Search

Responses of lateral hypothalamic neurons to 8 odors were studied in chronic unanesthetized old world monkeys (Macaca irus). Many neurons (54.5%) responded to a single odor only, and the number of neurons responding to 2, 3 and 4 odors decreased successively. No neuron responded to as many as 5 odors. Thus, the presence of olfactory input and a highly discriminative ability for odors were found in the lateral hypothalamic area (LHA). Neuronal responses to the same odors were also studied in the septum (Spt). In anesthetized old world monkeys, evoked potentials were recorded in the LHA and in areas of the Spt and the nucleus accumbens (Acc) during stimulation of the olfactory bulb (OB). When the Spt (and probably the Acc with it) was subsequently destroyed, OB-evoked potentials in the LHA disappeared. Next, by injecting horseradish peroxidase (HRP) into the LHA, an olfactory pathway to the LHA was examined. Labeled neurons were found mainly in the Spt and the Acc, and only partly in other areas. However, labeled neurons were scarcely found in the prepyriform (PPF)-entorhinal (ER) area or in the olfactory tubercle (OT). The present study thus shows that an olfactory pathway to the LHA passes through the Spt and probably also the Acc, but not through the PPF-ER areas nor through the OT in the old world monkey.

The neuronal pathways responsible for the fast inhibitory postsynaptic potentials (IPSPs) elicited in principal cells in the pyriform cortex (PC) by volleys from the olfactory bulb (OB), the lateral olfactory tract (LOT), the anterior commissure (AC), and the deep-lying structures of the PC (DPC) were studied in the rabbit. The central latencies of the fast IPSPs (measured from the onset of the monosynaptic excitatory postsynaptic potential (EPSP) elicited by volleys through the LOT) ranged between 3.0 and 9.3 ms (5.5 +/- 1.3 (SD) ms; n = 54) in the case of OB shocks and between 4.5 and 6.5 ms (5.1 +/- 0.7 (SD) ms; n = 7) in the case of LOT shocks. The onset latencies of the fast IPSPs were between 2.5 and 11.8 ms (5.1 +/- 1.8 (SD) ms; n = 66) in the case of DPC shocks and between 3.5 and 10.1 ms (5.8 +/- 1.5 (SD) ms; n = 61) in the case of AC shocks. The conditioning OB or LOT shocks almost completely eliminated the LOT-evoked fast IPSP when the testing shock was applied at the peak period of the conditioning slow IPSP. The conditioning OB shocks also eliminated the initial part of the OB-evoked fast IPSP, leaving the later part of the fast IPSP almost unchanged. Thus, the onset latency of the OB-evoked fast IPSP was lengthened by 7.1 +/- 2.9 (SD) ms (n = 35) by the conditioning OB shock. The conditioning OB or DPC shocks left the peak amplitude of the DPC-evoked fast IPSP almost unaffected. Similarly, the conditioning OB or AC shocks left the peak amplitude of the AC-evoked fast IPSP almost unaffected. The conditioning OB, DPC, or AC shocks had only a slight influence on the onset latency of the DPC- or AC-evoked fast IPSPs. Rhythmical steps at intervals of 3-5 ms were observed in the rising phase of the OB-evoked fast IPSP. This was interpreted as a result of a repetitive impingement of interneuronal discharges on the impaled cells. Spatial facilitation was observed among the fast IPSPs evoked by volleys from the OB, DPC, and AC when shocks were applied at suitable intervals. A slight facilitation was also seen between the LOT-evoked fast IPSP and the DPC- or AC-evoked fast IPSP. These results were interpreted as a result of the convergence of excitatory synaptic inputs onto the presumed inhibitory interneurons from the four structures of the brain. A temporal facilitation of the fast IPSPs was observed when the OB, DPC, or AC shocks were applied repetitively at short intervals. This suggests a temporal facilitation of the spike discharges of the presumed inhibitory interneurons under similar conditions. From these results, criteria were determined for identifying the inhibitory interneurons.

From the results of our electrophysiological and HRP studies in the old world monkey, multiple olfactory pathways have been clarified. The old world monkey has two neocortical olfactory areas, but no functional vomeronasal system. The response patterns to odors in various olfactory areas have also been studied. On the other hand, in the rabbit (Onoda and Iino, 1980) and dog (Onoda et al., 1981, 1982), which do have active vomeronasal systems, only one neocortical olfactory area was found. This important difference had already been indicated in three previous papers in which Takagi (1979, 1980, 1981) theorized that mammals can be divided into two groups according to their olfactory nervous mechanisms. One group includes old world monkeys, higher primates and man, and the other new world monkeys and lower mammals.

1. Evoked potentials restricted to the magnocellular portion of the mediodorsal nucleus (MDmc) of the thalamus were recorded after stimulation of the olfactory bulb (OB) and the posterior orbital cortex of the frontal lobe (OFC). Potentials evoked by stimulation of OB were probably trans-synaptically elicited, while potentials evoked by stimulation of OFC were probably a result of antidromic activation. 2. The area in which stimulation could elicit antidromic evoked potentials in MDmc was located in the centroposterior portion of OFC (CPOF). This area corresponds approximately to Walker's (80) area 13 and to von Bonin and Bailey's (9) area FF, and is situated medial and just anterior to a previously identified olfactory area, the lateroposterior portion of OFC (LPOF), which receives olfactory impulses through the hypothalamus. 3. Using extracellular microelectrodes, 58 neurons that responded with short latencies to OFC stimulation were identified in MDmc. To determine whether these neurons were activated antidromically by CPOF stimulation, three conventional neurophysiological criteria were applied; 20 of 58 neurons satisfied all the three criteria. Hence, they were concluded to be thalamocortical relay (TCR) neurons. 4. Intracellular recording of MDmc neurons disclosed that CPOF stimulation elicits an antidromic spike potential accompanied by an afterhyperpolarization. This hyperpolarization was presumed to be due to concurrent stimulation of inhibitory orbitothalamic fibers. It was also shown that EPSP-like depolarizations with superimposed spike potentials often occurred in the middle of the afterhyperpolarization. 5. Intracellular recording of MDmc neurons strongly suggested that the remaining 38 neurons that did not satisfy one of the three criteria were also TCR neurons. 6. These studies provide electrophysiological evidence for a transthalamic olfactory pathway from OB through MDmc to CPOF. 7. Using an extracellular recording technique, responses of neurons to eight odors were examined in CPOF and MDmc of unanesthetized awake monkeys. When these results were compared with the responses of neurons to the same odors in OB, prepyriform-amygdaloid area, and LPOF, it was concluded that the newly found transthalamic olfactory pathway to CPOF is very different in function from the previously demonstrated transhypothalamic olfactory pathway to LPOF.

Afferent fiber projections to the two orbitofrontal olfactory areas of monkeys were studied using the horseradish peroxidase (HRP) technique. After injections of HRP into the lateroposterior (LPOF) or centroposterior (CPOF) area of the orbitofrontal cortex, some differences were found in the distribution of labeled cells between the projections to the LPOF and CPOF. These results, along with those of previous electrophysiological investigations, suggest the following conclusions: (1) the extrathalamic olfactory pathway to the LPOF identified by Tanabe et al. has relay neurons primarily in the substantia innominata and the amygdala and, secondarily, in the prorhinal cortex and the hypothalamus; (2) direct fibers to the LPOF from the amygdala and the prorhinal cortex pass through the areas ventral to the thalamus; (3) the transthalamic olfactory pathway to the CPOF identified by Yarita et al. has relay neurons concentrated primarily in the magnocellular portion of the mediodorsal nucleus of the thalamus.

The olfactory epithelia of the bullfrog and the carp were stimulated with various odorants in the gaseous and/or aqueous phases. Some ionic solutions were also applied. 1. By recording the induced waves in the olfactory bulb, it was shown that the bullfrog responds to odorants in the two phases, but the carp can respond only to odorants in the aqueous phase. 2. Comparison of the stimulative effects of the same odorants in the two phases were made in the bullfrog. Although various response patterns to gaseous and aqueous stimuli of single cells in the olfactory bulb were found in most cases, a difference in the application methods of vapors and solutions made such comparison difficult. 3. The same single cells in the olfactory bulb usually responded to various odorous vapors and solutions as well as to ions in different patterns. When the responses to the vapors and solutions of the same odorants of different concentrations were compared, different patterns were also found.

Effects of various gustatory stimulants upon the olfactory spithelia were examined in the olfactory bulb of the bullfrog and the carp. 1. The olfactory epithelia of the two animals responded to the salty, bitter- and acid-tasting substances, but not to the sweet ones. 2. The olfactory epithelium of the bullfrog responded immediately to sodium solutions of high concentrations (the "initial response"), but the response to those of low concentrations showed long latency (the "delayed response"). In the carp, the "initial response" was found, while the "delayed response" was not in most cases. A "negative" delayed response was found only infrequently. 3. Responses only to high alkali or acid solutions were found in the two animals. 4. When 0.05 M NaCl was added to HCl solutions, an enhancing effect was found in the bullfrog, while a reducing effect occurred in the carp. On the contrary, when 0.05 M NaCl was added to NaOH solutions, an enhancing effect occurred in the carp, while no consistent result was founding the bullfrog. 5. Many amino acids were effective stimuli in the bullfrog, but only betaine and 1-aspartic acid were found effective in the carp. 6. Changes in temperatures beyond 35 degrees C or under 10 degrees C elicited responses. Mechanical stimuli were effective in the carp, but not in the bullfrog. 7. The "water response" was found in the bullfrog, but not in the carp. 8. Sensitivities of the olfactory epithelia of the two animals were compared and discussed.

Interneurons mediating the fast IPSPs in principal cells in the pyriform cortex (PC) of the rabbit were searched for using criteria derived from the analysis of the properties of the fast inhibitory postsynaptic potentials (IPSPs). Thirty units were identified as inhibitory interneurons. The interneurons were activated synaptically by volleys from the olfactory bulb (OB), the lateral olfactory tract (LOT), the anterior commissure (AC), and deep-lying structures of the PC (DPC). The interneurons showed a tendency to discharge repetitively in response to shocks applied to these structures of the basal forebrain (OB, LOT, AC, and DPC). The conditioning OB shocks eliminated the testing LOT-evoked discharges of the interneurons. The conditioning OB shocks eliminated the initial part of the testing OB-evoked discharges, leaving the later part relatively unchanged. On the other hand, the conditioning OB shocks did not completely eliminate the testing DPC- or AC-evoked discharges. A temporal facilitation of discharges in the interneurons was observed in response to volleys from the OB, DPC, or AC. A spatial facilitation of discharges in the interneurons was observed in response to a combination of shocks applied to the OB, DPC, and AC. The interneurons were recorded at depths 525-2,755 microns deep to the turnover point of the component 2 wave of field potentials evoked by volleys through the LOT fibers. They were located mostly in the deeper part of layer III of the PC. Intracellular recordings from the presumed inhibitory interneurons showed that OB stimulation elicited two successive excitatory postsynaptic potentials (EPSPs) on which the bursting discharges were superimposed. These EPSPs were followed by a long-lasting hyperpolarizing potential. A comparison of the latencies of the antidromic activation of the principal cells and the synaptic activation of the inhibitory interneurons following OB or DPC stimulation suggested that the inhibitory interneurons are activated at least partly through the axon collaterals of the principal cells, which project their main axons to the OB or DPC. A circuit diagram was proposed for the neuronal pathways responsible for the fast IPSPs of principal cells in the PC.

In the orbitofrontal olfactory area (LPOF) which was delineated in a previous paper, the capacity for odor discrimination was studied and compared with that in the anterior pyriform cortex (AP), the medial portion of the amygdala (MA), and the olfactory bulb (OB). Unanesthetized monkeys were used and eight odors were applied. 1. In the OB, 12.5% of the cells responded to only one odor, and the cells which responded to five odors were most numerous (25%). The total of the cells which responded to two, three, and four odors was 52%, which was less than the total of the cells responding to three, four, and five odors (67.5%). A small number oc cells responded to all eight odors (2.5%). The responses were classified as an increase (+type), a decrease (-type), or no change (no-type) in the rate of spike discharge. 2. In the AP and MA, no difference in the response patterns was found. The cells which responded to only one odor were 12.3% of the total, and the cells which responded to three different kinds of odors were most numerous (34.3%). The total of the cells responding to two, three, and four odors was 80%, much more than that in the OB. In addition, no cell responded to all eight odors. Concerning the response types, an increase followed by a decrease, or vice versa, in the rate of spike discharges (mixed-type) was observed which did not appear in the OB. Thus, an advance was found in the processing of olfactory information when compared with the OB. 3. A most striking finding in the LPOF was that 50% of the cells responded to only one odor. The cells which responded to two, three, and four odors decreased in this order, and no cell responded to more than five odors. These cells never responded to light or sound. 4. Using three very similar odors and five very different odors, it was apparent that the ability to discriminate odors of the same category is far more advanced in the LPOF than in the lower olfactory areas; and, in contrast, the lower olfactory areas also play a significant role in the discrimination of odors which belong to different categories. 5. It was concluded that the capacity for odor discrimination definitely improves along the olfactory nervous system from the lower to the higher areas. It is highly probable that a fine and sepcific discrimination of odors is performed in the LPOF.

Recent work in our laboratory has demonstrated that prostaglandin (PG) D2 and the enzyme activities for its biosynthesis and inactivation are highly concentrated in the olfactory bulb and that the mitral cell layer of the bulb is enriched with PGD2-binding protein. We therefore investigated the role of PGD2 in the processing of odor signals in the rabbit olfactory bulb by an electrophysiological technique. Iontophoretic (-100 nA, 20 s), intra-arterial (0.0125-0.1 mg/kg) and intravenous (i.v., 0.05-0.3 mg/kg) administration of PGD2 enhanced and prolonged the responses of mitral cells to some of the olfactory stimuli tested. The extent and duration of granule cell inhibition of mitral cells were assessed by recording field potential responses in the bulb to paired lateral olfactory tract volleys. The i.v. administration of indomethacin or diclophenac, both of which are inhibitors of PG biosynthesis, resulted in prolongation of the granule cell inhibition of mitral cells without any significant change of the conditioning amplitudes. It also caused the reduction of the spike responses of mitral cells to olfactory stimuli. After treatment with indomethacin, the i.v. administration of PGD2 (1 mg/kg) rapidly reduced the duration of the granule cell inhibition of mitral cells. These results indicate that PGD2 plays a modulatory role in the mitral cell responses to odor stimuli by suppressing the inhibitory synaptic inputs from granule cells to mitral cells.

This paper reviews work on olfactory function performed in the author's laboratory over the last 10 years. The following aspects of this work are covered. Neocortical olfactory areas were studied in old world monkeys. Olfactory responses were found in the lateroposterior and centroposterior portions of the orbitofrontal cortex (LPOF and CPOF). The routes of the olfactory nerve pathways to the LPOF and the CPOF were examined. An olfactory pathway to the lateral hypothalamic area (LHA) was also studied. Using unanesthetized monkeys, information processing of odors was studied in the OB, PPF-MA, LPOF, MDmc, CPOF, and LHA. In the LPOF and LHA, half or more of the cells responded differentially to one odor. We have thus been able to clearly demonstrate discrimination of odors at the cell level in these areas.

An olfactory projection area was studied in monkeys anesthetized with Nembutal. 1. Evoked potentials were recorded when the olfactory bulb (OB) was electrically stimulated in the lateroposterior portion of the orbitofrontal cortex (LPOF). However, those potentials disappeared when the anterior pyriform cortex (AP) (probably together with the medial portion of the amygdala (MA)) was aspirated or electrically destroyed. 2. In nearly the entire hypothalamic region, evoked potentials were recorded by the same stimulation of the OB. When the hypothalamic region was stimulated, evoked potentials were recorded in the LPOF. 3. The evoked potentials in the LPOF due to the OB stimulation never disappeared even when the thalamus was extensively aspirated or destroyed electrically, but they did disappear when the anterolateral and dorsoposterior portions of the hypothalamus were absorbed or electrocoagulated. 4. Evoked potentials in the mediodorsal nucleus (MD) of the thalamus were recorded when the OB was stimulated. When this nucleus was stimulated, evoked potentials were observed in the broad extent of the orbitofrontal cortex anterior to the LPOF, but never in the LPOF itself. 5. Monkeys were conditioned to discriminate two odors. When the LPOF was removed, such ability strikingly decreased; but when other areas in the prefrontal cortex were removed, the ability decreased only slightly. 6. It was concluded that there exists an olfactory pathway from the OB to the LPOF through the AP (and probably the MA) and the hypothalamus, but none through the thalamus, and that the LPOF plays an important role in the discrimination of odors. 7. It was proved that the entorhinal cortex (ER) is neither located as an intermediate olfactory area nor is it situated as a higher area than the LPOF in the newly found olfactory pathway stated above. It may be a link between the high olfactory area and the limbic system.

1. In the rabbit olfactory bulb, intracellular potentials were recorded from mitral cells and from neurones in the granule cell layer (g.c.l.) following lateral olfactory tract (l.o.t.) stimulation. 2. Most recordings from mitral cells showed large (5-21 mV) and prolonged (60-650 msec) i.p.s.p.s subseuqent to the antidromic spikes. These i.p.s.p.s decreased in amplitude and then reversed in polarity by progressive increase in hyperpolarizing current applied intracellularly. They were accompanied by a prominent and long lasting (up to 100 msec) conductance increase of the mitral cell membrane. 3. Reversed i.p.s.p.s of mitral cells having quite different time courses from the original hyperpolarizing i.p.s.p.s suggest that the inhibitory synapses are widely distributed on the soma and dendrites. 4. E.p.s.p.s could be recorded from g.c.l. cells whose onset latency was approximately 0.6 msec shorter than that of mitral cell i.p.s.p.s. Comparison of the behaviour of e.p.s.p.s in g.c.l. cells and that of mitral cell i.p.s.p. under various conditions of l.o.t. stimulation suggests that these g.c.l. cells are the inhibitory interneurones mediating mitral cell inhibition. 5. The results support the hypothesis of dendrodentritic pathways for activation of granule cells and subsequent inhibition of mitral cells.

1. Electrophysiological study disclosed that chloroform and some other odours generate long-lasting positive potentials in the olfactory epithelium of the bullfrog, while electron microscopical study showed that they elicit vigorous protrusion of the distal cytoplasmic portion of the supporting cell containing the secretory granules.2. The secretory process of the supporting cell is as follows: The first detectable indication is the protrusion of the apical portion of the supporting cell (Pls. 3 and 4); the protruded part detaches from its maternal supporting cell (Pl. 7 B), floats as a droplet in the mucus (Pl. 5), and finally the secretory granules inside the droplet disintegrate into the mucus (Pls. 6 and 7 A).3. The secretion was not elicited by the odours which elicit the negative potentials.4. In Cl(-)-free Ringer solution, neither the positive potential nor the protrusion and secretion occurred.5. When 1-2 mM-Ba(2+) in Ringer solution was dripped on the epithelium, both the positive potential and the protrusion and secretion resulted. Subsequent application of chloroform vapour only advanced further disintegration of the secretory granules, but it elicited neither a new protrusion of the granules nor the positive potential.6. In the olfactory epithelium in which the olfactory cells had degenerated but the supporting cells survived, both the positive potential and the protrusion and secretion occurred, but the negative potential did not.7. It is concluded that Cl(-) entry which mainly generates the longlasting positive potential triggers the secretion of the supporting cell.

To elucidate the organization of synaptic inputs to pyriform cortex neurons, intracellular and extracellular responses of single units were analyzed in urethane-anesthetized rabbits. The lateral olfactory tract (LOT) or the olfactory bulb (OB) was electrically stimulated. Intracellular recordings revealed two types of cells (type I and type II cells), according to the types of EPSP evoked by the LOT or OB shock. The EPSP in the type I cells had shorter latencies (0.0 to 0.9 ms) from the onset of the component 2 (C2) wave of the field potential (which signals the onset of the synaptic depolarization of the apical dendrites of the pyramidal cells in the PC), and that in the type II cells had longer latencies (1.0 to 6.0 ms). A conditioning LOT or OB shock did not suppress the testing EPSP in the type I cells, whereas the conditioning stimulation greatly suppressed the testing EPSP in most of the type II cells. Extracellular recordings from units responding synaptically to the LOT or OB shock revealed a group of units which had short latencies (0.7 to 1.9 ms) of spike discharges. Those units, which were likely to be the same cells as the type I cells, are believed to mediate excitatory synaptic inputs to the type II cells. On the basis of these results, we concluded that type I cells are monosynaptically activated by LOT volleys, whereas type II cells are activated di- or polysynaptically by way of a relay from type I cells. The type I cells were recorded in both the superficial and the deep parts of the pyriform cortex, although they were recorded more frequently in the superficial part. On the other hand, most of the type II cells were recorded in the deep part of the PC. These results support and extend the previous model, in which the monosynaptically activated superficial pyramidal cells give rise to excitatory inputs to other pyramidal cells and neurons in deep layers.

The stimulating effects of mono- and divalent cations and anions were studied in the olfactory epithelia of the bullfrog and the carp. The rhythmic waves induced by these ions were recorded in the olfactory bulb. 1. Many mono- and divalent cations and anions showed stimulating actions in the bullfrog and the carp. 2. Microelectrode studies disclosed that the olfactory receptor cells respond to different ions differently. 3. When many ions were applied with various concentrations, responses appeared with long latencies while the concentration was very low (the "delayed responses"). The responses nearly disappeared at the intermediate concentrations, but then responses with short latencies appeared at the higher concentrations (the "initial responses"). Thus, many ions showed dual responses in the bullfrog, although some exceptional cases were found (choline+, Tl+, La3+, Cd2+). 4. Cd2+ and other heavy metal ions showed depressive actions upon the responses induced by other ions in the olfactory epithelium. 5. Tetrodotoxin of even 10 (-5) g/ml was found ineffective in depressing the rhythmic waves induced by ions. 6. Chemoreceptive activities of the olfactory epithelia of the bullfrog and the carp were compared with the activities of the gustatory receptors. They were also compared with the other chemoreceptors of the fish, namely the palatal organ, external chemoreceptors over the snout region and the lateral-line organ. Chemical senses of the fish were discussed.

Responses of 50 olfactory bulb neurons of the bullfrog to odorous solutions of ethyl n-butyrate were composed mainly of facilitation, inhibition and no response, whereas those of the same neurons to water were mainly of facilitation and delayed facilitation. The responses to water tended to continue longer than those to the odorous solutions.

The water response elicited by application of distilled water on the olfactory epithelium was recorded extracellularly from single olfactory bulb neurones. Characteristics of the water response in the frog olfactory epithelium were examined in comparison with those of the water response in the gustatory and palatal organs. 1. Effects of various electrolyte solutions on the generation of the water response were studied by dripping distilled water on the olfactory epithelium after adaptation to each of these electrolyte solutions. Number of the olfactory bulb cells responding to distilled water increased with increasing the charge of the adapting cations and also with decreasing the size of the cations with a few exceptions. 2. Magnitude of the 'water response' increased with decreasing concentration of salt in the solution which was dripped after adaptation to the isotonic solution of the same salt. 3. The water response was effectively depressed by an electrolyte solution but not by a non-electrolyte solution. An electrolyte also depressed effectively the water response which was produced after adaptation to an organic salt solution. 4. The water response was blocked by treatment of the olfactory epithelium with the uranyl ions which had high affinity for phospholipids. A tentative hypothesis on the generating mechanism of the water response in the frog olfactory epithelium is presented on the basis of the present experimental results and the water responses of the gustatory and palatal organs so far reported.

Extracellular responses were recorded from neurons in the thalamic mediodorsal nucleus (MD) of the lightly anesthetized rabbit. Eighty-seven neurons responded to electrical stimulation of the lateral olfactory tract (LOT shocks). They were located in the medial portion of the MD. In the same portion, negative field potentials with a short latency were evoked by the electrical stimulation of the olfactory projection area in the neocortex (OPA shocks). Fifty-nine MD neurons responded both to LOT and to OPA shocks. Among them, 17 thalamocortical relay neurons (which responded antidromically to OPA shocks) were found to respond transsynaptically to LOT shocks. Of the 87 LOT-responsive MD neurons, 48 responded to the odors applied. Eight odor-sensitive neurons were found to be the thalamocortical relay neurons. Thus, it was proven for the first time that a portion of the olfactory input to the OPA is mediated via relay neurons in the MD. Characteristics of response of MD neurons to odor stimulation were compared with those of OPA neurons. MD neurons did not show a selectivity of response to odors of urine, feces, or dry food pellets, to which OPA neurons responded exclusively. These results were discussed in relation to the functional role of the MD-OPA projection system in the discrimination of specific odors.

(1) Regions which exert centrifugal influences on the olfactory bulb activity were studied by applying systematic stimulation to various areas of the ipsilateral telencephalon in the rabbit. By delivering electric stimuli to the anterior commissure (AC), the deep lying structures in the projection areas of the lateral olfactory tract (LOT) and the medial forebrain bundle situated between the lateral hypothalamic area and the lateral preoptic area, negative field potentials were evoked in the granule cell layer (GCL) of the bulb. (2) Intracellular recordings from the mitral cells and the GCL neurons in the olfactory bulb were performed in order to clarify the modes of the centrifugal influences on the olfactory bulb neurons. (3) EPSPs were recorded in the GCL neurons by stimulation of the deep-lying structure of the prepiriform cortex as well as by stimulation of the AC. The onset time and duration of the EPSPs corresponded well to those of the negative field potentials in the GCL. Thus, it was suggested that these negative potentials were caused by the EPSPs of the number of granule cells. (4) In almost all of the mitral cells, IPSPs were recorded by stimulation of the AC and the deep-lying structures of the LOT projection areas. The onsets of the IPSPs were found with delays of several milliseconds from those of the negative field potentials in the GCL. (5) It was postulated that the excitation of the centrifugal system mainly exerts a depressive influence on the activity of the mitral cell, and that the GCL neuron (presumably the granule cell) seems to be an inhibitory interneuron interpolated between the extrinsic fibers from the telencephalon and the mitral cell.

1. In the rabbit olfactory bulb, analysis has been carried out on intracellular potentials recorded from mitral cells and neurones in the granule cell layer (g.c.l. cells) in addition to the extracellular field potentials in the olfactory bulb elicited by anterior commissure (a.c.) stimulation. 2. Most mitral cell recordings showed i.p.s.p.s with latency of 7-11 msec following a.c. stimulation. These i.p.s.p.s were similar to those evoked by lateral olfactory tract (l.o.t.) stimulation in their sensitivity to internally applied current and showed asymmetrical reversal during application of the hyperpolarizing current. 3. Volleys in the a.c. elicited e.p.s.p.s in type 1 g.c.l. cells whose characteristics were in agreement with those of inhibitory interneurones inferred from the analyses of mitral cell i.p.s.p.s. It has been suggested that these type 1 g.c.l. cells may be the common inhibitory interneurones (presumably granule cells) mediating both a.c.-evoked and l.o.t.-evoked i.p.s.p.s in mitral cells. 4. Conditioning a.c. stimulation depressed the test l.o.t.-evoked i.p.s.p.s in mitral cells and test l.o.t.-evoked e.p.s.p.s in type 1 g.c.l. cells. These observations are in good agreement with the hypothesis that l.o.t.-evoked i.p.s.p.s are mainly mediated by the dendrodendritic reciprocal synapses between mitral cell dendrites and peripheral processes of granule cells. 5. The results are discussed in relation to the inhibitory mechanisms controlling mitral cell activity in the olfactory bulb.

In the negative EOG-generating process a cation which can substitute for Na(+) was sought among the monovalent ions, Li(+), Rb(+), Cs(+), NH(4) (+), and TEA(+), the divalent ions, Mg(++), Ca(++), Sr(++), Ba(++), Zn(++), Cd(++), Mn(++), Co(++), and Ni(++), and the trivalent ions, Al(+++) and Fe(+++). In Ringer solutions in which Na(+) was replaced by one of these cations the negative EOG's decreased in amplitude and could not maintain the original amplitudes. In K(+)-Ringer solution in which Na(+) was replaced by K(+), the negative EOG's reversed their polarity. Recovery of these reversed potentials was examined in modified Ringer solutions in which Na(+) was replaced by one of the above cations. Complete recovery was found only in the normal Ringer solution. Thus, it was clarified that Na(+) plays an irreplaceable role in the generation of the negative EOG's. The sieve hypothesis which was valid for the positive EOG-generating membrane or IPSP was not found applicable in any form to the negative EOG-generating membrane. The reversal of the negative EOG's found in K(+)- , Rb(+)- , and Ba(++)-Ringer solutions was attributed to the exit of the internal K(+). It is, however, not known whether or not Cl(-) permeability increases in these Na(+)-free solutions and contributes to the generation of the reversed EOG's.

The ionic mechanism of the electropositive olfactory receptor potential was studied in the bullfrog and the swamp frog. The positive receptor potential strikingly decreased in amplitude in chloride-free solution. When the olfactory epithelium was immersed in high-KCl-Ringer's solution and then in Cl-free, high-K solution, the polarity of the positive potential could be reversed. This is supposed to be due to the exit of the increased internal chloride ion. From the above two experiments it is concluded that the positive olfactory receptor potential depends primarily upon the influx of the chloride ion through the olfactory receptive membrane. Some contribution by potassium and possibly other ions may occur. The ability of other anions to substitute for chloride was examined. It was found that only Br-, F-, and HCO2- could penetrate the olfactory receptive membrane. The sieve hypothesis in the inhibitory post-synaptic membrane (Coombs, Eccles, and Fatt, 1955) is not applicable to the olfactory receptive membrane on the basis of the size of hydrated ions, but it may be applicable on the basis of the sizes of naked ions.

In order to clarify whether or not the electronegative olfactory mucosal potentials (EOG) are generator potentials, the effects of changed ionic enviroment were studied. The EOG decreased in amplitude and in some cases nearly or completely disappeared, when Na(+) in the bathing Ringer solution was replaced by sucrose, Li(+), choline(+), tetraethylammonium(+) (TEA), or hydrazine. In the K(+)-free Ringer solution, the negative EOG's initially increased and then decreased in amplitude. In Ringer's solution with increased K(+), the negative EOG's increased in amplitude. When K(+) was increased in exchange for Na(+) in Ringer's solution, the negative EOG's decreased, disappeared, and then reversed their polarity (Fig. 6). Next, when the K(+) was replaced by equimolar sucrose, Li(+), choline(+), TEA(+), hydrazine, or Na(+), the reversed potentials recovered completely only in Na(+)-Ringer's solution, but never in the other solutions. Thus, the essential role of Na(+) and K(+) in the negative EOG's was demonstrated. Ba(++) was found to depress selectively the electropositive EOG, but it hardly decreased and never increased the negative EOG. Hence, it is concluded that Ba(++) interferes only with Cl(-) influx, and that the negative EOG's are elicited by an increase in permeability of the olfactory receptive membrane to Na(+) and K(+), but not to Cl(-). From the ionic mechanism it is inferred that the negative EOG's are in most cases composites of generator and positive potentials.

A correlation between the length of the olfactory cilia and the electrical activity of the olfactory epithelium was studied in newts living in water and on land. The olfactory cilia grew when newts were transferred onto land. The cilia in the olfactory bud became longest in 108 hours after the transfer and then became shorter, while those in the interstitium only gradually elongated. Slow potentials were evoked in the epithelium by the application of odorous fluids but not by odorous vapors for 20 hours after the transfer. Thereafter, the slow potential began to appear in response to odorous vapors and reached maximal magnitude between 60 and 70 hours after the transfer, while it was not evoked by odorous fluids in this period. In the later stage, the slow potential to odorous vapors decreased in magnitude and disappeared 120 hours after the transfer, while it began to reappear in response to odorous fluids. When these changes in the slow potential were compared with those in the cilium, a discrepancy was found between the period of maximal potential magnitude and that of maximal cilium length.

Electrical activity and histological changes were studied in the degenerating olfactory epithelium of the bullfrog after the olfactory nerve had been sectioned. After nerve section, the electrical responses to odors disappeared in the olfactory epithelium in 8 days in the summer, in 11 days in the early autumn, and in 16 days in the early winter. In the degenerating olfactory epithelium a striking decrease in the number of olfactory cells was found, but not of supporting cells. The ratio of the number of olfactory cells to that of supporting cells was found to decrease from 5 or 6 to below 2 after the nerve section. At a ratio below 2, the electrical responses to odor disappeared. The histological changes in the bullfrog are compared with those in the mouse and rabbit. The localization of the olfactory pigment and the electrical activity of the supporting cell are discussed. It was concluded that all three types of responses to odors originate from the activity of the olfactory cell.

The effect of electrotonus on the slow potential of the olfactory epithelium of the frog was studied. The "on"-slow potential induced by a general odor like amyl acetate increased its magnitude in accordance with increase of anodal current, while it decreased its magnitude with increase of cathodal current. Similar relations were also found in the case of the vapors of organic solvents like ethyl ether of low concentrations. Conversely, the on-slow potential induced by the vapors of organic solvents of high concentration decreased its magnitude in accordance with the increase of anodal current, while it increased its magnitude with the increase of cathodal current. The "off"-slow potential induced by the vapors of organic solvents of high concentration showed a potential change under the action of electrotonic currents which is similar to the change of the on-slow potential induced by general odors. It was concluded that there are two receptive processes in the olfactory cell. One is an ordinary excitatory process which produces an electronegative slow potential in response to general odors. The other is a process of a different kind which is activated only by the vapor of an organic solvent of high concentration and which shows an entirely opposite reaction from that generally found in excitable tissues when an electrotonic current is applied.