Motor neurons are lost during embryonic development, but it remains controversial whether motor neuron cell death occurs during postnatal life. In this study we investigated the effect of postnatal maturation on the number of intact spinal motor neurons in the rat using retrograde labelling with model-based counting, and an unbiased stereological counting technique. To determine the number of motor neurons innervating a specific forelimb muscle in rats of different postnatal ages FluoroGold was injected into the flexor carpi radialis. Before postnatal day 21 there were higher numbers of retrogradely labelled motor neurons than in adult rats, suggesting a ‘loss’ with postnatal maturation. This loss may be attributed to tracer diffusion to adjacent muscles and to the permeability of the muscle spindle capsule in younger animals. To obtain an unbiased estimate of the number of motor neurons in the C7 and C8 segments of the postnatal rat cervical spinal cord the fractionator/optical disector counting technique was used. This method did not show a loss of spinal motor neurons between birth and adulthood. The main conclusion from this study is that there is no loss of spinal motor neurons during postnatal maturation.

Eight normal human spinal cords were studied. Spinal segments were identified and embedded in paraffin wax. Serial cross sections were cut at 25 μm and stained by cresyl violet. Motor columns were reconstructed adapting Elliott's (1942) methods. Motor columns were classified into the medial and lateral divisions and were numbered sequentially from medial to lateral at the level of Cl. In the cervical cord, 8 motor columns were traced. Column 1, corresponding to the medial column, presented 3 subdivisions designated as 1a, 1b and 1c with ventral, dorsal and lateral positions respectively. Columns 1a and 1b extended throughout the cervical region while 1c was confined to 3rd, 4th and 5th cervical segments. At the level of C3, 1c was a discrete column situated lateral to 1a and 1b but at C4 and C5 it became displaced medially close to the medial margin of the ventral horn. In cross section, it presented smaller medial and large lateral part. With the help of clinical and developmental evidence an attempt was made to correlate column 1c with the phrenic nucleus.