Gracilo‐diencephalic relay cells: A quantitative study in the cat using retrograde transport of horseradish peroxidase

Abstract In a quantitative study in the cat, gracilo‐diencephalic relay cells were labeled by the use of retrograde axonal transport of horseradish peroxidase injected into the ventroposterolateral nucleus of the thalamus. An initial series comprising 22 animals with survival periods varying between 2 h and 4 days showed maximal labeling in the gracile nucleus after 24–48 h. The earliest appearing peroxidase‐positive neurons were found after only 6 h, implying a transport rate in the medial lemniscus of at least 100 mm/day. In a second series of five cats, which were killed 24 h after injection, serial sections from the gracile nucleus were embedded in Epon following peroxidase processing, and cut at 2 μm. By stratified random sampling, about 2,000 cells were selected for light microscopic examination. The total number of nerve cells in a single gracile nucleus was calculated to be about 50,000, out of which about 15,000 were retrogradely labeled. In agreement with previous reports, rostrocaudal differences were observed. Thus, less than a third of the neurons in the rostral part of the nucleus were peroxidase‐positive, whereas about half of those in the middle and caudal regions were retrogradely labeled. The total number of labeled neurons was, however, about the same in the rostral as in the middle‐caudal part of the nucleus. The size of the nerve cell bodies, measured as the cross‐sectional area in the nucleolar plane, differed significantly between labeled and unlabeled neurons. The estimated average diameter of the former was 22 μm and of the latter 17 μm. There was a considerable overlap between the two groups, however, and labeled cells as small as 12 μm in diameter were found. Mainly because of the uneven distribution of labeled neurons, cells in the rostral region were, on the average, significantly smaller than those in the middle and caudal regions; these mean diameters were calculated to be 18, 23, and 21 μm, respectively. The results of this study support the idea of a heterogeneous organization of the gracile nucleus. However, a much larger proportion of the gracilo‐diencephalic relay cells is situated in the rostral part of the gracile nucleus than has previously been thought. The concept that only medium‐sized neurons project to the diencephalon also seems to need revision. It is concluded that although gracilo‐diencephalic relay cells are, on the average, larger than the nonlabeled neurons, single nerve cells cannot be identified on the basis of size alone. The function of the unlabeled neurons is discussed. Although many of these might relay to various extra‐diencephalic sites, it is suggested that a large number are internuncial neurons.