BACKGROUND: Targeting of the motor thalamus for the treatment of tremor has traditionally been achieved by a combination of anatomical atlases and neuroimaging, intraoperative clinical assessment, and physiological recordings. OBJECTIVE: To evaluate whether thalamic nuclei targeted in tremor surgery could be identified by virtue of their differing connections with noninvasive neuroimaging, thereby providing an extra factor to aid successful targeting. METHODS: Diffusion tensor tractography was performed in 17 healthy control subjects using diffusion data acquired at 1.5-T magnetic resonance imaging (60 directions, b value = 1000 s/mm2, 2 × 2 × 2-mm3 voxels). The ventralis intermedius (Vim) and ventralis oralis posterior (Vop) nuclei were identified by a stereotactic neurosurgeon, and these sites were used as seeds for probabilistic tractography. The expected cortical connections of these nuclei, namely the primary motor cortex (M1) and contralateral cerebellum for the Vim and M1, the supplementary motor area, and dorsolateral prefrontal cortex for the Vop, were determined a priori from the literature. RESULTS: Tractogram signal intensity was highest in the dorsolateral prefrontal cortex and supplementary motor area after Vop seeding (P < .001, Wilcoxon signed-rank tests). High intensity was seen in M1 after seeding of both nuclei but was greater with Vim seeding (P < .001). Contralateral cerebellar signal was highest with Vim seeding (P < .001). CONCLUSION: Probabilistic tractography can depict differences in connectivity between intimate nuclei within the motor thalamus. These connections are consistent with published anatomical studies; therefore, tractography may provide an important adjunct in future targeting in tremor surgery.