In rabbits, anesthetized with urethane‐chloralose. the activation of the granule cells of the dentate area by perforant path fibres from the entorhinal area was studied by intracellular and extracellular recording. Intracellular recording showed that the perforant path produced a large EPSP in the granule cells, followed after 2–4 msec by a large IPSP, lasting some 100 msec. Field potential studies indicated that the synapses responsible for the EPSP were located on the middle third of the dendritic tree, whereas the IPSP was generated by synapses at, or very close to, the cell bodies. The most likely explanation is that the inhibition is mediated by a recurrent inhibitory pathway, in which the collaterals of the granule cell axons excite the basket cells These are inhibitory in nature and send their axons to terminate upon the somata of many granule cells. The inhibition of the granule cells produced by the perforant path is resistant to strychnine in doses up to 0.6 mg/kg. The efficiency of the perforant path excitatory synapses was greatly increased by raising the rate of stimulation from 1 to 10 a second.

In rabbits anesthetized with urethane‐chloralose, the responses produced in the field CA1 by stimulation of the entorhinal area were studied with intracellular and extracellular recording. The most important pathway involved was found to be the three‐synaptic path: Entorhinal area — dentate granule cells— CA3 pyramids— CA1 pyramids. This conclusion is based upon the following experimental data: The main CA1 activity occurred later than expected for a monosynaptic path, it was always secondary to dentate granule cell discharges, it disappeared after a section severing the Schaffer collaterals at the CA2/CA1 border, it showed longer latency the longer the recording electrode was placed from the fimbria. The final link in this three‐synaptic pathway is the Schaffer collaterals. Their synapses produce a large negative field potential at the level of the profuse branching of the apical dendrites. The negativity is most likely due to the inward current flowing across the activated excitatory synapses. Extracellular recording of the population spike, produced by the Schaffer collateral synaptic activity, showed the spike to be initiated in that part of the dendritic membrane which borders the territory of the Schaffer synapses. The spike was conducted towards the soma with a speed of about 0.4 m/sec. Intracellular recording from cat CA1 pyramids gave as a rule no sign of soma epolarization in spite of the appearance of synaptically initiated action potentials. The results indicate that synapses located to the dendrites can initiate a local spike that may be conducted towards the soma with a low conduction velocity, and probably, with a low safety factor.

Previous papers in this series have been concerned with identification of primary afferents making monosynaptic connections with dorsal spino-cerebellar tract (DSCT) neurons and also with inhibitory action converging onto these neurons (LAPORTE. LUND- BERG and OSCARSSON 1956 a. b, LAPORTE and LUNDBERG 1956. LUXDBERG and OSCARSSON 1956.I n the present paper will be presented results showing that DSCT neurons receive excitatory action from interneurons. which