The goal of this thesis is to investigate the functional communication between primary motor (M1) and somatosensory (S1) cortices during the ethologically relevant behavior of hand grasping. Performing motor behaviors in our normal life requires the interplay between complex motor control and precisely timed somatosensory feedback. While many anatomical studies have shown cortico-cortical connections between M1 and S1, how somatosensory signals, in particular proprioception, functionally interact with motor area to guide natural hand movements is yet to be discovered. We sought to test the hypothesis that sites in M1 and S1 that shared similar somatotopic representations were more likely to be connected than paired sites that were somatotopically dissimilar. We used a multi-camera motion capture system to track the kinematics of reflective markers placed on the hand and arm of the monkey, from which we reconstructed joint kinematics. High-density multi-electrode arrays were implanted in non-human primates to record neural signal from both M1 and S1 simultaneously while the monkeys engaged in a trained grasp task. Multivariate generalized linear modeling (GLM) was used to characterize the functional connectivity between signals recorded from different electrode channels. To provide more causal evidence for these interactions, intracortical microstimulation (ICMS) was used to infer the short-latency connectivity between M1 and S1 sites. This technique was applied to investigate cortical dynamic connectivity and the causal relations between local cortical networks. Using both statistical and stimulation techniques, we found evidence for our hypothesis that sites in M1 and S1 that had similar receptive/projections fields were more likely to be functionally connected. Such connections may facilitate the synergistic coordination of movement with sensation.