On the concept of spinocerebellum

From an evolutionary point of view, the cerebellum consists of three regions: the archicerebellum, which is dominant in aquatic animals; the paleocerebellum, which developed during terrestrial locomotion in parallel with the spinal cord; and the neocerebellum, which dominates in mammals and developed along with the cerebral hemispheres. The paleocerebellum, identified mainly with the vermis, is often called the spinocerebellum, although spinal afferents project only to the vermal area of the anterior lobe and to a subregion of the posterior lobe. Therefore, the name spinocerebellum should not be synonymous with vermis and should not imply a lack of cortical afferents. The article by Coffman et al. (1) demonstrated, by means of retrograde viral tracing, that the cerebellar vermis—predominantly the posterior region—receives projections from the motor areas in the cerebral cortex. The authors concluded that their results challenge the classical view on the vermis and indicate that this structure should no longer be considered as entirely isolated from the cerebral cortex. In the interest of the scientific debate regarding how afferent information is organized over the cerebellar cortex, we would like to mention that (i) projections from the motor cortex to the vermal area of the anterior lobe and (ii) their complex distribution have been described in an article published in 1968 (2). In this study we investigated the spinal and cortical projections to the cat anterior lobe both in the vermis and in the pars intermedia along lobuli I–V. We analyzed electrophysiologically the projections mediated by the mossy fiber (MF) and the climbing fiber (CF) pathways from the forelimb (F-CORT) and the hindlimb (H-CORT) areas of the sensorimotor cortex and made a comparison with the known projections from forelimb and hindlimb peripheral nerves. Stimulation of F-CORT and H-CORT evokes in the anterior lobe an early response at a latency of 3–3.5 ms due to the MF input and a later response at a latency of 13–16 ms due to the CF input.

In the vermis we found that (i) both cortical areas project through MF and CF afferents; (ii) the topography of MF and CF responses overlaps; (iii) at least for the sharper CF projections the H-CORT is represented as a lateral longitudinal strip neighboring a more medial one under the control of the F-CORT; and (iv) there is a coincidence between activity evoked from each cortical area and that from stimulation of forelimb or hindlimb peripheral nerves. These results have been quoted in many books and reviews, including works by Brodal (3) and Brooks (4).

In conclusion, the article by Coffman et al. (1) added many interesting details regarding the relationships between the cerebral cortex and the vermis, and their results agree with the old view that the name spinocerebellum has its origin in the fact that it developed in parallel with the spinal cord and that during evolution it acquired a corticocerebral control input along with brain development. We are confident that the present scientific considerations will add interest to the interpretation and discussion of the results of Coffman et al. (1).