A rose by any other name... on 'naming of neurones' by Rowe and Stone.

Rowe and Stone dispose of descriptive terminologies for retinal units as 'essentialist' and argue for a 'neutral' multifactorial system redefined from previous W, X. Y nomenclatures. It is argued here that descriptive terminology need not be either essentialist or undesirable and that the W, X, Y terminology is too confused with false associations to be satisfactory. The synthetic power of the W, X, Y nomenclature was lost with the discovery of concentric W cell receptive fields, the projection of the W cells to the LGN and the suggestion of overlap between X and W conduction velocity groups. It was made methodologically unsound by its employment for the naming of Nissl stained soma classes before the identity of individual somas with the physiological entities can be demonstrated. Although now redefined, the identity of symbols in the several vintages of the W, X, Y system leads to inevitable confusion. The current tripartite division of a multidimensional quality space without cluster analysis is not neutral because it emphasises conduction velocity but represses the diversity of W receptive fields. The concise, apt and memorable descriptive receptive field terminologies are more desirable in that they are operationally based and approximate the unit's transfer function--the parameter most significant to the central nervous system in its information processing task; thus they are more suited for cross species comparison because of a probably bountiful source of functional hypotheses. The lack of taxonomic neutrality in the W, X, Y system has recently been emphasised by its failure to adequately classify rabbit retinal receptive field categories.