ABSTRACT
The columnar structure of the cerebral cortex has now been appreciated for well over 40 years as a fundamental architectural feature of cerebral cortex organization (Mountcastle, 1957; Powell and Mountcastle, 1959; Hubel and Wiesel, 1962). Cells situated in a column, perpendicular to the cortical surface, share a number of important functional properties such as receptive field position or stimulus selectivity. While we now appreciate the limitations of this scheme (in that cells situated in different layers are in fact known to differ to some degree in their connections and receptive field properties), it is still clear that in many important respects neurones located in a single column from pia to white matter share key receptive field properties. This columnar organisation is mirrored by the anatomical organisation of cerebral cortex: in any area one chooses to investigate, there is a dense vertical focus of connections linking cells at different depths within a single column of cortex. This has been appreciated for even longer than the physiological columnar organisation: golgi
(Cajal, 1922; O’Leary, 1941; Lund and Boothe, 1975). Connections running horizontally, parallel to the cortical surface, are difficult to trace with the Golgi technique. In such studies, axons are generally traced no more than a few hundred microns from the soma. However, it was clear, even from these relatively crude techniques, that there did exist neural circuits to mediate communication among cells at different locations across the cortex (i.e. in different columns).
