Abstract.-Centrifugal ("antidromic") discharges in cat sensory fibers are observed consistently in a variety of experimental preparations and with many different surgical and recording techniques. As is well known, they can be either "spontaneous" or induced by afferent volleys in other sensory fibers. In addition, it is shown here that they can be elicited by antidromic motoneuron activation when the latter is conditioned by natural sensory stimuli or by shocks to the dorsal roots. The latency of the centrifugal dorsal root response to ventral root stimulation is shorter than that of the monosynaptic reflex mediated by the same fibers. An "antidromic" coupling, probably of an electrical nature, between motoneurons and presynaptic terminals is postulated.The presence of nerve impulses leaving the spinal cord via dorsal roots (centrifugal sensory discharges) was discovered almost 80 years ago'0 and has been repeatedly confirmed by later investigators.2 5,6,9,12,18,19 Nevertheless, in spite of all the evidence encountered, it may be said that at present these discharges are still considered by neurobiologists to be a most uncommon type of sensory fiber activity. Although the reasons for this attitude are varied,1' 11 one of the most important probably is the disturbing nature of the phenomenon vis-A-vis classical neuronal theory. However, even if one does not accept the existence of this type of dorsal root (DR) activity as a normal physiological phenomenon, the question about the mechanism producing it remains perfectly valid.Dorsal root volleys generate a long-lasting depolarization, the dorsal root potential, in the activated as well as in neighboring DR fibers." 15 During the beginning of the depolarization, a mass discharge of "antidromic" sensory impulses, the DR reflex, can often be observed.19 Because of its relation to the DR potential, this type of efferent DR activity has been proposed to be a direct result of the presynaptic depolarization.4' 20 A marked increase in electrical excitability, maximal at the terminals and with a time course similar to the DR potential, has also been found in these afferent fibers.2'Although the concept of depolarization of presynaptic terminals has been linked with presynaptic inhibition, the mechanism by which the DR potential is generated still remains unclear. Eccles7' 8 has proposed a multisynaptic neuronal chain ending in axo-axonic synapses (chemical) on the primary afferent fibers. On the other hand, Barron and Matthews3 proposed that the depolarization that involved both stimulated and adjacent passive fibers was due to the extracellular accumulation of some ionic substance liberated by the excited terminals; some type of ionic interaction would then be present inside the cord.This communication is concerned with some aspects of the centrifugal dis-58