An earlier retrograde double-labeling study in cat showed that up to 30% of the corticospinal neurons in the medial and anterior parts of the precruciate motor area represent branching neurons which project to both the spinal cord and the reticular formation of the lower brain stem. These neurons were found to be concentrated in the rostral portion of the motor cortex, from where axial and proximal limb movements can be elicited. In the present study the findings in the macaque monkey are reported. The fluorescent retrograde tracer DY was injected unilaterally in the spinal cord at C2 and the fluorescent tracer FB was injected ipsilaterally in the medial tegmentum of the medulla oblongata. In the contralateral hemisphere large numbers of single DY-labeled corticospinal neurons and single FB-labeled corticobulbar neurons were present. A substantial number of DY-FB double-labeled corticospinal neurons were also found, which must represent branching neurons projecting to both the spinal cord and the bulbar reticular formation. These neurons were present in: 1. The anterior portion of the "cingulate corticospinal area" in the lower bank of the cingulate sulcus; 2. The supplementary motor area (SMA); 3. The rostral part of precentral corticospinal area; 4. The upper portion of the precentral face representation area; 5. The caudal bank of the inferior limb of the arcuate sulcus; 6. The posterior part of the insula. In these areas 10% to 30% of the labeled neurons were double-labeled. The functional implications of the presence of branching corticospinal neurons in these areas is discussed.

Earlier studies showed that Nuclear Yellow (NY), True Blue (TB) and Fast Blue (FB) are transported retrogradely through axons to their parent cell bodies. NY produces a yellow fluorescent labeling of the neuronal nucleus at 360 nm excitation wavelength, while TB and FB produce a blue fluorescence of the cytoplasm at this same wavelength. Therefore, NY may be combined with TB or FB in double-labeling experiments demonstrating the existence of axon collaterals. However, retrograde neuronal labeling with TB or FB requires a relatively long survival time, while NY requires a short survival time since NY migrates rapidly out of the retrogradely labeled neurons. This complicates double-labeling experiments since TB and FB must be injected first and NY later, a short time before the animal is sacrificed. We report a new yellow fluorescent tracer which labels mainly the nucleus and migrates much more slowly out of the retrogradely labeled neurons than NY. This new tracer can be used instead of NY in combination with TB or FB in double-labeling experiments and unlike NY can be injected at the same time as TB or FB. The new tracer is a diamidino compound (no. 28826) which is commercially available. It will be referred to as Diamidino Yellow Dihydrochloride (DY . 2HCl). According to the present study DY . 2HCl is transported over long distances in rat and cat, and produces a yellow fluorescence of the neuronal nucleus at 360 nm excitation wavelength, resembling that obtained with NY. When combined with TB or FB, DY . 2HCl is as effective as NY in double labeling of neurons by way of divergent axon collaterals.

Layer VI of macaque striate cortex contains a number of large solitary neurones called Meynert cells. It has been shown earlier that these Meynert cells project to the posterior bank of the superior temporal sulcus (area V5), but it has also been shown that they project to the superior colliculus. In retrograde fluorescent double-labelling experiments, it was found that Meynert cells represent a class of neurones which distribute divergent axon collaterals to the posterior bank of the superior temporal sulcus and to the superior colliculus, i.e. to a distant cortical and a subcortical structure. This feature appears to be unique among projecting neurones in monkey visual cortex.

The fluorescent retrograde double-labeling technique has been used to determine whether corticospinal neurons in the cat sensorimotor cortex distribute collaterals to the lower brain stem reticular formation. In this study the fluorescent tracers Nuclear Yellow and Diamidino Yellow 2HCl were used in combination with Fast Blue. One tracer was injected unilaterally in the spinal cord and the other was injected ipsilaterally in the bulbar medial reticular formation. The distribution of the retrogradely labeled neurons was studied in the contralateral hemisphere. In the sensorimotor cortex a large population of neurons was found which were labeled from the spinal cord and were double-labeled from the brain stem. These branching neurons were concentrated in the rostromedial part of the area 4 and the adjoining lateral part of area 6. In this region the percentages of corticospinal neurons which were double-labeled from the brain stem ranged from 5% laterally to 30% medially. In two cats it was demonstrated by means of the anterograde transport of HRP that the corticobulbar fibers from this region which must include the corticospinal collaterals are distributed to the reticular formation of the lower brain stem. In view of the fact that the double-labeled neurons are concentrated in the anterior part of the motor cortex, those branching neurons are in all likelihood involved in the control of neck, back and shoulder movements. This control is probably exerted by way of two routes i.e. by way of the direct corticospinal connections to spinal interneurons, and by way of the indirect cortico-reticulospinal connections established by the cortical fibers to the bulbar reticular formation. The present findings suggest that this dual control may be exerted by one and the same cell.

No abstract