We operate in the three-dimensional (3D) environment in which objects move in depth, thus estimating the 3D motion of objects is important for many everyday activities. Neural processing of visual image motion within the fronto-parallel plane has been studied extensively (Andersen, 1997, Born and Bradley, 2005), however, relatively little is known about how cortical neurons compute object motion in depth. There are two binocular cues to motion-in-depth: change in horizontal disparity over time (CD) and interocular velocity differences between the two eyes (IOVD). Psychophysical studies have demonstrated that both IOVD and CD cues contribute to perception of motion-in-depth (Cumming and Parker, 1994, Harris et al., 2008). We recently reported selectivity for the direction of motion-in-depth (approaching vs. receding) in macaque area MT. This selectivity is driven primarily by the IOVD cue with a small contribution of the CD cue. In this study, we explored the neural mechanisms that underlie selectivity for motion-in-depth based on IOVD and CD cues. To explore the mechanism for coding IOVD cues, we measured interocular speed interaction profiles by presenting motion stimuli having all 64 combinations of 8 speeds presented to each eye. A weighted sum of monocular responses generally fit the speed interactions well, but the weights were not predictable from the ocular dominance of each neuron. To test the mechanism for coding CD cues, disparity-time maps were measured using reverse correlation. A tilted response profile in the disparity-time maps generally predicted selectivity for motion-in-depth based on CD cues. These experiments reveal distinct mechanisms by which single neurons in area MT signal motion in depth based on IOVD and CD cues.