Two psychophysical measures of tactile sensitivity--grating-orientation (GO) and smooth-grooved (SG) discrimination--were used to determine tactile spatial acuity with and without an intermediate surface (latex glove) interposed between the contactor and the skin. Measures were made at three locations that varied in sensitivity and in density of innervation of the primary afferent fibers: the index fingerpad (fingertip), palmar surface of the proximal phalanx (fingerbase), and the thenar eminence (palm). Neurophysiological studies have suggested that the density of innervation of SAI fibers is a limiting factor in spatial acuity. In the present study, without a glove, the GO thresholds varied as a function of location. With the glove, increases in the GO thresholds were relatively uniform and modest. Without a glove, however, the SG thresholds were well below the GO thresholds, and changing the site of stimulation had little effect on the threshold. With a glove, the SG thresholds increased by 100% to more than 300% as in comparison with those in the no-glove condition. The largest increases occurred at the less sensitive locations. The results of the GO task are consistent with the view that GO is a valid measure of spatial acuity. The results of the SG task, however, are inconsistent with previous results and suggest that both spatial and intensive factors are involved in this task.