Cerebellar Purkinje Cell Projection to the Peripheral Vestibular Organ in the Frog Abstract. Neurones located 200 to 300 microns from the surface of the auricular lobe of the frog cerebellar cortex, and identified as Purkinje cells, were activated antidromically from the eighth cranial nerve. A parallel anatomical study confirmed the existence of this projection. On the basis of these findings the existence of a cerebello-vestibular efferent system is postulated, the precise significance of which is as yet unclear. However, since Purkinje cells in other species have an inhibitory action on their target cells, the Purkinje efferent system to the vestibular organ may have an ac- tion similar to that ascribed to the olivo-cochlear bundle upon the cochlea, that is, to serve as an inhibitory control system. An outstanding characteristic of the neural organization of the cerebellum throughout evolution has been the presence of only one efferent system from the cerebellar cortex—the Purkinje cell axons. While most of these axons project directly to the underlying cerebel- with micropipettes filled with 444 NaCl ascribed to the antidromic invasion of Purkinje cells (7, 12). and 3M KCI, respectively. The average direct-current resistance of the extra- Giant extracellular positive-negative cellular recording electrode was 5 meg- action potentials were recorded from the frog auricular lobe after threshold electrical stimulation of the eighth nerve (Fig. 1, B and C). Four criteria were used to identify these action po- ohms; that of the intracellular one was 10 megohms. As in previous experiments (8), the threshold current for electrical stimulation of the eighth nerve was measured attheinitiation of the experiment and was neverincreased above 2.5 times threshold. Electrical activation of the eighth nerve evokesactivation of Purkinje cells in the frog auricular lobe via climbing and mossy fibers (8). The activation of cat (9) and frog Purkinje cells (70) by climbing fibers has been characterized by an all-or-none burst. of spikes having a very regular pattern of discharge and an almost constant latency. Activation by mossyfibers, on the other hand, produces, in the cat, activation of Purkinje cells which has variable tentials as those generated by Purkinje cells: (i) their antidromic activation from extracerebellar regions (6, 10, 12, 13), (ii) their characteristic large extra- cellular action potentials, the so-called “giant spikes” (73), (iii) their ortho- dromic activation by stimulation of parallel fibers (7, 10, 14), and (iv) in the auricular lobe, their depth from the piaglial membrane, that ‘is, 200 to 300 » from the surface (8). Forty-seven cells were studied; their latency is illustrated in Fig. 1K. A large numberof Purkinje cells activated at longer latencies were not included since their latency (71). Similar findings have been reported for the frog cerebellum (7, 8. 12), Our report, however, deals almost latencies put them outside the scope of tion of short latency, which has been tion of the eighth nerve (Fig. 1K). The exclusively with Purkinje cell activa- this paper. There appear to be two groups of action potentials activated by stimula- G lar nuclei (7, 2), anatomists have de- scribed many extranuclear projections in different species (2, 3). Recent physi- neat =x ological evidence has confirmed the existence of these extranuclear projections (4) and, in addition, has shown that Purkinje cells have an inhibitory E action on cerebellar and vestibular nuclei (4). Furthermore, the axon collaterals of Purkinje cells are inhibitory on cerebellar basket cells (5) and di- I (fe F Ke So far, however, the cerebellar pro- mary sensory systems. Our experiments, on the other hand, indicate that, in the frog, auricular Purkinje cells project directly to the peripheral vestibular organ. The general experimental procedures have been described (7, 8). Bullfrogs (Rana catesbiana) were anesthesized with pentobarbital sodium (60 mg per kilogram of body weight), and the eighth nerve was stimulated electrical- ly by means of bipolar electrodes lo- cated extracranially in its anterior or posterior branches (Fig. cellular and intracellular 1A). Extra- recordings from Purkinje cells were performed 1328 | —— 4 jections described have been restricted between cerebellar efferents and pri- 4h J rectly to Purkinje cells themselves (6). to systems involved in motor control, no direct relation having been found a Smee Vmeac tov Vmaec ro nv | 5 muer Fig. 1. (A) Diagram of frog brainstem, labyrinth, and experimental arrangement; BC, bipolar ganglion cells; CER, cerebellum; CF, climbing fiber; GC, granule cell; ME, recording microelectrode; MF, mossy fiber; OL, optic lobe; PC, Purkinje cell; RC, receptor cell; .S, peripheral nerve stimulating electrode; V, trigeminal nerve; VI/, facial nerve; VIII, stato-acoustic nerve; PB, posterior branch; arrows indicate the direction of impulse conduction. (B~F) Antidromic extracellular action potentials from Purkinje cells in the auricular lobe, recorded at a depth of 250 » and evoked by stimulation of the eighth nerve. (B and C) All-or-none spike responses at threshold intensity from two different Purkinje cells. (D-F) Purkinje cell action potentials evoked by activation of the eighth nerve at 1.3 times the threshold. (D) Control. (E) Refractory period for antidromic invasion of the Purkinje cell after double shocks with a short interval between stimuli. (F) As in E, but with longer interval between stimuli, showing second Purkinje spike. (G and H) Extracellular and intracellular recording from a Purkinje cell in the auricular lobe at a depth of 300 » activated antidromically and synaptically by stimulation of the eighth nerve. (G) Extracellular record. (H and I) Intracellular records of the same cell showing the antidromically excited action potential and an excitatory postsynaptic potential. (J) At slightly higher stimulus intensity the Purkinje cell shows also a transsynaptic response. (K) Latency histogram of the early spike responses after stimulation of the eighth nerve. Abscissa, time in milliseconds. Ordinate, number of Purkinje cells recorded. The arrows in B-J indicate stim- ulus artifact location. Time and voltage calibration as indicated. The polarity is ex- pressed by the positive and negative signs on the amplitude calibrating bar. SCIENCE, VOL. 158