How are the neurons that directly influence the motoneurons of a muscle distributed in the primary motor cortex (M1)? To answer this classical question we used retrograde transneuronal transport of rabies virus from single muscles of macaques. This enabled us to define cortico-motoneuronal (CM) cells that make monosynaptic connections with the motoneurons of the injected muscle. We examined the distribution of CM cells that project to motoneurons of three thumb and finger muscles. We found that the CM cells for these digit muscles are restricted to the caudal portion of M1, which is buried in the central sulcus. Within this region of M1, CM cells for one muscle display a remarkably widespread distribution and fill the entire mediolateral extent of the arm area. In fact, CM cells for digit muscles are found in regions of M1 that are known to contain the shoulder representation. The cortical territories occupied by CM cells for different muscles overlap extensively. Thus, we found no evidence for a focal representation of single muscles in M1. Instead, the overlap and intermingling among the different populations of CM cells may be the neural substrate to create a wide variety of muscle synergies. We found two additional surprising results. First, 15-16% of the CM cells originate from area 3a, a region of primary somatosensory cortex. Second, the size range of CM cells includes both "fast" and "slow" pyramidal tract neurons. These observations are likely to lead to dramatic changes in views about the function of the CM system.