Spatial convergence and divergence between cutaneous afferent axons and dorsal horn cells are not constant.

We have proposed a quantitative model of the development of dorsal horn cell receptive fields (RFs) and somatotopic organization (Brown et al. [1997] Somatosens. Motor Res. 14:93-106). One component of that model is a hypothesis that convergence and divergence of connections between low-threshold primary afferent mechanoreceptive axons and dorsal horn cells are invariant over skin location and dorsal horn location. The more limited, and more easily tested, hypothesis that spatial convergence and divergence between cutaneous mechanoreceptors and dorsal horn cell are constant was examined. Spatial divergence is the number of dorsal horn cells whose RFs overlap the RF center of a primary afferent, and spatial convergence is the number of afferent RF centers that lie within the RF of a dorsal horn cell. Innervation density was determined as a function of location on the hindlimb by using peripheral nerve recording and axon counting. A descriptive model of dorsal horn cell receptive fields (Brown et al. [1998] J. Neurophysiol. 31:833-848) was used to simulate RFs of the entire dorsal horn cell population in order to estimate RF area and map scale as a function of location on the hindlimb. Previously reported correlations among innervation density, map scale, and RF size were confirmed. However, these correlations were not linear. The hypothesis that spatial convergence and divergence are constant was rejected. The previously proposed model of development of dorsal horn cell somatotopy and RF geometries must be revised to take variable spatial convergence and divergence into account.