffin oil than in water. Thus, there does not appear to be any consensus about the effect of viscosity upon suprathreshold intensity of taste.The present experiment was designed to determine a functional relation between viscosity and perceived sweetness, and differs from most of the studies cited above ...