Eye movements evoked by electrical stimualtion of the brain in anesthetized fishes.

Brain Behav. Evol. 11: 109-129 (1975)

Eye Movements Evoked by Electrical Stimulation of the Brain in Anesthetized Fishes L. S. D e m s k i and D. H. B auer Departments of Anatomy and Biology, University of New Mexico School of Medicine, Albuquerque, N. Mex.

Key Words. Eye movements • Brain stimulation • Vestibular nuclei • Cerebellum • Sunfish • Goldfish • Oculomotor complex • Trochlear nerve • Medial longitudinal fasciculus Abstract. Several eye movements were evoked by electrical stimulation of the brain in anesthetized sunfish and goldfish. Conjugate lateral rolling movements, similar to eye movements observed when an unoperated fish is rotated about its long axis, were evoked from the acoustico-lateral area of the medulla and the eminentia granulans and an adjacent medial portion of the cerebellum. Bilateral and unilateral backward rotations, similar to the eye movements observed when unoper ated fish are rotated forward about the interpupillary axis, were evoked from the medial longitudinal fasciculus and areas related to the oculomotor nerve. Bilateral forward rotations, comparable to the eye movements resulting when unoperated fish are rotated backward about the interpupillary axis, were elicited by stimulation near the trochlear nerve roots in the valvula of the cerebellum; unilateral responses re sulted from stimulation near the exiting trochlear nerves. Convergence was elicited by stimulation in the midline near the oculomotor complex and the medial longitu dinal fasciculus while unilateral vcrgcnce responses were triggered by stimulation in the medial longitudinal fasciculus and areas lateral to the oculomotor nucleus. Con jugate eye movements in the horizontal plane were frequently evoked but were not studied in detail.

Studies on fishes have contributed greatly to the understanding of peri pheral vestibular mechanisms [L o w e n s t e in , 1971]. More recently these

Downloaded by: Univ. of California Santa Barbara 128.111.121.42 - 3/24/2018 1:39:22 PM

Introduction

110

D emski/B auer

animals have been used in studies on the effects of various environmental influences on posture and orientation in three-dimensional space [von B aumgarten el al., 1972]. Central vestibular mechanisms in fishes, how ever, have not been as well studied. For this reason, we have begun to study them in sunfish and goldfish, anticipating that our results will con tribute to a better understanding of how vestibular systems affect behav ior. Eye movements arc frequently used as a measure of vestibular activ ity. Consequently, we have chosen to identify the regions of the fish brain involved in their control and then determine if these areas also mediate more complete postural and orientational responses. The technique of electrical stimulation of the brain in anesthetized animals has been used as the first step in this analysis. This paper reports the results of studies using this method.

A total of 29 green sunfish (Lepomis cya/iellus), ranging in standard length from 11 to 22.5 cm and 5 goldfish (Carassitts entrains), ranging in standard length from 12 to 13.5 cm, were used in this study. Sunfish were obtained from a local pond and goldfish were purchased from a commercial supplier. Animals were stimulated while anesthetized by a 0.3-pcrccnt urethane solution that was circulated through a hollow mouth holder and passed over their gills. Fish were held by a stainless steel apparatus previously described [D emski and K nigce , 1971; D emski and G erald, 1972]. Monopolar electrodes were made from tapered 00 stainless steel insect pins insulated with Epoxylite except al the lip. The stimula tion was provided by a Nuclear Chicago 7150 constant current stimulator and con sisted of 50-Hz, 2-mscc square wave pulse pairs of opposite polarity with currents equal to or less than 50 it A. The indifferent electrode was a bare wire placed in the solution bathing the animal. The stimulus current was measured on an oscilloscope as the voltage drop across a resistor in series with the fish. Most electrode tracks (160) were run in a dorsoventral direction; however, some (33) were run in an ob lique lateromcdial direction. Using a micromanipulator, electrodes were lowered slowly into the brain with the stimulation set at 50 ,«A. When certain eye move ments (see Results for details) were observed, the threshold at that site was deter mined. The electrode was then lowered in 0.1-mm increments and the threshold for eye movements was recorded at each point. After passing the lowest threshold site by at least 0.2 mm, the electrode was returned to this site and the area was marked with iron ions by passing a 20-/< A anodal current through the electrode for 10-20 sec. In some cases, the top and bottom of an electrode track were also marked in order to permit identification and plotting of various segments of the track (dashed lines in fig. 2, 4 and 7). Some responses were elicited at the lowest setting of the


Methods

Eye Movements Evoked from Fish Brain

111

stimulator (10 /uA), and when more than 1 10-,aA response was consecutively evoked, the midpoint of these sites along the track was marked at the lowest thresh old region. In a few instances (points indicated by arrows in fig. 4), the anatomical sites from which responses were evoked were obtained by reconstruction of an en tire electrode track rather than by directly marking them with Prussian blue. Points of interest were plotted in relation to known marked sites using a shrinkage factor calculated by measuring the distance between at least two Prussian-blue-marked sites along the track. Many negative tracks (current at 50 «A) were run, a few of which were marked for identification by deposition of iron along the entire track. All evoked eye move ments were recorded by the observer as written descriptions and diagrams. In some cases, responses were also recorded on videotape or 16-mm movie film. Following testing, animals were decapitated and their brains fixed in 10-percent formalin, embedded in paraffin, serially sectioned in the frontal plane at 10 ¡xm and stained, using Prussian blue [A kert and W elker , 1961] for localization of iron ions deposited at the stimulation sites and neutral red [H umason, 1967] for identification of nuclear groups. Marked stimulation points are plotted on representative frontal sections of the brains of green sunfish (fig. 2, 4 and 7) and goldfish (fig. 3). Anatom ical terminology was adopted from A riens K appers el at. [1936] and D emski and Knigge [1971],

Results

Lateral Rolling This response is a conjugate movement of the eyes similar to that ob served when an unoperated fish is rotated about its long axis (tilted to the side). In this case, the pupil of the eye on the side to which the animal is tilted is directed dorsally, while the pupil of the other eye points ventrally


During the course of this study, many different eye movements were evoked. We have chosen to concentrate on three of the responses (lateral rolling and forward and backward rotation) which in our opinion are like ly to be more affected by otolith rather than visual or other sensory input. Other responses were studied (horizontal and vergence movements) but not to the extent of those listed above. All evoked eye movements were highly repeatable with short latencies (usually less than 1 sec) and ended abruptly with the termination of the stimulation. Most of the experiments were carried out in sunfish because of our familiarity with evoked respon ses in this group of fishes; however, at least some responses were also studied in goldfish since these animals are probably the most readily available teleosts and have been used by many other investigators.

D emski/B auer

112

Fig-. 1. Dorsal view of the head of a sunfish before (a) and during (b) stimulation of the left side of the brain. A typical evoked ipsilateral lateral rolling of the eyes is indicated by the arrows. The left pupil moves up while the right one moves down (see text for details).

Table I. Stimulation sites and parameters used to evoke conjugate lateral rolling movements of the eyes in anesthetized goldfish Point No. and region stimulated

Fish No.

Threshold, (l‘A)

Additional eye movements associated with the stimulation

Cerebellum 1 2

5 2

32 25

3

2

38

none slight horizontal conjugate movement and possible bilateral forward rotation none

4 6 2 3 6 5

11 10 15 10 10 12

possible bilateral forward rotation none none none none none

4 5 6 7 8 9

[Lowensteln, 1936, fig. 4]. Compensatory body and fin movements to lateral tilting have also been described [Lowenstein, 1936, fig. 6]. Electrically evoked eye movements appear identical to those observed in the intact fish tilted to the side (fig. 1) and are maintained continuously during stimulation. At threshold stimulation, the eyes may move only


Medulla

113

slightly, but suprathreshold stimulation usually results in a maximal ex cursion of each eye. For most evoked responses, the eye on the stimulated side was directed upward and the contralateral pupil pointed downward (fig. 1); however, some opposite responses, e.g. the pupil on the stimulated side was directed downward and the other pupil upward, were evoked from a portion of the cerebellum (see details below). The more typical eye-up on the stimulated side will be referred to as an ipsilateral response, while the eye-down on the stimulated side will be called a contralateral response. Other responses such as horizontal conjugate movements could also be associated with evoked lateral rolling (fig. 2; table I). In these cases, the resulting deviation of the eyes was complex; however, the later al rolling component was clearly discernible. Lateral rolling of only one eye was not observed. Fin bending, as has been described as a response to tilting in normal fish [L o w e n s t e in , 1936], was evoked from the medulla with evoked ipsilateral rolling of the eyes. In this case the posterior soft part of the dorsal fin and upper part of the caudal fin were bent toward the stimulated side while the posterior part of the anal fin and lower por tion of the caudal fin were bent toward the opposite side. The anatomical distribution of sites from which lateral rolling was evoked in sunfish is illustrated in figure 2. Many of the positive sites in the medulla formed a continuous distribution that extends caudally from the level of the entrance of the lateral line and eighth cranial nerves (sec tion I, fig. 2) to almost the level of the obex (sections J-L, fig. 2). The points appear to be grouped at the entrance of the nerves (section I, fig. 2) as well as in the central portion of the acoustico-lateral area (sec tions J-L, fig. 2) as defined by P ea rso n [1936]. Several more ventrally situated points are shown in figure 2, sections J and L. A similar distribu tion of positive sites was found in the goldfish medulla (points 4-9, fig. 3) with points at the level of the eighth nerve and in the central portion of the acoustico-lateral area. All responses evoked from the medulla were ipsilateral. Lateral rolling was also evoked from portions of the body of the cere bellum in both sunfish (sections G and H, fig. 2) and goldfish (points 1-3, fig. 3). Positive sites along dorsoventrally directed tracks were located at the border of the molecular and granular layers just medial and rostromedial to the eminentia granularis (sections G and H, fig. 2, 3), as well as in the eminentia itself (section H, fig. 2). In the case of stimulation in the more medial area, both ipsilateral (solid circles in fig. 2 and 3) and contralateral (open circles in fig. 2) responses were observed. In fact, the



D emski/B auer

Fig. 2. Summary of eye movements evoked from points along dorsoventrally directed electrode tracks in anesthetized sunfish. Points indicated by the various symbols were histologically identified using the Prussian-blue technique and are plotted on representative frontal sections. Each site represents the lowest threshold area along an electrode track for eliciting either a single specific response or two simultaneous responses (overlapping symbols). Dashed lines (section F) indicate electrode tracks reconstructed to show several stimulation sites. Arrows (section F) indicate the region along the adjacent electrode track from which bilateral forward rotation was evoked; symmetrical responses were elicited only from the region indicated by the solid triangle while the ipsilateral eye showed greater movement than the contralateral eye during stimulation of the area near the upper arrow and the opposite situation resulted from stimulation of the region indicated by the lower arrow (sec text lor details). A-LL=Aeouslico-lateral lemniscus; EG=eminentia granulans of cerebellum; GI=ganglion islhmi; GL=granulc cell layer of cerebellum; IL=inferior lobe of hypothalamus; ML “ molecular layer of cerebellum; MLF=modial longitudinal fasciculus; MV=ventricle of the midbrain; N 3“ nucleus of oculomotor nerve; NPR = nucleus prerotundus; NPRM “ nucleus prerotundus pars medialis; NR “ nucleus rotundus (also called nucleus glomcrulosus); N R3 = root of oculomotor nerve; O L=optic lobe; SV = saccus vasculosus; TEG “ tegmentum of the midbrain; T L =torus longitudinalis; TS = torus semicireularis; V4= fourth ventricle; VAL=valvuln of the cerebellum. « “ Lateral rolling, eye up on stimulated side; O“ lateral rolling, eye down on stimulated side; A = bilateral forward rotation; T = bilateral backward rotation; v = unilateral backward rotation; —“ horizontal conjugate movements; / “ convergence;\ “ unilateral vergencc movement.


1 14

Eye Movements Evoked front Fish Brain

115

two sites on section H of figure 2 designated by partially overlapping open and solid circles, represent points along electrode tracks in which the lateral rolling response reversed direction by simply lowering the elec trode a few tenths of a millimeter. In one case, stimulation of this area at 10 (jlA evoked a contralateral response which, following passage of the direct current for the Prussian blue marking (20 ¡.iA for 20 sec), changed to an ipsilateral response with a threshold of 100 uA. It can be suggested that the area from which the original response was evoked was lesioned by the direct current and that the subsequent higher stimulus current spread to and stimulated adjacent areas which mediate the response in the opposite direction. In order to better define the cerebellar areas involved in lateral rolling responses, angled electrode tracks were run through the eminentia granularis and adjacent granular layer in sunfish (fig. 4). As was the case for the dorsoventrally directed tracks, responses were evoked from the cere bellar area medial to the eminentia as well as from the eminentia itself. As before (see above), lateral rolling in both directions was evoked from the medial cerebellar area but not from the eminentia. In the case of one electrode track which was reconstructed to allow the plotting of un marked positive stimulation sites (see Methods), stimulation at 22 uA in


Fig. 3. Summary of histologically identified sites from which lateral rolling responses were evoked in anesthetized goldfish. Points are plotted on representative frontal sections of the brain. Numbers on the stimulation sites refer to table I which gives additional de tails for each point. For explanation of abbreviations, see figure 2.

116

D emski/B auer

the ventral granular region evoked a contralateral response (double ar rows in fig. 4), while stimulation throughout the adjacent eminentia re sulted in ipsilateral responses at currents of 10-15 nA (region between single arrows in fig. 4). These results correlate with those obtained from the dorsoventrally directed tracks and suggest that both cerebellar areas are involved in the control of lateral rolling movements. In the course of running angled tracks through the eminentia granularis it appeared that the strongest responses were obtained when tracks were run at more ros tral levels. In fact, we were able to run negative tracks (current at 50 ,uA) through the eminentia at its caudal and middle levels. Figure 5 shows the distribution of thresholds for evoking lateral rolling responses from the Prussian-blue-marked sites in the sunfish medulla and cerebellum illustrated in figures 2 and 4. A significant difference (0.005level using a two-tailed t-test) between the mean thresholds for evoking the responses from the cerebellum and medulla was found. Higher thresholds for cerebellar sites were also found to occur in goldfish (table I). In fact,


Fig. 4. Summary of the anatomical distribution of sites from which lateral rolling re sponses were evoked along oblique electrode tracks (dashed lines) in anesthetized sunfish. Points are plotted on representative frontal sections through the eminentia granularis of the cerebellum, o = Prussian-blue-marked sites from which ipsilateral responses were evoked; O = marked sites from which contralateral responses were elicited. Low threshold (10-15 /tA) ipsilateral rolling responses were evoked from the area between the single ar rows along the electrode track run through the eminentia granularis while contralateral responses at 22 /¿A were elicited at the point indicated by the double arrow. All regions indicated by arrows were determined by reconstruction of the entire electrode track (see Methods). For explanation of abbreviations, see figure 2.


117

Fig. 5. Relative thresholds for evoking lateral rolling responses from the medulla and cerebellum in anesthetized sunfish. n = Number of points;T-bar=standarderror; * = lcvel of significance (0.005) of the difference in mean thresholds for evoking responses from medulla compared to cerebellum computed using a two-tailed t-test.

Forward Rotation This response is characterized by a rotation of the eyes about the interpupillary axis with the dorsal surface of the eyeball moving in a rostral direction (fig. 6A). It appears to be the same movement that results from tipping (pitching) unoperated fishes in a backward or head-over-tail direc tion [L owenstein , 1936, fig. 3; T raill and M ark, 1970; D emski and Bauer, unpublished]. In sunfish, positive bilateral responses were evoked from sites in the valvula of the cerebellum (sections E^G, fig. 2). All of these points are within or adjacent to the crossing trochlear nerves [see anatomical de scription in A riens K appers et al., 1936; P earson , 1936]. Forward rota tion was not evoked from the area of the trochlear nucleus. This area was, however, probably not adequately tested, although in one case other responses (convergence and backward rotation) were elicited as an elec-


for fish No. 2 and 5, positive responses evoked from the cerebellum were followed by lower threshold responses evoked from medullary sites (table I).

I 18

D p.mski/B auer

I'ig. 6. Lateral view of the head of a stmfish, illustrating forward (A) and backward (B) rotation of the eyes.

Backward Rotation This response is the same as forward rotation with the exception that the dorsal surface of the eyeball moves in the caudal direction. Backward rotation appears to be the same as the response of the eyes observed


node which evoked forward rotation in the valvula was lowered into the underlying tegmentum at approximately the level of the trochlear nucleus (see midline track in section F, fig. 2). In some eases, non-symmetrical rotation of the eyes was evoked. The lateral track in section F of figure 2 illustrates an instance in which stimulation at the upper arrow produced greater rotation in the ipsilateral eye, while greater rotation in the contra lateral eye resulted from stimulation at the lower arrow. Activation of the area between the arrows (triangle) resulted in a bilaterally symmetrical ro tation. Similar results were obtained using angled electrodes. The triangle along the angled track through the valvula (fig. 7) represents a site from which bilaterally symmetrical forward rotation was evoked. Non-symmelrical rotation of the eyes occurred during stimulation of points above (to the left on the figure) or below (to the right on the figure) this marked site. There was greater movement of the ipsilateral eye at the higher points while the opposite occurred at the lower sites. Unilateral forward rotation of the ipsilateral eye was evoked in several cases in which the lat eral edge of the brain at the junction of optic lobes and cerebellum (isth mus) was stimulated in goldfish (not illustrated). This area corresponds to the region from which the fourth nerve exits.


119

Fig. 7. Lowest threshold points from which forward (▲) and backward (Y) rotations of the eyes and convergence (/) were evoked along oblique electrode tracks in anesthetized sunfish. Points arc plotted on representative frontal sections of the sunfish brain. For ex planation of abbreviations, see figure 2.

Bilateral backward rotation of the eyes in sunfish occurred during stimulation of the midline tegmentum in areas within or adjacent to the oculomotor nuclear complex (sections C, D, F, fig. 2), as well as from a point partially within the medial longitudinal fasciculus (MLF) in the medulla (section I, fig. 2). Convergence movements were also frequently evoked with backward rotations, thus resulting in a complex movement with two clearly discernible components (sections C, D, fig. 2). Unilateral backward rotation of the ipsilateral eye in sunfish resulted from stimula tion of sites lateral to the oculomotor nuclei, points in the vicinity of the exiting oculomotor roots as well as a single site in the MLF (sections D, E, G, fig. 2). Unilateral vergence of the same eye occurred with many of these unilateral backward rotations (sections D, E, G, fig. 2). Several an gled tracks were run through the oculomotor nuclei in sunfish (fig. 7). Bi lateral backward rotation was evoked only at points in the midline. Sites lateral to these midline points resulted in unilateral responses (not shown in figure). Convergence movements were also associated with stimulation of the two midline sites illustrated in figure 7.


when unoperated fishes are tipped forward [T raill and M ark, 1970; D emski and B auer , unpublished].

120

D emski/B auer

Horizontal Movements Conjugate movements of the eyes in the horizontal plane to either side were frequently observed during these experiments. However, these movements were not systematically studied since they have been pre viously associated, at least in part, with visuomotor areas such as the op tic lobe [A kert , 1949; H ermann , 1971a; D emski and B auer , unpub lished]. As stated above, these responses frequently occurred with lateral rolling when the medulla and cerebellum were stimulated (sections G-L, fig. 2). A few horizontal movements independent of other responses were observed when sites in the MLF were stimulated (sections I and L, fig. 2). It must be stressed that horizontal movements were evoked from many stimulation sites which were not marked for identification. Vergence Movements These responses, consisting of a medial deviation of the rostral portion of the eyeball, were unilateral as well as bilateral (convergence). Converg ence in sunfish resulted from stimulation in the midline within or adjacent to the oculomotor nucleus (sections C, D, F, fig. 2, 7). Unilateral vergence movements of the ipsilateral eye were elicited by stimulation of sites in the tegmentum lateral to the oculomotor nucleus, points near the ocu lomotor roots (sections D, E, fig. 2) as well as from a point in the MLF (section G, fig. 2). As mentioned above, the points from which vergence responses were evoked had the same general anatomical distribution as points from which backward rotation was elicited and in many cases both responses were evoked simultaneously from the same point.

Several different eye movements have been evoked in this study, and as pointed out, emphasis was placed on those movements thought to be most reflective of otolith activity. It should also be kept in mind that var ious responses observed in this investigation, although considered sepa rately to facilitate their study, can occur together as complex movements both in response to electrical stimulation of the brain as well as to natural environmental influences. Our failure to evoke nystagmus, which has been elicited in teleosts by vestibular nerve stimulation [K orn and B ennett , 1971] may be due to the anesthesia blocking these responses. We feel that mapping studies similar to ours only using unanesthetized preparations [see details in


Discussion


121

Lateral Rolling This evoked conjugate response which relates to ocular compensation to a static tilt to the side [L owenstein, 1936, fig. 4] may be mediated by the inferior and superior rectus muscles. Its counterpart in monkeys could be a tonic movement that has been evoked (along with other movements) by utricular nerve stimulation. During stimulation, the pupil of the eye on the stimulated side was directed upward and the other pupil moved down ward. In these experiments it was demonstrated that the left superior and the right inferior rectus muscles were active during the movement [Suzuki et al., 1969], In our studies we also observed fin movements which are used to compensate for lateral tilting in the intact fish [L owenstein , 1936], thus suggesting that at least in some cases the stimulation is trig gering neural mechanisms which control both ocular and postural systems related to equilibrium. This idea has recently been given further support by experiments in free-swimming fish with chronically implanted stimu lating electrodes [D emski and Bauer, 1974, unpublished]. Stimulation of the medullary and cerebellar regions from which lateral rolling of eyes was evoked in the anesthetized sunfish during implantation resulted in both lateral rolling movements of the eyes and tilting to the side (at least up to 90°) when the fish were tested while swimming free in a 50 gal aquarium [see D emski and K nigge , 1971; D emski and G erald, 1974, for details of this procedure]. Most of the lateral rolling responses were evoked from the medulla in regions which are likely to correspond to vestibular nuclei, including the tangential nucleus, as have been described for various teleosts [Ramón y Cajal, 1908; A riens K appers et al., 1936; P earson, 1936; L arsell, 1967; M aler et al., 1973; H inojosa, 1973]. It is significant that the only study listed above [M aler et al., 1973], based on modern methods of fiber tracing, was done on Gnathonemus petersii, a fish with a very specialized medulla that at least superficially appears quite unlike that of the species in this study. Anatomical studies of the distribution of vestibular nerve fi bers in sunfish and goldfish using modem reliable methods [see N auta and E bbesson, 1970, for a discussion of techniques] must be done before one can state with complete assurance that positive medullary sites from this study correspond to the vestibular nuclei. If the comparison of the rolling movements in fishes to similar move


K orn and B e n n e t t , 1971] are needed to investigate the anatomical sites from which nystagmus can be evoked in fishes.

Demski/B auer

ments evoked by utricular nerve stimulation in monkeys is valid (see above), this would suggest that the areas stimulated in the fish medulla also receive ipsilateral utricular afferents. Based on our stimulation re sults, this presumed vestibular nuclear area probably has connections with the cerebellum (the other major region from which lateral rolling was evoked), extraocular nuclei and the spinal cord (for control of tilting and fin bending). Direct support for vestibular connections to extraocular nuclei in fish comes from neurophysiological studies in which the activity of teleost oculomotor neurons was influenced by vestibular nerve stimula tion [K id o k o r o , 1969; K o r n and B e n n e t t , 1971]. Connections from vestibular nuclei to spinal cord have been clearly demonstrated in am phibians, as has their extreme importance for the maintenance of normal posture [B arale et cil., 1971; C orvaja et al., 1973]. From these results, it can be suggested that similar mechanisms may be present in teleosts. Connections between the vestibular nuclei and portions of the cerebellum are well known in mammals; in fact, it may be significant that the vestibu lar areas involved also appear to be the primary region of termination of primary afferents from the otoliths [see review by C o h e n , 1971], A dis tinct vestibular input into the cerebellum has also been demonstrated in amphibians [P r e c h t and L u n a s , 1969; M e h l e r , 1972] and our data would suggest a similar situation in teleosts. Lateral rolling has also been evoked by electrical stimulation of a por tion of the body of the cerebellum, namely, the eminentia granularis and the adjacent area just medial to it. Responses from the eminentia were always ipsilateral, and in this respect were similar to those evoked from the medulla. Consistent with this finding are reports that, like the vestibu lar nuclei, the eminentia in teleosts also receive primary vestibular affer ent fibers [L a r sell , 1967] and that a possibly homologous region in am phibians clearly receives vestibular inputs [P r e c h t and L l in a s , 1969]. The other region from which responses were evoked is mainly in the granular layer but it also contains several large fiber tracts, including the cerebellar peduncle and a commissure that appears to interconnect the eminentia of either side. In addition, just below this area are cells which according to P e a r so n ’s [1936] description in trout could constitute a deep cerebellar nucleus. Thus, it is not possible from our data to state exactly which of these structures are responsible for the evoked eye movements. Activa tion of the cerebellar peduncles seems an unlikely explanation since they would also probably have been stimulated in other portions of the cere bellum which were always negative. One of the distinctive features of


122

123

responses in this area was that they could be either ipsilateral or contra lateral and, in some cases, reversed with only slight movement of the electrode. Activation of the commissural fibers seems more consistent with this observation. Hypothetically, stimulation of fibers from one eminentia would elicit a response to one side, and moving the electrode into adjacent fibers from the other eminentia could result in the opposite response. Our finding of cerebellar activation of eye movements is consistent with observations that in at least some teleosts Purkinje cells are known to end directly on oculomotor neurons [K id o k o r o , 1969], This finding also suggests that a deep cerebellar nucleus may not necessarily be in volved in our evoked movements. Lesion studies have also demonstrated cerebellar projections to extraocular nuclei in both goldfish [B r a f o r d , 1970], and nurse sharks [E bbesson and C a m pbe ll , 1973], The finding that, in general, greater stimulation was necessary to evoke lateral rolling responses from the cerebellum than from medullary regions cannot be ad equately explained until further anatomical and electrophysiological stud ies are carried out. The cerebellum has been stimulated electrically in free-swimming te leosts. The results of these studies show that circling in the horizontal plane is the primary response evoked from the cerebellar cortex in several species [C lark et al., 1960] while tilting to the side has been observed during stimulation of the area adjacent to the eminentia in sunfish [D e m ski and B a u e r , unpublished]. Thus, the areas of the cerebellum related to eye movements are probably also involved in more complete postural res ponses as well. Comparison of our results in teleosts with those of similar studies in mammals is made especially difficult because the eyes in fishes are lateral, while those in mammals have a frontal position. This makes comparison of certain eye movements in the two groups extremely difficult and for this reason we will not attempt to do this. However, it is instructive to compare the general areas of the fish brain from which lateral rolling (the most thoroughly studied response) was evoked with areas from which conjugate eye movements have been similarly elicited in mammalian species. In general, our findings that eye movements can be evoked from presumed vestibular areas of the medulla and a restricted portion of the cerebellum compare favorably to the observations that various eye move ments can be evoked by stimulation in the vestibular nuclei in monkeys [T o k u m a su et al., 1969] and from portions of the cerebellum in both monkeys [R on and R o b in so n , 1973] and cats [C oh en et al., 1965],



Demski/IU uf.r

Forward and Backward Rotation These evoked responses consist of ocular rotations about the interpupillary axis. Similar eye movements have been described in unoperated lelcosls as a compensatory response to being tilted in a head-over-tail or tail-over-hcad manner [L o w e n s t e in , 1936, fig. 3; T raill and M a r k , 1970; D em ski and B a u e r , unpublished]. In some of these experiments eye deviations of over 30° were observed and curves were generated for eye deviations through a full 360° till of the fish. It was observed that blinded fish showed more counter-rolling than normals, and that most of the response was lost following labyrinlhcctomy. Thus, the authors con cluded that vision and other cues arc involved, but most of the response is due to the vestibular input [T ra ill , and M a r k , 1970]. Consistent with this interpretation arc reports that blinded goldfish subjected to changing G-forccs and weightlessness perform diving, climbing or constant looping responses depending on the specific conditions imposed on them. The sig nificant factor in these studies is that in all cases the response is one of tilting (head over tail or tail over head) and that this is probably mediated primarily through the otolith organs [ von B aum g a rten et al., 1969, 1971, 1972]. From the above, it can be suggested that the rotational eye movements observed in this study on anesthetized fishes may be involved in postural and orientational responses to gravitational and centrifugal forces via input from the otoliths. Further support for this idea comes from experiments in which electrodes were implanted chronically in re gions which resulted in eye rotation in anesthetized sunfish |D em sk i and B a u e r , unpublished]. During stimulation while the fish were swimming free, the animals made diving or climbing movements which followed the evoked eye movements, e.g. the animal made backward hcad-over-tail movements accompanied by backward rotation of the eyes or made for ward tail-over-hcad movements accompanied by forward rotation of the eyes. In one case, a fish showed continuous hcad-over-tail backward looping during the stimulation. The direction of the rotational responses was dependent on the area stimulated (see discussion below). Based on dissected specimens of sunfish, forward and backward rota tional movements are probably mediated primarily by contractions of the superior and inferior oblique muscles respectively [P it t m a n , and D e m s k i , unpublished]. It is not surprising, therefore, that we evoked unilateral for ward rotation from the area where the trochlear nerve exits and bilateral forward rotation from the areas of the valvula and tegmentum in which the trochlear roots cross. As predicted, backward rotations were evoked


124


125

Horizontal Conjugate Movements These eye movements have been associated with turning movements in free-swimming fishes [H erm a n n and C o n st a n t in e , 1971], and sponta neous [E aster , 1971; H e r m a n n and C o n st a n t in e , 1971], optokinetic and pursuit movements [E a ster , 1972] in restrained fishes. They also oc cur as components of nystagmus induced by vestibular nerve stimulation [K orn and B e n n ett , 1971]. Tonic horizontal conjugate movements were evoked in this study by stimulation of areas of the medulla, presumed to be vestibular nuclei (see discussion under Lateral Rolling) as well as from the MLF and eminentia granularis and adjacent portions of the cerebellum (medial region). As mentioned in Results, these responses were not studied in detail. We also evoked responses from the optic tectum (points not identified histological ly), that appear to be similar to those reported by A k er t [1949] for stim ulation of the trout optic lobes. Thus, it appears that several areas related to either the vestibular or visual systems may be primary substrates from which horizontal movements can be evoked. The restricted distribution of positive sites in the cerebellum in our study on anesthetized fishes proba bly does not accurately reflect the role of this structure in the control of horizontal eye movements, since C lark et al. [1960] observed turning with associated eye movements in response to stimulation of many areas of the cerebellar surface in sunfish as well as in several other free-swim ming teleosts. Electrical recording studies in several teleosts have also indicated that tectal [J o h n sto n e and M a rk , 1969; H e r m a n n , 1971a], cerebellar [H e r m ann , 1971a] and vestibulo-oculomotor systems [H e r m a n n , 1971b; K orn and B e n n e t t , 1971] are all involved in the mediation of these


from the midline in the oculomotor nucleus and the MLF, presumably by either stimulation of both sides simultaneously or possibly through the in trinsic connections of the oculomotor system [see discussions in K orn and B e n n ett , 1971, 1972; W axman and P a ppa s , 1971]. Unilateral res ponses were triggered by stimulation either lateral to the oculomotor nu cleus or in the vicinity of the nerve itself. One unilateral response resulted from stimulation through a laterally situated electrode that was in the MLF on one side only. In regard to general comparisons, it can be stated that various eye movements have been evoked from areas in the monkey brain similar and possibly homologous to the regions from which we evoked eye rotations in teleosts [B e n d e r and S h a n z e r , 1964],

126

D emski/B auer

movements. In addition, central modulation of vestibular organ activity is also probably involved [D ic h g a n s et al, 1972], Vergence Movements These responses presumably primarily involve contractions of the medial rectus muscles (anterior rectus in fishes). They were observed dur ing stimulation of areas associated with the oculomotor nerves and nucle ar complex or the nearby MLF. Unilateral responses occurred when stim ulation was lateral to the midline or near one of the nerves. Convergence was associated only with midline stimulation. It is not known whether or not this was triggered by simultaneous activation of similar structures on both sides of the brain or by direct activation of structures on one side and subsequent triggering of the contralateral area through intrinsic connections in the oculomotor nucleus itself [see discussions in K orn and B ennett , 1971, 1972; W axman and P appas, 1971]. Our finding of a mid line convergence area in the vicinity of the oculomotor nucleus is consis tent with the results of electrical recording studies in this general area in goldfish which revealed single units that increase their firing rate during vergence movements [H ermann , 1971b]. Possibly similar midbrain areas involved with the control of vergence movements have been described in mammals including man [B ender and Shanzer , 1964; H oyt and D aroff , 1971]. A cknowledgements The authors would like to express their sincere appreciation to Dr. Robert S. K ellogg for his assistance throughout this study. Financial support was provided by the Air Force Office of Scientific Research through contract 73-2491.

References Akert , K.: Der visuelle Greifreflex. Helv. physiol, pharmac. Acta 7: 112-134

(1949). A kert , K. and W elker , W. I.: Problems and methods of anatomical localization; in Sheer Electrical stimulation of the brain, pp. 251-260 (University of Texas

Press, Austin 1961). my of the nervous system of vertebrates, including man (MacMillan, New York 1936).


Ariens Kappers , C. U.; H uber , G. C., and Crosby, E. C.: The comparative anato


127

Barale, F.; C orvaja, N., and P ompeiamo, O.: Vestibular influences on postural ac

tivity in frog. Archo ital. Biol. 109: 27-36 (1971). ponses to short periods of lowered gravitational force in blind goldfish. Space Life Scis. 1: 554-564 (1969). Baumgarten, R. J. von ; Baldrighi, G.; A tema , J., and Shillinger , G. L., jr.: Be havioral responses to linear accelerations in blind goldfish. Space Life Scis. 3: 25-33 (1971). Baumgarten, R. J. von ; Baldrighi, G., and Shillinger , G. L., jr.: Vestibular be havior of fish during diminished G-force and weightlessness. Aerospace Med. 43: 626-632 (1972). Bender , M. B. and S hanzer, S.: Oculomotor pathways defined by electric stimula tion and lesions in the brainstem of monkey; in Bender The oculomotor system, pp. 81-140 (Harper & Row, New York 1964). Braford, M. R. jr.: Projections of the corpus cerebelli in the goldfish. Anat. Rec. 166: 282 (1970). C lark, S. L.; C hung , M. Y.; Shine , L., and C lark, M. R.: Responses in free-swim ming fishes to electrical stimulation of the cerebellum. Am. J. Anat. 106: 121-132 (1960). C ohen , B.: Vestibulo-ocular relations; in Bach-y-R ita and C ollins The control of eye movements, pp. 105-148 (Academic Press, New York 1971). C ohen , B.; K azuyoshi, G.; Shanzer , S., and W eiss , A. H.: Eye movements induced by electric stimulation of the cerebellum in the alert cat. Expl. Neurol. 13: 145-162 (1965). C orvaja, N.; G rofova, L, and P ompeiano , O.: The origin, course and termination of vestibulospinal fibers in the toad. Brain Behav. Evol. 7: 401^123 (1973). D emski, L. S. and Bauer, D. G.: Eye movements and postural changes evoked by electrical stimulation of the fish brain. Soc. Neurosci., abstract 160 (1974). D emski, L. S. and G erald, J. W.: Sound production evoked by electrical stimula tion of brain in toadfish, Opsamts beta. Anim. Behav. 20: 507-513 (1972). D emski, L. S. and G erald, J. W.: Sound production and other behavioral effects of midbrain stimulation in free-swimming toadfish Opsanus beta. Brain Behav. Evol. 9: 41-59 (1974). D emski, L. S. and K nigge , K. M.: The telencephalon and hypothalamus of the bluegill (Lepomis macrochirus). Evoked feeding, aggressive and reproductive be havior with representative frontal sections. J. comp. Neurol. 143: 1-16 (1971). Dichgans, J.; Schm idt , C. L., and W ist , E. R.: Frequency modulation of afferent and efferent unit activity in the vestibular nerve by oculomotor impulses; in Brodal and P ompeiano Progress in brain research, vol. 37, pp. 449-456 (Else vier, Amsterdam 1972). Easter, S. S., jr.: Spontaneous eye movements in restrained goldfish. Vision Res. 11: 333-342 (1971). E aster, S. S., jr.: Pursuit eye movements in goldfish (Carassius aurattts). Vision Res. 12: 673-688 (1972). E bbesson, S. O. E. and C ampbell , C. B. G.: On the organization of cerebellar effer ent pathways in the nurse shark (Ginglymostoma cirratum). J. comp. Neurol. 152: 233-254 (1973).


Baumgarten, R. J. von ; A tema , J.; H ukuhara, T., and R ocker, M.: Behavioral res

128

D emski/B auer

H ermann, H. T.: Saccade-correlated potentials in optic tectum and cerebellum of

Carassius auratus. Brain Res. 26: 240-244 (1971a). Eye movement correlated units in mesencephalic oculomotor complex of goldfish. Brain Res. 35: 240-244 (1971b). H ermann, H. T. and C onstantine , M. M.: Eye movements in goldfish. Vision Res. 11: 313-331 (1971). H inojosa, R.: Synaptic ultrastructure in the tangential nucleus of the goldfish (Carassius auratus). Am. J. Anat. 137: 159-186 (1973). H oyt, W. F. and D aroff , R. B.: Supranuclear disorders of ocular control systems in man; in Bach-y -R ita and C ollins The control of eye movements, pp. 175-235 (Academic Press, New York 1971). H umason, G. L.: Animal tissue techniques (Freeman, San Francisco 1967). H ermann, H. T.:

J ohnstone , J. R. and M ark, R. F.: Evidence for efference copy fo r eye movem ents in fish. Comp. Biochem. Physiol. 30: 931-939 (1969). K idokoro, Y.: Cerebellar and vestibular control of fish oculomotor neurons; in L li-

Neurobiology of cerebellar evolution and development, pp. 257-276 (Ameri can Medical Association, Chicago 1969). K orn , H . and B ennett , M. V. L.: Dendritic and somatic impulse initiation in fish oculomotor neurons during vestibular nystagmus. Brain Res. 27: 169-175 (1971). K orn , H. and Bennett , M. V. L.: Electrotonic coupling between teleost oculomotor neurons; restriction to somatic regions and relation to function of somatic and dendritic sites of impulse initiation. Brain Res., Osaka 38: 433-439 (1972). L arsell, O.: The comparative anatomy and histology of the cerebellum from myxinoids through birds (University of Minnesota Press, Minneapolis 1967). L ow enstein , O.: The equilibrium function of the vertebrate labyrinth. Biol. Rev. 11: 113-145 (1936). Low enstein , O.: The labyrinth; in H oar and R andall Fish physiology, vol. 5, pp. 207-240 (Academic Press, New York 1971). M aler, L.; K arten, H. J., and Bennett , M. V. L.: The central connections of the anterior lateral line nerve of Gnathonemus petersii. J. comp. Neurol. 151: 67-84 (1973). M ehler , W. R.: A comparative anatomical survey of the vestibular nuclear complex in submammalian vertebrates; in Brodal and P ompeiano Progress in brain re search, vol. 37, pp. 55-67 (Elsevier, Amsterdam 1972). N auta, W. J. H. and E bbesson, S. O. E.: Contemporary research methods in neu roanatomy (Springer Verlag, New York 1970). Pearson, A. A.: The acustico-lateral centers and the cerebellum, with fiber connec tions of fishes. J. comp. Neurol. 65: 241-294 (1936). P recht , W. and L linás, R.: Comparative aspects of the vestibular input to the cere bellum; in L linás Neurobiology of cerebellar evolution and development, pp. 677-702 (American Medical Association, Chicago 1969). Ramón y C ajal, S.: Sur un noyau spécial du nerf vestibulaire des poissons et des oi seaux. Trab. Lab. Invest, biol. Univ. Madr. 6: 1-20 (1908). R on , S. and R obinson , D. A.: Eye movements evoked by cerebellar stimulation in the alert monkey. J. Ncurophysiol. 36: 1004-1022 (1973). Suzuki, J.; T okumasu, K., and G oto , K.: Eye movements from single utricular nerve stimulation in the cat. Acta oto-lar. 68: 350-362 (1969).


nás


129

T okumasu, K.; G oto , K., and C ohen , B.: Eye movements from vestibular nuclei

stimulation in monkeys. Ann. Otol. Rhinol. Lar. 78: 1105-1119 (1969). T raill, A. B. and M ark, R. F.: Optic and static contributions to ocular counter-ro

tation in carp. J. exp. Biol. 52: 109-124 (1970). W axman, S. G. and P appas, G. D.: An electron microscopic study of synaptic mor

L eo S. D emski, Ph. D., Department of Anatomy, LSU Medical Center, Bldg. 137,

1100 Florida Avenue, New Orleans, LA 70119 (USA)


phology in the oculomotor nuclei of three inframammalian species. J. comp. Neurol. 143: 41-72 (1971).

Eye and head movements evoked by electrical stimulation of monkey superior colliculus.

Adaptive Acceleration of Visually Evoked Smooth Eye Movements in Mice.

Character and origin of rotatory movements evoked by electrical stimulation of the caudate nucleus in cats.

Electrical responses evoked from the human brain.

Eye movements evoked by superior colliculus stimulation in the alert cat.

Eye movements elicited by electrical stimulation of area PG in the monkey.

Eye movements and brainstem neuronal responses evoked by cerebellar and vestibular stimulation in chicks.

Blurring by fixational eye movements.

Brain activation of eye movements in subjects with refractive error.

Neurohypophysial electrical activity in the anesthetized cat.

Functional linkage between the electrical activity in the vermal cerebellar cortex and saccadic eye movements.

Versive eye movements elicited by cortical stimulation of the human brain.

Self-face hallucination evoked by electrical stimulation of the human brain.

Sperm release evoked by electrical stimulation of the fish brain: a functional-anatomical study.

Proceedings: Eye movements and brain-stem dysfunction after head injury.

Gaze-evoked involuntary movements.

Electrical activity of superior colliculus associated with eye movements in alert cats.

Phenothiazine effects on cerebral-evoked potentials and eye movements in acute schizophrenics.

Processes controlling visually evoked movements.

Hypnosis and eye movements.

Dyslexia and eye movements.

Eye movements in Alzheimer's disease.

Locomotor-like leg movements evoked by rhythmic arm movements in humans.

Lightning eye movements.