Retinotectal Projection of the Adult Winter Flounder ( h eudop /euro nec tes a rn er ic a n us) L LUCKENBILL EDDS ANI) S C 5HARMA '' ' Department of Neuroputhology, tlarvurd Medtcal School and Department of h'ev.mxctence, Mental Returda taon Research Center, Children's HoPpttal Medtcul Center, Boston, Musmchuyetts 022 15 Department of Ophthalmology, net^ York iMedical College, N e w York, N P u York 10029 and Thp Munne Biological Laboratory, Woods Hole, Massachusetts 0254 3
ABSTRACT The winter flounder shifts the orientation of its body 90" at metamorphosis so that its left side is functionally ventral and its right side functionally dorsal. Concomitantly the left eye migrates onto the right side. The net result of these complex metamorphic changes is that the dorsoventral axes of the visual fields are perpendicular to the body rather than parallel as in most other teleosts. The developing flatfish may provide a resource for studying the formation of neural connections. for the change in orientation may necessitate some shift in connections in visrioinotor pathways. As a baseline for developmental studies, we have established thc retinotectal projectiori in adult winter flounder by means of anatomical tracing techniques (autoradiography and degeneration staining) and electrophysiological mapping techniques. The histological pattern of retinal afferents to the tectum is similar to that of other teleosts; affrrents are confined to the superficial white and gray zone, with a few fibers coursing in the deep white zone. Electrophysiological mapping shows that the visuotectal projection is complete over the entire extent of the tectum, symmetrical for right and left fields and patterned normally.
The winter floundvr (Pseudopkuronectes umericunus, Walbaum, fam. Pleuronectidae) is a species of flatfish which ranges from the tide mark to the edge of the great fishing banks of the eastern coast of North America. In its natural habitat the flounder lies partially submerged in silt so that only its large, somewhat turreted eye5 protrude above the substratum (Rigelow and Schroeder, '53).It feeds on polychaetes, amphipods and mollusks (Pearcy, '63, Richards, '63) apparently using visual cues, although to our knowledge there is no experimental proof for the visual basis of feeding behaviour in this species. In the laboratory it may remain motionless and follow a lure in the aquarium with its eyes. It is also capable of extremely rapid movement when disturbed or when capturing prey. Adaptation of all flatfish to their demerJ m M P NEUR, 173 307 318
sal habitat depends on the orientation of the body and the position of both eyes on either the right or the left side of the body, depending on the species. The winter flounder lies or swims with its left or blind side down and its right or eyed side up. (We use the ordinary anatomical axes to refer to the body of the winter flounder to eliminate confusion regarding terminology.) In order to understand how the natural orientation of the flounder is related to its anatomical axes, one can imagine that the dorsoventral axis of the body has been flipped 90" with respect to the dorsoventral axes of the eyes. If the eyes of other teleosts are rotated experimentally to produce a similar asymmetry of the visual fields with respect to the body, then misdirected visuomotor behavior results (Sperry, '48). But a group of fish as successful in exploiting their environment
307
308
L. LUCKENBILL-EDDS 4ND S. C . SHARMA
as the flatfish obviously are not misdirected. Thus, the unusual relationship between visual fields and the body axis raises the possibility that a component of the flounder’s visuomotor pathway departs from the usual telostean pattern in order to compensate for the asymmetry. Furthermore, flatfish do not begin life with this peculiar orientation; rather, as embryos and then planktonic larvae they are oriented like other teleosts. At the end of larval life, the fish transforms its posture and the position of its eyes during a period of metamorphosis, a time when developmental mechanisms in the visuomotor system might well compensate for the shift in Fymmetry. Before embarking on such a developmental study, we have determined the retinotectal projection of the adult winter flounder using degeneration and autoradiographic methods, as well as electrophysiological mapping techniques. MATERIALS AND METHODS
Adult winter flounder were collected from Long Island Sound, Cape Cod Sound and Vineyard Sound. The fish were maintained in aerated or running sea water or “Instant Ocean” at 8-10°C. Once acclimated to the laboratory they were fed minced clams. The sizes of the animals varied from 10-15 cm (young adults) to 2835 cm (sexually mature adults). Both size classes will be referred to as adult winter flounder. For routine histology, brains were fixed in Bouin’s fluid, embedded in paraffin, sectioned serially at 10 p either transversely or sagittally and stained with Bodian method for fibers, or with 1% thionin in acetate buffer (pH 4.0) for cell bodies. The optic tract was traced anatomically by removing either the right or the left eye from a series of 12 fish lightly anesthetized with MS-222. The retinal afferents were allowed to degenerate for periods of 2 to 17 days. At the time of sacrifice, the fish were anesthetized and the brain removed, fixed in formalin and processed according to Roth’s modification of the Nauta technique for teleosts (Ebbesson, ’70). Both
sagittal and transverse sections were examined from four favorable cases. Retinofugal fibers were also identified in a series of 14 flounder by injecting either the right eye or the left eye with 3H-proline (specific activity 43.6 Ci/mmole) and allowing the labeled amino acid to be incorporated into protein and transported to the brain. The amount of injected isotope and the period of transport varied as follows: one group of fish was injected with 5 pCi and sacrificed 18 hours later, the second with 25 pCi and sacrificed at 11 days and the third group with 40 pCi and sacrificed at seven days. The brains were processed for autoradiography (Landreth et al., ’75) and stained with 0.1%toluidine blue. The pattern of the visuotectal projection in 14 flounder was mapped by photically stimulating either the right or left eye and recording visually evoked potentials in the contralateral optic tectum. The fish was anesthetized with MS-222 (Tricaine), immobilized with tubocurare (0.01 cc), and its pharynx continuously perfused with aerated sea water. The stimulus was a white spot of light which subtended 5” of visual space and was placed 34 cm from the center of the eye. The exposed dorsal surface of the tectum was covered with parafin oil, then systematically penetrated with recording electrodes, platinum coated 2-4 p tip glass pipettes filled with Woods metal. The techniques used to produce the visuotopic maps were the same as those described previously (Gaze and Sharma, ’70). RESULTS
The gross organization of the visuotectal system and the histology of the optic tectum will be described first. Then the pattern of retinofugal fibers to the tectum will outlined as determined by degeneration and autoradiography experiments, followed by electrophysiological mapping of the visuotectal projection. To document both anatomical and physiological results, we have used material from the adult winter flounder, however, histological results include observations from brains of
FLOUNDER RETINOTECTAI, PROJECTIOK
309
A bbreuiations
c, cerebellum CGZ, central gray zone DM, dorsomedial lobe, optic tectum DWZ. deep white zone il, inferior lobe Is, lateral sulcus
ot, optic tract PCZ, periventricular gray Lone SWGZ, superficial white and gray zone tel. telencephalon VL, ventrolateral lobe, optic tectum
Fig. 1 The anterior end of a young adult winter flounder, hovering in its natural orientation, pigmented right side upward. Note that the dorsoventral (dv) axes of the eyes or visual fields are perpendicular to the dorsoventral (DV) axis o f t h e body. In this photo the fish is swimming in the water column. If the fish were lying on the bottom camouflaged with a layer of silt, only the eyes and the stalked external nares would protrude. (Courtesy of W. Saidel.)
newly metamorphosed (2-3 cm) winter flounder.
Organization of visuotectal system The eyes of the winter flounder are located well forward of the rostra1 end of the midbrain, a landmark anterior to which both internal and external structures are twisted to the right. The orientation of the eyes is such that the dorsoventral axes of the resting eyes are included in a plane that lies perpendicular to the anatomical dorsoventral axis of the body (fig. 1). The eyes are turretlike and extremely mobile, occasionally converging or diverging and often moving independently. The visual fields of both eyes overlap about 30"when the eyes are relaxed and looking out toward the dorsal and ventral fins respectively.
Since the eyes lie well anterior to the brain, the optic nerves are long, with the left nerve from the migrated eye being 36% shorter than the right (Murray, '74). At the chiasm the left nerve is dorsal in this species, so the nerves do not entwine during metamorphosis when the left eye migrates over to the right side (Parker, '03). Crossing at the chiasm is complete; neither degenerated nor radioactively labeled fibers appear in the optic tract ipsilateral to either an enucleated or an injected eye, respectively, and visually evoked responses in the ipsilateral tectum are lacking. Rostra1 to the tectum, the optic tract bifurcates into two large trunks, the dorsal trunk entering the dorsomedial (DM) region of the optic tectum, and the ventral trunk, the ventro-lateral (VL) region. The anatomical axes of the optic tectum
310
L. LUCKENBILL-EDDS AND S. C. SHAHMA
Fig. 2 Lateral view of the right side of the brain of an adult (35cm) winter flounder; the arrow points anteriorly. The optic tectum lies dorsal to the inferior lobe (il), posterior to the telencephalon (tel) and anterior to the cerebellum (c).The rostral optic tectum is divided by a lateral sulcus (Is) into dorsomedial (DM) and ventrolateral (VL) lobes. The anterior end of the D M lobe is recessed, providing a niche that accomodates the posterior region of the dorsal thalamus. The telencephalon (tel) is lifted here to show the optic tracts (ot).The bars at the top indicate the levels from which sections in figures 3 and 6 were cut. Note that a section through level 6 cuts the D M lobes more anteriorly than the VL lobe. Mm rule at bottom.
correspond to those of the body; thus, the roof of the midbrain lies ventral to the dorsal fin, as in other fish. Furthermore, despite the migration of the eyes at metamorphosis and position of both eyes on the right side of the adult fish, the right and left optic tecta are bilaterally symmetrical. For this reason we shall use the term “tectum” to apply to either side. The tectum is oval shaped when viewed dorsally, but when seen in section, it forms a roof that arches around the ventricle and slopes ventrally alongside part of the tegmentum (fig. 3 ) . The rostral tectum bulges forward dorsal and lateral to the dorsal thalamus and pretectal region; the cerebellar valvula in turn bulges into the ventricle of the caudal tectum. An unusual feature of the winter flounder is a lateral sulcus running anteroposteriorly in the rostral third of the tectum, dividing it into dorso-medial (DM) and ventrolateral (VL) lobes (figs.2,6A). These two lobes are positioned so that the rostral end of the VL lobe begins farther anterior than the DM lobe (figs. 2, 6A). Caudal to the lateral sulcus the posterior two-thirds of the tec-
tum curves smoothly from dorsal to ventral regions (fig. 3). The ventral region disappears near the end of the tectum, leaving only a dorsolateral crescent adjacent to the cerebellum.
Histology of the optic tectum The pattern of cells and fibers in the winter flounder tectum is similar to that of other teleosts (e.g., Leghissa, ’55; Ebbesson, ’68; Campbell and Ebbesson, ’69; Sharma, ’72; Vanegas and Ebbesson, ’73). Leghissa’s terminology is used regarding different layers of the optic tectum. Figure 4,a transverse section from the middle of the tectum stained with the Bodian method, illustrates the four main zones. From the ventricle to the pial surface they are: 1. periventricular gray zone (PGZ); 2. deep white fiber zone(DWZ); 3. central gray zone (CGZ); and 4. superficial white and gray zone (SWGZ). Since the SWGZ is the main target for retinal afferents in the tectum, it will be described in detail. The SWGZ is subdivided into two strata - a narrow marginal stratum with thin, lightly staining fibers and a deeper stratum
FLOUNDEH RETINOTECTAI. PROJECTION
containing external white and plexiform layers (fig. 4). The plexiform layer, (3) in figure 4, (stratum fibrosum et griseum superficialis of Ariens Kappers) of the deeper stratum is one-half to two-thirds the width of the entire SWGZ. The plexiform layer is bounded by three fibrous laminae on its su erficial side and by one fibrous lamina, (4 in figure 4, on its deep side adjacent to the CGZ. All three superficial laminae contain large caliber fibers; however, the outer and inner laminae, (1) and (2) in figure 4, also contain additional fibers of medium caliber which appear to arise in the retina. The histological pattern described above is the same along the entire anteroposterior axis of the tectum, although the extreme curvature and division into two lobes of the rostral region alter the three-dimensional pattern of layers (fig. 6). Variation does occur in the thickness of bundles of fibers in the SWGZ in a given mediolateral arc of the tectum. This variation is associated with changes in the orientation of the bundles with respect to the curvature of the tectum, and with the course of tracts within the tectum. In the mid-tectum, bundles of fibers in the extreme dorsomedial (DM) and ventrolateral (VL) portions are large and run anteroposteriorly. Bundles adjacent to the DM and VL extremes run obliquely with resepct to the anteroposterior axis, while those in the lateral region are parallel to the surface. In the rostral tectum near the posterior limit of the lateral sulcus the fibrous layers parallel the contour of the sulcus.
P
Retinal uferents in the optic tectum Both degeneration and autoradiographic experiments for tracing retinal input to the tectum gave the same labeling pattern. Since autoradiography provides more dramatic illustrations, we shall use results of this technique to document the location of optic fibers in the tectum. The major tectal area which retinal fibers innervate is the SWGZ; however, some optic fibers were found in the DWZ (figs. 3B, 5,6). No retinal afferents were encountered in other zones,
311
including the marginal layer of the SWGZ. Most of the degeneration or labeling in the superficial fibrous laminae (stratum opticum) of the SWGZ is associated with bundles of fibers of medium caliber within the outer and inner laminae, (1)and (2) of figure 5,while the large caliber fibers in both laminae are spared (fig. 5 ) . Labeled fibers of medium caliber are also found at the base of the SWGZ. This pattern of retinal afferents in the SWGZ prevails in both the dorsomedial (DM) and ventrolateral (VL) portions of the tectum, and along its entire rostrocaudal extent (figs. 3B, 6A). In the lateral tectum postefior to the lateral sulcus, the autoradiographic and degeneration results differ in the relative amount of labeled fibers. Few profiles of degenerated axons appear in the sector where the DM and VL regions abut, while the band of grains seems as intense as in the DM and VL regions (fig. 38). In the rostral tectum a few optic fibers innervate the DWZ; these fibers could not be traced in the caudalmost region of the tectum. If we correlate the histology of the tectum with the localization of retinal afferents, three major points emerge. First, retinal afferents run in fiber bundles and enter the external white and plexiform layer of the SWGZ; a few retinal afferents accompany fibers of the deep white zone. Second, the pattern of degeneration and labeling extends throughout the anteroposterior axis of the tectum. Third, retinal afferents are present in the lateral sector (between the DM and VL extremes), but there appear to be fewer of them, and when grouped into bundles, the fibers are oriented parallel to the tectal surface. The conclusion that retinal afferents spread along both the anteroposterior and mediolateral axes of the tectum is confirmed by electrophysiological mapping experiments presented below.
The zjisuotectal projection In order to “map” the visuotectal projection, the flounder was oriented with its anatomically left side down on a platform. In this, the natural position of the adult
312
I.. LIICKENBILL-EDDS AND S. C. SHARh4A
flounder, each eye looks outward, the right eye toward the ventral fin and the left eye toward the dorsal fin (fig. 1).A week before visuotopic mapping, one eye of a few fish had been removed to avoid spurious results due to overlapping visual fields. This precaution proved unnecessary since no visually-evoked potentials could be recorded from the ipsilateral tectum. The projection of either the right or left visual field to the contralateral optic tectum was complete and orderly, and patterned similarly to that of other teleosts (Schwassmann, ’68; Gaze and Sharma, ’70). Thus, the “nasal” portion of the sisual field is represented rostrally on the optic tectum and the “temporal” field caudally. The dorsal part of the visual field projects to the medial region of the tectum and the ventral field to the lateral and ventral parts of the tectum (fig. 7 ) . The visually evoked potentials recorded at each tectal point were usually from multiple units with receptive fields of 1020”wide, and occasionally from single units 6-8”wide. The receptive fields at electrode positions near the lateral sulcus overlapped to a greater extent than those outside this area. This resulted in significantly smaller receptive fields in the area surrounding the sulcus. A certain non-linearity exists between visual field points in the map and equally spaced electrode positions on the optic tectum. The receptive field points projecting to the lateral sulcus were spaced closer to each other and overlapped considerably as compared to the points in the rest of the visual field, which were more or less equally spaced.
time one eye migrates over the dorsal midline to the opposite side of the body. The migrating eye does not rotate upon its own axis, in so far as we can determine by the position of lesions made in eyes during migration. Although the other eye does not migrate, it does rotate with respect to the body so that its dorsoventral axis corresponds to that of the migrated r y e (fig. 1). The change in anatomical relationship between eyes and body at metamorphosis results in the dorsoventral axes of the visual fields being oriented perpendicular to the dorsoventral axis of the body, rather than parallel as in most other teleosts. Given this relationship in the adult flatfish, we have asked the following: Does the unusual anatomical relationship between eyes and body imply a different visuotectal projection as compared with other teleosts? Our results show that the winter flounder’s visuotectal projection is typically teleostean. That is, the projection is symmetrical with respect to right (non-migrated eye) and left (migrated eye) visual fields; the projection is complete in all three tectal axes and it is orderly. Thus, even though the dorsoventral axis of each eye is perpendicular to the dorsoventral axis of the body (rather than parallel as in other teleosts), the receptive fields lying in this plane of visual space project to the mediolateral optic tectum in an orderly way. That the projection is complete was demonstrated by mapping the receptive fields of points throughout the rostrocaudal axis of the tectum and in two-thirds of the dorsoventral axis, the usual area accessible to electrophysiological exploration. One observation did indicate a speDISCUSSION cialization in the visual system similar to The flatfish are teleosts which depart that found in species of the sea bass family from the usual body orientation by turning (Schwassmann, ’68). In the region of the onto one side, either the anatomically right lateral sulcus of the rostra1 tectum we or left, depending on the species. The side found units with overlapping receptive of the body turned downward becomes fields subtending about 10”of visual space, functionally “ventral”; the opposite side is in contrast to the units in the rest of the functionally “dorsal” and bears both eyes. tectum whose receptive fields subtend The orientation of the body and position of about 20” of space. This suggests that the the eyes on the same side is an adaptation temporal retina may contain an area cento the bottom-dwelling existence the adult tralis or similarly specialized region. Alassumes at the end of its larval life. At this though the position of the eyes on the head
FLOUNDER RETINOTECTAL PROJECTION
and their great mobility suggest binocular convergence for binocular fixation, recordings from the ipsilateral tectum gave no evidence of any projection. Lack of “ipsilateral projection” was confirmed by recording from animals with one enucleation. The results of anatomical tracing techniques, namely, the Roth stain following axonal degeneration and autoradiography of ganglion cell axons labeled with ‘Hproline, show that the main target of retinal afferents within the contralateral tecturn is the superficial white and gray zone (SWGZ), excluding the marginal layer. A few fibers supply the region of the deep white zone (DWZ) adjacent to the periventricular zone. Within the external white and plexiform layers of the SWGZ, the degenerated profiles are fewest in the lateral sector. This pattern arises apparently because the large bundles of fibers which are located in the dorsal and ventral regions give off fascicles to the fibrous laminae of the more lateral region of the tectum. From the superficial fibrous laminae the retinofugal axons move deeper in the external white and plexiform layer where they apparently terminate. The evidence for terminal fields in the deep two-thirds of the SWGZ is supported by two observations; fine degeneration appears there at short intervals following enucleation and relatively heavy labeling occurs there within a brief period following intra-ocular injection of 3H-proline (Ebbesson, ’70; and Landreth et al., ’75). The visuotectal projection in the winter flounder differs from the retinotectal projection in two flatfish of the sole family (Achirus lineatus and Trinectes masculatus) (Gulley et al., ’75).These investigators, using degeneration and autoradiographic techniques, found that the entire dorsoventral retinal projection is confined to the medial portion of the tectum in both species. In the peacock flounder, Bothus tunutus (Gulley et al., ’7S), the projection appears to resemble that of the winter flounder, particularly considering the sparseness of degenerated retinal afferents in the lateral sector of the
313
optic tectum of both species. One possible explanation for differences between sole and flounder may lie in the evolutionary divergence of the two families, each adapted to different life styles. The sole (Soleidae) are day burrowers and night feeders relying on senses other than vision to seek rey, whereas the flounder (Pleuronectidae are fierce predators who sight prey and swim either along the bottom or in the water column to capture it (Norman, ’35; Bigelow and Schroeder, ’53).The behavioral experiments necessary to verify the differences between Soleidae and Pleuronectidae have yet to be done. If our hypothesis is correct that the shift in orientation of the body of the winter flounder with respect to its visual fields necessitates some adjustment in the central nervous system to account for adaptive behavior, then the present results lead to at least two further possibilities. Either a shift has occurred in the visuotectal projection, but our methods are not sensitive enough to measure it, or central function has changed within the tectum or in other visuomotor integrative centers. If the latter is true, then it would be in line with Platt’s (’73) results on the vestibular system of bothid and pleuronectid flatfish in which a change in central function of the vestibular system appears to be responsible for postural change.
7
ACKNOWLEDGMENTS
We thank the New York City Aquarium, Doctor James Wallace of Harvard University and the Marine Biological Laboratory for supplying fish, and William Saidel, M.I.T., for providing figure 1. L.L.-E. supported by the Faculty Scholarship Fund of Smith College and NIH Grant NS-0970406. S.C.S. supported by NIH Grant EY 01426, NSF Grant GB 43506 and a summer fellowship by Marine Biological Laboratory, Woods Hole, Massachusetts. LITERATURE CITED Bigelow, H. B., and W. C. Schroeder 1953 Fishes of the Gulf of Maine. Fishery Bull. Fish and Wildlife Service, 53: 74. Campbell, C. B. G.. arid S. 0. E. Ebbesson 1969 The
314
L. LUCKENRII,I,.EDI)S 4YD b C SIIARMA
optic system of a teleost: Holocentrtcs re-examined. Brain Behav. and Evol., 2: 415-430. Ehbesson, S. 0. E. 1968 Retinal projections in two teleost fishes (Opsanus tau and Gymnothorax funehrid An experimental study with silver impregnation methods. Brain Behav. and Evol.. 1 : 134-154. 1970 The selective silver-impregnation of degenerating axons and their synaptic endings in non-mammalian species. In: Contemporary Research Methods in Neuroanatomy. W. J. H. Nauta and S. 0. E. Ebbesson, eds. Springer, New York, pp. 132-161. Gaze, R. M.,and S. C . Sharma 1970 Axial differences in the reinnervation of the goldfish optic tectum hy regenerating optic nerve fibers. Exp. Brain Res., 10: 151-181. Gulley, H. L.. M. Cochran and S. 0.E. Ebbesson 1975 The visual connections of the adult flatfish, Achims lineatus. J. Comp. Neur., 162: 309-320. Landreth, G. E., E. A. Neale, J. H. Neale, R. S. Duff, M. R. Braford, Jr., R. G. Northcutt and B. W. Agranoff 1975 Evaluation of 'H-proline for radioautographic tracing of axonal projections in the teleost visnal system. Brain Rcs., 91: 25-42. Leghissa, S. 1955 La struttura microscopica e la citoarchitettonica del tetto ottico dei pesci teleostei. Z. Anat. Entwcklgesch., 118: 427-463. Murray, M. 1974 Axonal transport in the asymmetric optic axons of flatfish. Exp. Neurol., 42: 636646. Norman, J. R. 1934 A systematic monograph of the flatfishes (Heterosomata). Vol. 1 , Psettodidae,
Bothidae, Plenronectidae. London, Trustees Brit. Museum, Johnson Reprint Corp. Parkcr, G. €1. 1903 The optic rhiasma in teleosts and its bearings on the asymmetry of the Hetcrosomata (flatfishes). Boll. Mus. Comp. Zool. Harv., 40: 22 1-24?,, Pearcy, W. G. 1963 Ecology of an estuarine population of winter flounder, Pseudopleuronec tes arnericanw Walbaum). IV. Food habits of larvae and juveniles. Bull. Bingham Oceanogr. Coll.. 18: 65-78, Platt, C. 1973 Central control of postural orientation in flatfish. I. Postural change dependcnce 011 central neural changes. J. Exp. Biol~,59: 491-521. Richards, S. W. 1963 The demersal fish population of Long Island Sound. 11. Food of the juveniles from a sand-shell locality (Station I). Bull. Bingham Oceanogr. Coll., 18: 32-71. 1963 The demersal fish population of Long Island Sound. 111. Food of the juveniles from a mud locality (Station 3A). Bull. Ringham Oceanogr. Coll., 18: 73-101, Schwassmann, H. 0. 1968 Visual projection upon the optic tectum in foveate marine tcleosts. Vision Res., 8 : 1337-1348. Sharma, S. C . 1972 The retinal projection in the goldfish: an cxperimental study. Brain Res., 39: 213223. Sperry, R. 1948 Patterning of central synapses in regeneration of the optic nerbe in teleosts. Phlsiol. Zool., 21. 351-361. Vanegas, H., and S. 0. E. Ebbesson 1973 Retinal projections in the perch-like teleost Eugerres plumieri. J. Comp. Neur., 151: 331-358.
PLATE 1 EXPLANATION OF FICIIRES
3 Transverse sections through the midtectum posterior to the lateral sulcus at level 3 indicated in figure 2. .4 is a Bodian stain showing that the pattern of layering is similar throughout the dorsoventral !DVj axis of the tectum. B is a dark field autoradiograph of another brain prepared after injection of the contralateral eye with 40 pCi "1-proline seven days beforc sacrifice. The zones of labeled retinal aEerents in the SWGZ and DWZ run across the entire UV axis. The grains over the PGZ may represent transynaptic labeling. 4 Bodian-stained transverse section of the tectum showing the four major tectal zones: SWGZ, CCZ, DWZ and PGZ. The SWGZ consists of a superficial marginal stratum (arrowhead) and an external white and plexiform stratum. The latter Ftratum is subdivided into outer fibrous laminae (1 and 2), a plexiform layer (3) and a deep fibrous lamina (4) adjacent to the CGZ. 5
Paired bright field and dark field autoradiographs of a dorsal sector (A and B) and a ventral sector (C and D) of the same tectum. Labeled retinal afferents are found in fibrous laminae ( 1 , 2 , 4 )and in the plexiform layer (3) of the SWGZ. The DWZ is too lightly 1ahr:led here to be illustrated. Because of the difference in width o f t h e tecta in figures 4 and 5, only the layers of the SWGZ are in register. Contralateral cye was injected with 5 pCi "-proline; survival time was 18 hours. Toluidine blue stain.
FLOUNDER RETINOTECTAL PROJECTION L. Luckenbill-Edds and S. C . Sharma
PLrZTE 1
PLATE 2 FXP1.4NkTION OF FIGURES
6 Toluidine blue-stained cross section through the rostral tectum in the region of the lateral sulcus (Is). All zones of the ventrolateral (VL) lobe are present at this level, whereas only the SWGZ and CGZ appear this far rostral in the dorsomedial lobe (DM).The contralateral eye was injected with 40 pCi of 3Hproline and the brain processed for autoradiography after a survival time of seven days. The band of silver grains in the SWGZ extends across both lobes and the lateral sulcus. Arrows indicate the sectors enlarged in B, C and D, all dark field micrographs. In all sectors heavy labeling appears in the white plexiform stratum of the SWGZ, specifically in fibrous layers (1, 2, 4) and in the plexiform layer (3). The thickness of the SWGZ varies in the sectors illustrated because of the differences in curvature of the three regions.
316
FLOUNDER RETINOTECTAL PROJECTION L. Luckenhill-Edcls and S . C. Sharma
PLATE 2
PLATE 3
FLOUNDER RETINOTECTAL PROJECTION L. Luckenbill-Edds and S. C. Sharma
Superior
-m L
Inferior
Rostra1
Left Visual Field
Right Optic Tectum
EXPLANATION OF FIGURES
7 The representation of the left visual field on the right optic tectum of an adult winter flounder. The numbers on the tectal diagram (right) represent electrode positions, for each of which the corresponding optimal stimulus position is indicated on the chart of the visual field (left).The chart extends for 100"outwards from the center of the visual field. The nasal visual field is represented rostrally on the tectum and the caudal field temporally. The superior periphery of the field projects to near the medial edge of the tectum.
318
7
ABSTRACT The winter flounder shifts the orientation of its body 90" at metamorphosis so that its left side is functionally ventral and its right side functionally dorsal. Concomitantly the left eye migrates onto the right side. The net result of these complex metamorphic changes is that the dorsoventral axes of the visual fields are perpendicular to the body rather than parallel as in most other teleosts. The developing flatfish may provide a resource for studying the formation of neural connections. for the change in orientation may necessitate some shift in connections in visrioinotor pathways. As a baseline for developmental studies, we have established thc retinotectal projectiori in adult winter flounder by means of anatomical tracing techniques (autoradiography and degeneration staining) and electrophysiological mapping techniques. The histological pattern of retinal afferents to the tectum is similar to that of other teleosts; affrrents are confined to the superficial white and gray zone, with a few fibers coursing in the deep white zone. Electrophysiological mapping shows that the visuotectal projection is complete over the entire extent of the tectum, symmetrical for right and left fields and patterned normally.
The winter floundvr (Pseudopkuronectes umericunus, Walbaum, fam. Pleuronectidae) is a species of flatfish which ranges from the tide mark to the edge of the great fishing banks of the eastern coast of North America. In its natural habitat the flounder lies partially submerged in silt so that only its large, somewhat turreted eye5 protrude above the substratum (Rigelow and Schroeder, '53).It feeds on polychaetes, amphipods and mollusks (Pearcy, '63, Richards, '63) apparently using visual cues, although to our knowledge there is no experimental proof for the visual basis of feeding behaviour in this species. In the laboratory it may remain motionless and follow a lure in the aquarium with its eyes. It is also capable of extremely rapid movement when disturbed or when capturing prey. Adaptation of all flatfish to their demerJ m M P NEUR, 173 307 318
sal habitat depends on the orientation of the body and the position of both eyes on either the right or the left side of the body, depending on the species. The winter flounder lies or swims with its left or blind side down and its right or eyed side up. (We use the ordinary anatomical axes to refer to the body of the winter flounder to eliminate confusion regarding terminology.) In order to understand how the natural orientation of the flounder is related to its anatomical axes, one can imagine that the dorsoventral axis of the body has been flipped 90" with respect to the dorsoventral axes of the eyes. If the eyes of other teleosts are rotated experimentally to produce a similar asymmetry of the visual fields with respect to the body, then misdirected visuomotor behavior results (Sperry, '48). But a group of fish as successful in exploiting their environment
307
308
L. LUCKENBILL-EDDS 4ND S. C . SHARMA
as the flatfish obviously are not misdirected. Thus, the unusual relationship between visual fields and the body axis raises the possibility that a component of the flounder’s visuomotor pathway departs from the usual telostean pattern in order to compensate for the asymmetry. Furthermore, flatfish do not begin life with this peculiar orientation; rather, as embryos and then planktonic larvae they are oriented like other teleosts. At the end of larval life, the fish transforms its posture and the position of its eyes during a period of metamorphosis, a time when developmental mechanisms in the visuomotor system might well compensate for the shift in Fymmetry. Before embarking on such a developmental study, we have determined the retinotectal projection of the adult winter flounder using degeneration and autoradiographic methods, as well as electrophysiological mapping techniques. MATERIALS AND METHODS
Adult winter flounder were collected from Long Island Sound, Cape Cod Sound and Vineyard Sound. The fish were maintained in aerated or running sea water or “Instant Ocean” at 8-10°C. Once acclimated to the laboratory they were fed minced clams. The sizes of the animals varied from 10-15 cm (young adults) to 2835 cm (sexually mature adults). Both size classes will be referred to as adult winter flounder. For routine histology, brains were fixed in Bouin’s fluid, embedded in paraffin, sectioned serially at 10 p either transversely or sagittally and stained with Bodian method for fibers, or with 1% thionin in acetate buffer (pH 4.0) for cell bodies. The optic tract was traced anatomically by removing either the right or the left eye from a series of 12 fish lightly anesthetized with MS-222. The retinal afferents were allowed to degenerate for periods of 2 to 17 days. At the time of sacrifice, the fish were anesthetized and the brain removed, fixed in formalin and processed according to Roth’s modification of the Nauta technique for teleosts (Ebbesson, ’70). Both
sagittal and transverse sections were examined from four favorable cases. Retinofugal fibers were also identified in a series of 14 flounder by injecting either the right eye or the left eye with 3H-proline (specific activity 43.6 Ci/mmole) and allowing the labeled amino acid to be incorporated into protein and transported to the brain. The amount of injected isotope and the period of transport varied as follows: one group of fish was injected with 5 pCi and sacrificed 18 hours later, the second with 25 pCi and sacrificed at 11 days and the third group with 40 pCi and sacrificed at seven days. The brains were processed for autoradiography (Landreth et al., ’75) and stained with 0.1%toluidine blue. The pattern of the visuotectal projection in 14 flounder was mapped by photically stimulating either the right or left eye and recording visually evoked potentials in the contralateral optic tectum. The fish was anesthetized with MS-222 (Tricaine), immobilized with tubocurare (0.01 cc), and its pharynx continuously perfused with aerated sea water. The stimulus was a white spot of light which subtended 5” of visual space and was placed 34 cm from the center of the eye. The exposed dorsal surface of the tectum was covered with parafin oil, then systematically penetrated with recording electrodes, platinum coated 2-4 p tip glass pipettes filled with Woods metal. The techniques used to produce the visuotopic maps were the same as those described previously (Gaze and Sharma, ’70). RESULTS
The gross organization of the visuotectal system and the histology of the optic tectum will be described first. Then the pattern of retinofugal fibers to the tectum will outlined as determined by degeneration and autoradiography experiments, followed by electrophysiological mapping of the visuotectal projection. To document both anatomical and physiological results, we have used material from the adult winter flounder, however, histological results include observations from brains of
FLOUNDER RETINOTECTAI, PROJECTIOK
309
A bbreuiations
c, cerebellum CGZ, central gray zone DM, dorsomedial lobe, optic tectum DWZ. deep white zone il, inferior lobe Is, lateral sulcus
ot, optic tract PCZ, periventricular gray Lone SWGZ, superficial white and gray zone tel. telencephalon VL, ventrolateral lobe, optic tectum
Fig. 1 The anterior end of a young adult winter flounder, hovering in its natural orientation, pigmented right side upward. Note that the dorsoventral (dv) axes of the eyes or visual fields are perpendicular to the dorsoventral (DV) axis o f t h e body. In this photo the fish is swimming in the water column. If the fish were lying on the bottom camouflaged with a layer of silt, only the eyes and the stalked external nares would protrude. (Courtesy of W. Saidel.)
newly metamorphosed (2-3 cm) winter flounder.
Organization of visuotectal system The eyes of the winter flounder are located well forward of the rostra1 end of the midbrain, a landmark anterior to which both internal and external structures are twisted to the right. The orientation of the eyes is such that the dorsoventral axes of the resting eyes are included in a plane that lies perpendicular to the anatomical dorsoventral axis of the body (fig. 1). The eyes are turretlike and extremely mobile, occasionally converging or diverging and often moving independently. The visual fields of both eyes overlap about 30"when the eyes are relaxed and looking out toward the dorsal and ventral fins respectively.
Since the eyes lie well anterior to the brain, the optic nerves are long, with the left nerve from the migrated eye being 36% shorter than the right (Murray, '74). At the chiasm the left nerve is dorsal in this species, so the nerves do not entwine during metamorphosis when the left eye migrates over to the right side (Parker, '03). Crossing at the chiasm is complete; neither degenerated nor radioactively labeled fibers appear in the optic tract ipsilateral to either an enucleated or an injected eye, respectively, and visually evoked responses in the ipsilateral tectum are lacking. Rostra1 to the tectum, the optic tract bifurcates into two large trunks, the dorsal trunk entering the dorsomedial (DM) region of the optic tectum, and the ventral trunk, the ventro-lateral (VL) region. The anatomical axes of the optic tectum
310
L. LUCKENBILL-EDDS AND S. C. SHAHMA
Fig. 2 Lateral view of the right side of the brain of an adult (35cm) winter flounder; the arrow points anteriorly. The optic tectum lies dorsal to the inferior lobe (il), posterior to the telencephalon (tel) and anterior to the cerebellum (c).The rostral optic tectum is divided by a lateral sulcus (Is) into dorsomedial (DM) and ventrolateral (VL) lobes. The anterior end of the D M lobe is recessed, providing a niche that accomodates the posterior region of the dorsal thalamus. The telencephalon (tel) is lifted here to show the optic tracts (ot).The bars at the top indicate the levels from which sections in figures 3 and 6 were cut. Note that a section through level 6 cuts the D M lobes more anteriorly than the VL lobe. Mm rule at bottom.
correspond to those of the body; thus, the roof of the midbrain lies ventral to the dorsal fin, as in other fish. Furthermore, despite the migration of the eyes at metamorphosis and position of both eyes on the right side of the adult fish, the right and left optic tecta are bilaterally symmetrical. For this reason we shall use the term “tectum” to apply to either side. The tectum is oval shaped when viewed dorsally, but when seen in section, it forms a roof that arches around the ventricle and slopes ventrally alongside part of the tegmentum (fig. 3 ) . The rostral tectum bulges forward dorsal and lateral to the dorsal thalamus and pretectal region; the cerebellar valvula in turn bulges into the ventricle of the caudal tectum. An unusual feature of the winter flounder is a lateral sulcus running anteroposteriorly in the rostral third of the tectum, dividing it into dorso-medial (DM) and ventrolateral (VL) lobes (figs.2,6A). These two lobes are positioned so that the rostral end of the VL lobe begins farther anterior than the DM lobe (figs. 2, 6A). Caudal to the lateral sulcus the posterior two-thirds of the tec-
tum curves smoothly from dorsal to ventral regions (fig. 3). The ventral region disappears near the end of the tectum, leaving only a dorsolateral crescent adjacent to the cerebellum.
Histology of the optic tectum The pattern of cells and fibers in the winter flounder tectum is similar to that of other teleosts (e.g., Leghissa, ’55; Ebbesson, ’68; Campbell and Ebbesson, ’69; Sharma, ’72; Vanegas and Ebbesson, ’73). Leghissa’s terminology is used regarding different layers of the optic tectum. Figure 4,a transverse section from the middle of the tectum stained with the Bodian method, illustrates the four main zones. From the ventricle to the pial surface they are: 1. periventricular gray zone (PGZ); 2. deep white fiber zone(DWZ); 3. central gray zone (CGZ); and 4. superficial white and gray zone (SWGZ). Since the SWGZ is the main target for retinal afferents in the tectum, it will be described in detail. The SWGZ is subdivided into two strata - a narrow marginal stratum with thin, lightly staining fibers and a deeper stratum
FLOUNDEH RETINOTECTAI. PROJECTION
containing external white and plexiform layers (fig. 4). The plexiform layer, (3) in figure 4, (stratum fibrosum et griseum superficialis of Ariens Kappers) of the deeper stratum is one-half to two-thirds the width of the entire SWGZ. The plexiform layer is bounded by three fibrous laminae on its su erficial side and by one fibrous lamina, (4 in figure 4, on its deep side adjacent to the CGZ. All three superficial laminae contain large caliber fibers; however, the outer and inner laminae, (1) and (2) in figure 4, also contain additional fibers of medium caliber which appear to arise in the retina. The histological pattern described above is the same along the entire anteroposterior axis of the tectum, although the extreme curvature and division into two lobes of the rostral region alter the three-dimensional pattern of layers (fig. 6). Variation does occur in the thickness of bundles of fibers in the SWGZ in a given mediolateral arc of the tectum. This variation is associated with changes in the orientation of the bundles with respect to the curvature of the tectum, and with the course of tracts within the tectum. In the mid-tectum, bundles of fibers in the extreme dorsomedial (DM) and ventrolateral (VL) portions are large and run anteroposteriorly. Bundles adjacent to the DM and VL extremes run obliquely with resepct to the anteroposterior axis, while those in the lateral region are parallel to the surface. In the rostral tectum near the posterior limit of the lateral sulcus the fibrous layers parallel the contour of the sulcus.
P
Retinal uferents in the optic tectum Both degeneration and autoradiographic experiments for tracing retinal input to the tectum gave the same labeling pattern. Since autoradiography provides more dramatic illustrations, we shall use results of this technique to document the location of optic fibers in the tectum. The major tectal area which retinal fibers innervate is the SWGZ; however, some optic fibers were found in the DWZ (figs. 3B, 5,6). No retinal afferents were encountered in other zones,
311
including the marginal layer of the SWGZ. Most of the degeneration or labeling in the superficial fibrous laminae (stratum opticum) of the SWGZ is associated with bundles of fibers of medium caliber within the outer and inner laminae, (1)and (2) of figure 5,while the large caliber fibers in both laminae are spared (fig. 5 ) . Labeled fibers of medium caliber are also found at the base of the SWGZ. This pattern of retinal afferents in the SWGZ prevails in both the dorsomedial (DM) and ventrolateral (VL) portions of the tectum, and along its entire rostrocaudal extent (figs. 3B, 6A). In the lateral tectum postefior to the lateral sulcus, the autoradiographic and degeneration results differ in the relative amount of labeled fibers. Few profiles of degenerated axons appear in the sector where the DM and VL regions abut, while the band of grains seems as intense as in the DM and VL regions (fig. 38). In the rostral tectum a few optic fibers innervate the DWZ; these fibers could not be traced in the caudalmost region of the tectum. If we correlate the histology of the tectum with the localization of retinal afferents, three major points emerge. First, retinal afferents run in fiber bundles and enter the external white and plexiform layer of the SWGZ; a few retinal afferents accompany fibers of the deep white zone. Second, the pattern of degeneration and labeling extends throughout the anteroposterior axis of the tectum. Third, retinal afferents are present in the lateral sector (between the DM and VL extremes), but there appear to be fewer of them, and when grouped into bundles, the fibers are oriented parallel to the tectal surface. The conclusion that retinal afferents spread along both the anteroposterior and mediolateral axes of the tectum is confirmed by electrophysiological mapping experiments presented below.
The zjisuotectal projection In order to “map” the visuotectal projection, the flounder was oriented with its anatomically left side down on a platform. In this, the natural position of the adult
312
I.. LIICKENBILL-EDDS AND S. C. SHARh4A
flounder, each eye looks outward, the right eye toward the ventral fin and the left eye toward the dorsal fin (fig. 1).A week before visuotopic mapping, one eye of a few fish had been removed to avoid spurious results due to overlapping visual fields. This precaution proved unnecessary since no visually-evoked potentials could be recorded from the ipsilateral tectum. The projection of either the right or left visual field to the contralateral optic tectum was complete and orderly, and patterned similarly to that of other teleosts (Schwassmann, ’68; Gaze and Sharma, ’70). Thus, the “nasal” portion of the sisual field is represented rostrally on the optic tectum and the “temporal” field caudally. The dorsal part of the visual field projects to the medial region of the tectum and the ventral field to the lateral and ventral parts of the tectum (fig. 7 ) . The visually evoked potentials recorded at each tectal point were usually from multiple units with receptive fields of 1020”wide, and occasionally from single units 6-8”wide. The receptive fields at electrode positions near the lateral sulcus overlapped to a greater extent than those outside this area. This resulted in significantly smaller receptive fields in the area surrounding the sulcus. A certain non-linearity exists between visual field points in the map and equally spaced electrode positions on the optic tectum. The receptive field points projecting to the lateral sulcus were spaced closer to each other and overlapped considerably as compared to the points in the rest of the visual field, which were more or less equally spaced.
time one eye migrates over the dorsal midline to the opposite side of the body. The migrating eye does not rotate upon its own axis, in so far as we can determine by the position of lesions made in eyes during migration. Although the other eye does not migrate, it does rotate with respect to the body so that its dorsoventral axis corresponds to that of the migrated r y e (fig. 1). The change in anatomical relationship between eyes and body at metamorphosis results in the dorsoventral axes of the visual fields being oriented perpendicular to the dorsoventral axis of the body, rather than parallel as in most other teleosts. Given this relationship in the adult flatfish, we have asked the following: Does the unusual anatomical relationship between eyes and body imply a different visuotectal projection as compared with other teleosts? Our results show that the winter flounder’s visuotectal projection is typically teleostean. That is, the projection is symmetrical with respect to right (non-migrated eye) and left (migrated eye) visual fields; the projection is complete in all three tectal axes and it is orderly. Thus, even though the dorsoventral axis of each eye is perpendicular to the dorsoventral axis of the body (rather than parallel as in other teleosts), the receptive fields lying in this plane of visual space project to the mediolateral optic tectum in an orderly way. That the projection is complete was demonstrated by mapping the receptive fields of points throughout the rostrocaudal axis of the tectum and in two-thirds of the dorsoventral axis, the usual area accessible to electrophysiological exploration. One observation did indicate a speDISCUSSION cialization in the visual system similar to The flatfish are teleosts which depart that found in species of the sea bass family from the usual body orientation by turning (Schwassmann, ’68). In the region of the onto one side, either the anatomically right lateral sulcus of the rostra1 tectum we or left, depending on the species. The side found units with overlapping receptive of the body turned downward becomes fields subtending about 10”of visual space, functionally “ventral”; the opposite side is in contrast to the units in the rest of the functionally “dorsal” and bears both eyes. tectum whose receptive fields subtend The orientation of the body and position of about 20” of space. This suggests that the the eyes on the same side is an adaptation temporal retina may contain an area cento the bottom-dwelling existence the adult tralis or similarly specialized region. Alassumes at the end of its larval life. At this though the position of the eyes on the head
FLOUNDER RETINOTECTAL PROJECTION
and their great mobility suggest binocular convergence for binocular fixation, recordings from the ipsilateral tectum gave no evidence of any projection. Lack of “ipsilateral projection” was confirmed by recording from animals with one enucleation. The results of anatomical tracing techniques, namely, the Roth stain following axonal degeneration and autoradiography of ganglion cell axons labeled with ‘Hproline, show that the main target of retinal afferents within the contralateral tecturn is the superficial white and gray zone (SWGZ), excluding the marginal layer. A few fibers supply the region of the deep white zone (DWZ) adjacent to the periventricular zone. Within the external white and plexiform layers of the SWGZ, the degenerated profiles are fewest in the lateral sector. This pattern arises apparently because the large bundles of fibers which are located in the dorsal and ventral regions give off fascicles to the fibrous laminae of the more lateral region of the tectum. From the superficial fibrous laminae the retinofugal axons move deeper in the external white and plexiform layer where they apparently terminate. The evidence for terminal fields in the deep two-thirds of the SWGZ is supported by two observations; fine degeneration appears there at short intervals following enucleation and relatively heavy labeling occurs there within a brief period following intra-ocular injection of 3H-proline (Ebbesson, ’70; and Landreth et al., ’75). The visuotectal projection in the winter flounder differs from the retinotectal projection in two flatfish of the sole family (Achirus lineatus and Trinectes masculatus) (Gulley et al., ’75).These investigators, using degeneration and autoradiographic techniques, found that the entire dorsoventral retinal projection is confined to the medial portion of the tectum in both species. In the peacock flounder, Bothus tunutus (Gulley et al., ’7S), the projection appears to resemble that of the winter flounder, particularly considering the sparseness of degenerated retinal afferents in the lateral sector of the
313
optic tectum of both species. One possible explanation for differences between sole and flounder may lie in the evolutionary divergence of the two families, each adapted to different life styles. The sole (Soleidae) are day burrowers and night feeders relying on senses other than vision to seek rey, whereas the flounder (Pleuronectidae are fierce predators who sight prey and swim either along the bottom or in the water column to capture it (Norman, ’35; Bigelow and Schroeder, ’53).The behavioral experiments necessary to verify the differences between Soleidae and Pleuronectidae have yet to be done. If our hypothesis is correct that the shift in orientation of the body of the winter flounder with respect to its visual fields necessitates some adjustment in the central nervous system to account for adaptive behavior, then the present results lead to at least two further possibilities. Either a shift has occurred in the visuotectal projection, but our methods are not sensitive enough to measure it, or central function has changed within the tectum or in other visuomotor integrative centers. If the latter is true, then it would be in line with Platt’s (’73) results on the vestibular system of bothid and pleuronectid flatfish in which a change in central function of the vestibular system appears to be responsible for postural change.
7
ACKNOWLEDGMENTS
We thank the New York City Aquarium, Doctor James Wallace of Harvard University and the Marine Biological Laboratory for supplying fish, and William Saidel, M.I.T., for providing figure 1. L.L.-E. supported by the Faculty Scholarship Fund of Smith College and NIH Grant NS-0970406. S.C.S. supported by NIH Grant EY 01426, NSF Grant GB 43506 and a summer fellowship by Marine Biological Laboratory, Woods Hole, Massachusetts. LITERATURE CITED Bigelow, H. B., and W. C. Schroeder 1953 Fishes of the Gulf of Maine. Fishery Bull. Fish and Wildlife Service, 53: 74. Campbell, C. B. G.. arid S. 0. E. Ebbesson 1969 The
314
L. LUCKENRII,I,.EDI)S 4YD b C SIIARMA
optic system of a teleost: Holocentrtcs re-examined. Brain Behav. and Evol., 2: 415-430. Ehbesson, S. 0. E. 1968 Retinal projections in two teleost fishes (Opsanus tau and Gymnothorax funehrid An experimental study with silver impregnation methods. Brain Behav. and Evol.. 1 : 134-154. 1970 The selective silver-impregnation of degenerating axons and their synaptic endings in non-mammalian species. In: Contemporary Research Methods in Neuroanatomy. W. J. H. Nauta and S. 0. E. Ebbesson, eds. Springer, New York, pp. 132-161. Gaze, R. M.,and S. C . Sharma 1970 Axial differences in the reinnervation of the goldfish optic tectum hy regenerating optic nerve fibers. Exp. Brain Res., 10: 151-181. Gulley, H. L.. M. Cochran and S. 0.E. Ebbesson 1975 The visual connections of the adult flatfish, Achims lineatus. J. Comp. Neur., 162: 309-320. Landreth, G. E., E. A. Neale, J. H. Neale, R. S. Duff, M. R. Braford, Jr., R. G. Northcutt and B. W. Agranoff 1975 Evaluation of 'H-proline for radioautographic tracing of axonal projections in the teleost visnal system. Brain Rcs., 91: 25-42. Leghissa, S. 1955 La struttura microscopica e la citoarchitettonica del tetto ottico dei pesci teleostei. Z. Anat. Entwcklgesch., 118: 427-463. Murray, M. 1974 Axonal transport in the asymmetric optic axons of flatfish. Exp. Neurol., 42: 636646. Norman, J. R. 1934 A systematic monograph of the flatfishes (Heterosomata). Vol. 1 , Psettodidae,
Bothidae, Plenronectidae. London, Trustees Brit. Museum, Johnson Reprint Corp. Parkcr, G. €1. 1903 The optic rhiasma in teleosts and its bearings on the asymmetry of the Hetcrosomata (flatfishes). Boll. Mus. Comp. Zool. Harv., 40: 22 1-24?,, Pearcy, W. G. 1963 Ecology of an estuarine population of winter flounder, Pseudopleuronec tes arnericanw Walbaum). IV. Food habits of larvae and juveniles. Bull. Bingham Oceanogr. Coll.. 18: 65-78, Platt, C. 1973 Central control of postural orientation in flatfish. I. Postural change dependcnce 011 central neural changes. J. Exp. Biol~,59: 491-521. Richards, S. W. 1963 The demersal fish population of Long Island Sound. 11. Food of the juveniles from a sand-shell locality (Station I). Bull. Bingham Oceanogr. Coll., 18: 32-71. 1963 The demersal fish population of Long Island Sound. 111. Food of the juveniles from a mud locality (Station 3A). Bull. Ringham Oceanogr. Coll., 18: 73-101, Schwassmann, H. 0. 1968 Visual projection upon the optic tectum in foveate marine tcleosts. Vision Res., 8 : 1337-1348. Sharma, S. C . 1972 The retinal projection in the goldfish: an cxperimental study. Brain Res., 39: 213223. Sperry, R. 1948 Patterning of central synapses in regeneration of the optic nerbe in teleosts. Phlsiol. Zool., 21. 351-361. Vanegas, H., and S. 0. E. Ebbesson 1973 Retinal projections in the perch-like teleost Eugerres plumieri. J. Comp. Neur., 151: 331-358.
PLATE 1 EXPLANATION OF FICIIRES
3 Transverse sections through the midtectum posterior to the lateral sulcus at level 3 indicated in figure 2. .4 is a Bodian stain showing that the pattern of layering is similar throughout the dorsoventral !DVj axis of the tectum. B is a dark field autoradiograph of another brain prepared after injection of the contralateral eye with 40 pCi "1-proline seven days beforc sacrifice. The zones of labeled retinal aEerents in the SWGZ and DWZ run across the entire UV axis. The grains over the PGZ may represent transynaptic labeling. 4 Bodian-stained transverse section of the tectum showing the four major tectal zones: SWGZ, CCZ, DWZ and PGZ. The SWGZ consists of a superficial marginal stratum (arrowhead) and an external white and plexiform stratum. The latter Ftratum is subdivided into outer fibrous laminae (1 and 2), a plexiform layer (3) and a deep fibrous lamina (4) adjacent to the CGZ. 5
Paired bright field and dark field autoradiographs of a dorsal sector (A and B) and a ventral sector (C and D) of the same tectum. Labeled retinal afferents are found in fibrous laminae ( 1 , 2 , 4 )and in the plexiform layer (3) of the SWGZ. The DWZ is too lightly 1ahr:led here to be illustrated. Because of the difference in width o f t h e tecta in figures 4 and 5, only the layers of the SWGZ are in register. Contralateral cye was injected with 5 pCi "-proline; survival time was 18 hours. Toluidine blue stain.
FLOUNDER RETINOTECTAL PROJECTION L. Luckenbill-Edds and S. C . Sharma
PLrZTE 1
PLATE 2 FXP1.4NkTION OF FIGURES
6 Toluidine blue-stained cross section through the rostral tectum in the region of the lateral sulcus (Is). All zones of the ventrolateral (VL) lobe are present at this level, whereas only the SWGZ and CGZ appear this far rostral in the dorsomedial lobe (DM).The contralateral eye was injected with 40 pCi of 3Hproline and the brain processed for autoradiography after a survival time of seven days. The band of silver grains in the SWGZ extends across both lobes and the lateral sulcus. Arrows indicate the sectors enlarged in B, C and D, all dark field micrographs. In all sectors heavy labeling appears in the white plexiform stratum of the SWGZ, specifically in fibrous layers (1, 2, 4) and in the plexiform layer (3). The thickness of the SWGZ varies in the sectors illustrated because of the differences in curvature of the three regions.
316
FLOUNDER RETINOTECTAL PROJECTION L. Luckenhill-Edcls and S . C. Sharma
PLATE 2
PLATE 3
FLOUNDER RETINOTECTAL PROJECTION L. Luckenbill-Edds and S. C. Sharma
Superior
-m L
Inferior
Rostra1
Left Visual Field
Right Optic Tectum
EXPLANATION OF FIGURES
7 The representation of the left visual field on the right optic tectum of an adult winter flounder. The numbers on the tectal diagram (right) represent electrode positions, for each of which the corresponding optimal stimulus position is indicated on the chart of the visual field (left).The chart extends for 100"outwards from the center of the visual field. The nasal visual field is represented rostrally on the tectum and the caudal field temporally. The superior periphery of the field projects to near the medial edge of the tectum.
318
7
