Purpose/Hypothesis: Our prior work established that reticulospinal neurons have activity related to preparation for reaching (P cells), performance of the reaching movement (M cells), or a combination of the two (MP cells). This study tests the hypothesis that classifcation of a cell as M, P, or MP can be explained by whether muscles influenced by that cell have activity that can be similarly classifed. Number of Subjects: Two adult male M. fascicularis monkeys. Materials/Methods: As the awake animals performed a reaching task, neural activity was recorded with microelectrodes in the pontomedullary reticular formation; chronic indwelling EMG electrodes recorded activity from flexor and extensor muscles of the scapulothoracic, shoulder, elbow, and wrist joints of both arms. The target was indicated for each trial by a colored box (instruction cue) shown on either side of a touch screen; the reach could not begin until a go cue was delivered. A repeated-measures ANOVA was used to classify cells as M, P, or MP. Cells with activity changing from baseline only after the instruction were P cells. M cells had activity modulated only during movement, and MP cells had activity modulated after the instruction and further modulated during movement. The same approach was used to classify muscles as M, P, or MP. Motor outputs of the cells were measured by the EMG responses to microstimulation at the recording sites. Results: Among muscles sampled, only biceps consistently showed M-like activity. All other muscles had MP-like activity patterns. Among neurons, however, there were 46% M cells, 29% MP cells, and 26% P cells. Stimulation in vicinity of the recording sites produced short-latency (<; 15 ms) EMG responses in the upper limb muscles. Regardless of the type of cell recorded at a given site, the effect of stimulation was consistent: ipsilateral limb flexors were facilitated, ipsilateral extensors were suppressed, and the opposite pattern was found contralaterally. Conclusions: The classifcation of a cell as M, MP, or P cannot be simply explained by the pattern of recruitment for muscles to which that cell projects. The vast majority of muscles recorded in the limbs had activity that was modulated during preparation and during movement, but many of the neurons had activity modulated only during movement. Thus, neural control signals related to the preparation versus performance of reaching may be segregated or combined at the level of the brainstem, but are typically combined in the muscles as the “final common pathway” of Sherrington. Clinical Relevance: Even in the brainstem, motor commands required to prepare for movement are often segregated from commands for the movement itself. Data in the literature now shows that preparatory and movement related motor command signals can be distinguished at multiple levels of the CNS, from cortex to spinal cord. Terefore, it is important as physical therapists that, when devising a task-specific rehabilitation program, we train for the preparation component of the task, not just the movement itself.