A GABAergic tecto–tegmento–tectal pathway in pigeons

Previous studies have demonstrated that the optic tecta of the left and right brain halves reciprocally inhibit each other in birds. In mammals, the superior colliculus receives inhibitory γ‐aminobutyric acid (GABA)ergic input from the basal ganglia via both the ipsilateral and the contralateral substantia nigra pars reticulata (SNr). This contralateral SNr projection is important in intertectal inhibition. Because the basal ganglia are evolutionarily conserved, the tectal projections of the SNr may show a similar pattern in birds. Therefore, the SNr could be a relay station in an indirect tecto–tectal pathway constituting the neuronal substrate for the tecto–tectal inhibition. To test this hypothesis, we performed bilateral anterograde and retrograde tectal tracing combined with GABA immunohistochemistry in pigeons. Suprisingly, the SNr has only ipsilateral projections to the optic tectum, and these are non‐GABAergic. Inhibitory GABAergic input to the contralateral optic tectum arises instead from a nearby tegmental region that receives input from the ipsilateral optic tectum. Thus, a disynaptic pathway exists that possibly constitutes the anatomical substrate for the inhibitory tecto–tectal interaction. This pathway likely plays an important role in attentional switches between the laterally placed eyes of birds. J. Comp. Neurol. 524:2886–2913, 2016. © 2016 Wiley Periodicals, Inc.     

Postsynaptic potentials in neurons of the pigeon''s optic tectum in response to afferent stimulation from the retina and other visual structures: an intracellular study

The responses of the cells in the pigeon's optic tectum to electrical stimulation of the contralateral optic nerve, the ipsilateral visual Wulst and the opposite optic tectum were intracellularly recorded. Optic nerve or visual Wulst stimulation elicited 3 types of responses: (1) a pure EPSP which gave rise to one or two action potentials; (2) an EPSP which sometimes gave rise to a spike, followed by an IPSP; and (3) a pure IPSP. Opposite tectum stimulation evoked in the tectal cells either a pure IPSP or a pure EPSP. The mono- or polysynaptic nature of the pathways involved in the excitatory and inhibitory responses of the tectal cells was assessed by increasing the frequency of the optic nerve stimulation. At low stimulus rates (2-6 Hz), all the excitatory events showing latencies longer than 5 ms were blocked suggesting that they were polysynaptic. Excitatory events having latencies shorter than 5 ms were generally able to follow high rate frequencies of optic nerve stimulation (40, 50 or 90 Hz) and we considered them to be monosynaptic. All but 3 IPSPs evoked by optic nerve stimulation, were blocked by stimulus rates beyond 5 Hz. Thus, although most IPSPs are generated through polysynaptic paths, direct retino-tectal inhibitory paths may also exist. The latency of the responses of individual cells to optic nerve, visual Wulst and opposite tectum stimulation show that the polysynaptic IPSPs to optic nerve stimulation did not involve relays in the visual Wulst or the opposite tectum.     

Tectal efferents in the blind cave fish Astyanax hubbsi.

Suction lesions were placed in the optic tecta of 36 blind cave fish. Three main bundles of tectal efferents were observed. A large, caudally directed fascicle distributes to the ipsilateral torus semicircularis, nucleus isthmi, and lateral tegmental areas of the mesencephalon and pons via the ipsilateral tectobulbar tract. Contralaterally, this fascicle descends to pontine levels as the contralateral tectobulbar tract. A second, rostrally directed bundle exits from the tectum at two levels. A small fascicle leaves from the caudal tectum and ascends rostrally as the commissura transversa. This bundle then joins with more rostrally exiting fibers and the combined fascicles collect in the area of the medial optic tract. They remain in this position until the level of the postoptic commissure where they decussate. Subsequently, this bundle moves caudally and enters the contralateral tectum at its most rostral extreme. The third bundle of tectofugal efferents leaves the tectum medially, at the level of the lesion, and enters the tectal commissure, through which it is distributed to the ipsilateral torus longitudinalis and contralateral optic tectum.     

Tectal efferents in the banded water snake,Natrix sipedon

Visual information reaches the dorsal thalamus by two distinct routes in most reptiles. Retinal efferents terminate directly in the dorsal lateral geniculate nucleus (DLGN). Retinal information is also channeled indirectly through the tectum to nucleus rotundus. Retinal projections to DLGN and tectum are also well esablished in snakes, but the status of the tecto-rotundal link of the indirect visual pathway is uncertain. Thus, tectal efferents were studied with Fink-Heimer methods in banded water snakes (Natrix sipedon). The tectum gives rise to crossed and uncrossed projections to the brainstem reticular formation. Commissural connections are effected with the contralateral tectum via the tectal and osterior commissures. tectum projects densely to the ipsilateral basal optic nucleus. Bilateral ascending projections reach the pretectal area, nucleus lentiformis mesencephali, lateral habenular nuclei, and posterodorsal nuclei. Ascending projections reach the ventral lateral geniculate and suprapeduncular nuclei. there is a diffuse projection to the central part of the caudal thalamus and a dense, bilaternal projection to the DLGN. These results indicate that the relation of the tectum to the dorsal thalamus is different in snakes than in other reptiles. Nucleus rotundus is either absent or poorly differentiated and there is a strong convergence of the direct and indirect visual pathways at DLGN.     

Bilaterally projecting neurons in the two visual pathways of chicks

By means of a double-labeling technique, we have investigated the organization of the bilateral thalamo-Wulst and tecto-rotundal projections in 2-day old chicks. After injecting fluorogold (FG) into one side of the visual Wulst and rhodamine B isothiocyanate (RITC) into the other side of the visual Wulst, the labeled neurons in the nucleus geniculatus lateralis pars dorsalis (GLd) were examined. Although the distribution areas of ipsilaterally and contralaterally labeled neurons overlap partly, very few double-labeled neurons were found (only 0.01% double-labeled neurons). This suggests that the ipsilateral and contralateral projections to the Wulst come from different neuronal populations of the thalamus. The FG and RITC were also injected into the rotundal nuclei (Rt) on each side of the thalamus and the labeled neurons in the optic tectum (TeO) were examined. In the TeO, the distribution areas of the neurons labeled ipsilaterally and contralaterally to Rt overlap completely and we found that up to 45% of the tectal cells were double-labeled by both FG and RITC. Therefore, many tectal neurons have axon collaterals so that they project to the Rt on both sides of the thalamus and must send information simultaneously to both sides of the brain. The differences in the structural organization of the two visual pathways are discussed with reference to the transmission of information to higher centers on both sides of the brain.     

Retinal projections in the cane toad, Bufo marinus.

The location and extent of retinorecipient areas in the cane toad, Bufo marinus, were established by anterograde transport of cobaltic-lysine complex from the cut optic nerve. Most of the labeled optic axons travelled in the marginal optic tract, while others were in the axial optic tract, and/or the basal optic tract. Retinal projections terminated in both contralateral and ipsilateral targets. In addition to the optic tectum, the main visual center, retinorecipient areas included the suprachiasmatic nucleus, rostral visual nucleus, neuropil of Bellonci, corpus geniculatum thalamicum, ventrolateral thalamic nucleus (dorsal part), posterior thalamic neuropil, uncinate neuropil, pretectal nucleus lentiformis mesencephali and basal optic nucleus. While all of these retinorecipient areas receive optic fibers from both eyes, the ipsilateral retinal projections were observed to be generally sparser than those from the contralateral retina. A sparse optic fiber projection covers the surface of the ipsilateral optic tectum and is most prominent rostromedially and caudolaterally. The position and the extent of each of the retinorecipient areas were determined in relation to a three-dimensional coordinate system. Morphometric analysis showed that 85.3% of the retinorecipient area is in the contralateral optic tectum, 10.4% in contralateral non-tectal areas, 1.6% in the ipsilateral optic tectum and 2.7% in ipsilateral non-tectal areas. The presence of an ipsilateral tectal projection and the well defined pretectal visual neuropil complex may be related to the highly developed visual behavior and visual acuity of Bufo marinus.     

Differential sensitivities of the two visual pathways of the chick to labelling by fluorescent retrograde tracers.

This study investigates the neurone structure-specific differences of sensitivities of fluorescent tracers. The tracers were used for retrograde labelling of contralateral projections in the two visual pathways of the chick. Rhodamine B Isothiocyanate (RITC), Fluorogold (FG) and True blue (TB) were injected into either the visual Wulst (thalamofugal pathway) or the nucleus rotundus (Rt; tectofugal pathway) and the retrogradely labelled neurones in the nucleus geniculatus lateralis pars dorsalis (GLd) or the optic tectum, respectively, were counted. Differential retrograde labelling in the two pathways was observed. In the thalamofugal pathway, both the contralateral and ipsilateral GLd cells were labelled by all three tracers (RITC, FG and TB). However, in the tectofugal pathway, whereas RITC labelled both the ipsilateral and contralateral tectal neurones, FG or TB labelled effectively only the ipsilateral tectal neurones. It was clear that FG and TB were taken up by the nerve endings and transported part-way along the axon but failed to be transported to the cell bodies of the contralateral tectal neurones. In addition, red beads and green beads were also injected into Rt and the differential labelling was also observed. Red beads labelled both ipsilateral and contralateral tectal neurones but green beads labelled only the ipsilateral tectal neurones. Since the contralateral tectal projections consist of divergent axon collaterals, the present study suggests that various retrograde tracers are not transported in these axon collaterals to label cell bodies. The contralaterally projecting neurones in the thalamofugal pathway are not axon collaterals and they were labelled by all of the tracers used.     

Topographic projections between the nucleus isthmi and the optic tectum in a teleost,Navodon Modestus

Following horseradish peroxidase injections into the optic tectum of a teleost,Navodon modestus, reciprocal and topographic projections between the nucleus isthmi and the ipsilateral optic tectum were determined. The isthmo-tectal fibers diverge to the optic tectum while maintaining the spatial arrangements of the isthmic cells from which the fibers originate. The tecto-isthmic projections also keep the spatial arrangements in the optic tectum. The tectal fibers converge near the nucleus isthmi and terminate in the non-cellular portion of the nucleus. The reciprocal topography is apparent in the combined results of 9 experiments with one tectal injection in each region. No labeled cells and fibers were found in the contralateral nucleus isthmi.     

Responses of the optic tectum to telencephalic stimulation in catfish

In order to test physiologically for cerebrotectal connections in a fish, averaged evoked potentials and unit responses were recorded from the optic tectum following electrical stimulation applied to the telencephalon in the siluroid teleost Ictalurus nebulosus. A single shock applied to the area dorsalis centralis (Dc) of the telencephalon, and only to this area, elicits a sequence of deflections in the ipsilateral optic tectum: an initial negative peak at about 8 ms, (= N8), a larger N25 and a slow P50-N95. The configurations, depth profiles, latencies and susceptibility to repetitive stimulation, together with the known tectal anatomy, suggest that the first wave is due to the afferent fibers from the telencephalon and that N25 is due to deep tectal neurons. Telencephalic input exerts a conditioning effect on the field potentials and unit responses evoked by direct optic nerve shock. Such a shock elicits, in the contralateral tectum, small negative, optic tract axon peaks followed by a large N6, believed to be postsynaptic, and a still later P12. As a first approximation it is argued that the telencephalic input and the retinal input are activating different sets of neuronal elements in the optic tectum, since the configuration and depth profile of the telencephalic and optic nerve shock-elicited potentials are different. A conditioning Dc stimulus has a long-lasting effect on the form of the optic nerve field potential, maximally when the pallial shock precedes the optic by about 90 ms. The effect, observed by subtracting the conditioned from the unconditioned tectal response to optic nerve shock, is a difference wave with N11 and P20. The unit activity from deep tectal laminae is either activated or accelerated following Dc stimulation, while superficially located neurons are not affected. In another group of tectal units, the optic nerve shock-induced response is depressed by a preceding pallial dorsalis centralis stimulus. The evidence is compatible with the assumption of direct projections from Dc to the deep layers of the tectum, but the timing could also permit indirect pathways. In any case, the influence is not simple or identical for different tectal cell classes.     

Projections of the optic tectum in the longnose gar,lepisosteus osseus

Efferent projections of the optic tectum were studied with the anterograde degeneration method in the longnose gar. Ascending projections were found bilaterally to 3 pretectal nuclei — the superficial pretectal nucleus, nucleus pretectalis centralis and nucleus pretectalis profundus — and to a number of targets which lie further rostrally — the central posterior nucleus, dorsal posterior nucleus, accessory optic nucleus, nucleus ventralis lateralis, nucleus of the ventral optic tract, rostral part of the preglomerular complex, suprachiasmatic nucleus, anterior thalamic nucleus, nucleus ventralis medialis, nucleus intermedius, nucleus prethalamicus and rostral entopeduncular nucleus. Projections of the tectum reach the contralateral side via the supraoptic decussation and are less dense contralaterally than ipsilaterally. Descending projections resulting from tectal lesions include: (1) a tectal commissural pathway to the core of the torus longitudinalis bilaterally and the contralateral tectum and torus semicircularis; and (2) a pathway leaving the tectum laterally from which fibers terminate in the ipsilateral torus semicircularis, an area lateral to the nucleus of the medial longitudinal fasciculus, lateral tegmental nucleus, nucleus lateralis valvulae, nucleus isthmi and the reticular formation. A component of this bundle decussates at the level of the lateral tegmental nucleus to project to the contralateral reticular formation.

On the basis of comparisons of these findings with the pattern of retinal projections in gars and other data, it is argued that the nuclei previously called the lateral geniculate and rotundus in fish are not the homologues of the nuclei of those names in land vertebrates but are rather pretectal cell groups. The overall organization of both retinal and tectal projections in gars is strikingly similar to that in land vertebrates; at present, the best candidate for a rotundal homologue is the dorsal posterior nucleus.     

Extratelencephalic projections of the avian visual Wulst. A quantitative autoradiographic study in the pigeon Columbia livia

The efferent projections of the pigeon visual Wulst upon the diencephalon and mesencephalon were investigated using the autoradiographic technique following the combined injection of [3H] proline and [3H] leucine into the rostral hyperstriatum accessorium. Repeated measures of silver grain densities were performed bilaterally in different brain structures using a computer-assisted system of image analysis. The density values were compared (Mann-Whitney U-Test) with those recorded in three homolateral control structures (tractus opticus, n. rotundus, n. pretectalis principalis) and in corresponding contralateral areas and nuclei. The data showed ipsilateral projections from the visual Wulst and via the tractus septomesencephalicus upon the dorsal thalamus (n.: dorsolateralis anterior superficialis parvocellularis), ventral thalamus (n.: intercalatus, ventrolateralis, geniculatus lateralis pars ventralis--GLv), pretectum (n.: superficialis synencephali, geniculatus pretectalis, griseus tectalis, pretectalis: diffusus, pars lateralis and pars medialis, area pretectalis) as well as to the nucleus of the basal optic root, n. spiriformis medialis and optic tectum (layer 2-4, 6, 7, 12 and 13). Crossed projections were observed to pass through the supraoptic decussation and the posterior commissure, however only the contralateral n. GLv was found to be significantly labeled. Interspecies variations in the organization of descending visual Wulst projections, related to the terminal distribution and relative size of the crossed components may be linked to differences in the degree of overlap of the binocular fields. Correspondingly, this may reflect the degree of bilateralization upon the Wulst of direct input from the visual thalamus.     

Connections of eye-saccade-related areas within mesencephalic reticular formation with the optic tectum in goldfish

Physiological studies demonstrate that separate sites within the mesencephalic reticular formation (MRF) can evoke eye saccades with different preferred directions. Furthermore, anatomical research suggests that a tectoreticulotectal circuit organized in accordance with the tectal eye movement map is present. However, whether the reticulotectal projection shifts with the gaze map present in the MRF is unknown. We explored this question in goldfish, by injecting biotin dextran amine within MRF sites that evoked upward, downward, oblique, and horizontal eye saccades. Then, we analyzed the labeling in the optic tectum. The main findings can be summarized as follows. 1) The MRF and the optic tectum were connected by separate axons of the tectobulbar tract. 2) The MRF was reciprocally connected mainly with the ipsilateral tectal lobe, but also with the contralateral one. 3) The MRF received projections chiefly from neurons located within intermediate and deep tectal layers. In addition, the MRF projections terminated primarily within the intermediate tectal layer. 4) The distribution of labeled neurons in the tectum shifted with the different MRF sites in a manner consistent with the tectal motor map. The area containing these cells was targeted by a high-density reticulotectal projection. In addition to this high-density topographic projection, there was a low-density one spread throughout the tectum. 5) Occasionally, boutons were observed adjacent to tectal labeled neurons. We conclude that the organization of the reticulotectal circuit is consistent with the functional topography of the MRF and that the MRF participates in a tectoreticulotectal feedback circuit.     

Connections of the tectum opticum in two urodeles, Salamandra salamandra and Bolitoglossa subpalmata, with special reference to the nucleus isthmi

Tectal connections were studied in two urodele species following horseradish peroxidase injections into the tectum opticum. In both species retrogradely labelled cells were observed: ipsilaterally in the corpus striatum, lateral amygdala, ventral and dorsal thalamus and nucleus of DARKSCHEWITSCH--bilaterally in the pretectal nucleus, dorsal tegmentum and nucleus reticularis medius--contralaterally in the tectum opticum and area octavo lateralis. Besides these nuclei the nucleus isthmi was bilaterally labelled. Rostral efferent projections of the tectum opticum terminated in the ipsilateral pretectal area and the ipsilateral dorsal and ventral thalamus ipsilaterally coursing to the contralateral tectum via the commissura postoptica. Caudal efferents formed the bilaterally organized tecto-bulbar tracts innervating the rhombencephalon. Comparison of the results of a series of tectal horseradish peroxidase injections differing in depth, tangential extension and location, indicated that tectal afferents from the telencephalon, the contralateral tectum opticum and the medulla were sparse and widely branching. Projections of the telencephalon and all diencephalic nuclei terminated deep in the rostral tectum opticum. Projections of the medulla terminated preferentially deep in the caudal tectum opticum. The tecto-isthmic projection was highly topographic forming a layered terminal field lateral to the nucleus isthmi. The isthmo-tectal projection innervated the whole tectum opticum on the ipsilateral side and was highly topographic. On the contralateral side the caudal part of the tectum opticum was not innervated. The isthmo-tectal fibers terminated superficially in the tectum opticum on both sides of the brain. The nucleus isthmi identified here is proposed to be homologe to that of other vertebrates.     

Tectal projections of an infrared sensitive snake,Crotalus viridis

Crotaline snakes have detectors for infrared radiation and this information is projected to the optic tectum in a spatiotopic manner. The tectal projections were examined in Crotalus viridis with the use of silver methods for degenerating fibers and the autoradiographic and horseradish peroxidase tracing methods. Large lesions included all of the tectal layers but not the underlying structures. Projections to the thalamus include a sparse input to the ipsilateral ventral and dorsal lateral geniculate nuclei, the ventromedial nucleus, and nucleus lentiformis thalami. Nucleus rotundus was not detected. The projections to the pretectal nuclei are primarily ipsilateral to the nucleus lentiformis mesencehali and pretectal nucleus. At the level of the mesencephalon, tectal efferents are bilateral to nucleus profundus mesencephali and the tegmentum. There is minimal input to the contralateral deep tectal layers. There are ispilateral terminations in a nucleus identified as the posterolateral tegmental nucleus. Descending fibers include the two major tracts—the ventral tectobulbar tract that terminates in the ipsilateral lateral reticular formation and the predorsal bundle that distributes throughout the contralateral medial reticular formation. Two small descending tracts were noted—the intermediate and dorsal tectobulbar tracts. All of these descending tracts appear to terminate by the time they reach the caudal medulla. After superficial lesions terminals could be found in the ventral lateral geniculate nucleus, the nucleus profundus mesencephali, and the posterolateral tegmental nucleus; the two major descending tracts contained degenerated fibers as well. The areas receiving tectal input in Crotalus were compared to those of other reptiles and discussed.     

Thalamo-tectal projections in the frog

The optic tectum of the leopard frog was examined for evidence of terminal degeneration following lesions in forebrain structures. No evidence was found of a telencephalic tectal projection. Degeneration following lesions in forebrain sites which give rise to tectal afferents are located in the dorsal thalamus, specifically the corpus geniculatum laterale in the rostral thalamus and the nucleus posterolateralis in the caudal thalamus. Corpus geniculatum projects to the ipsilateral tectum where it distributes in both the superficial and deep laminae over the rostromedial one-third of that structure. The rostral portion of the nucleus posterolateralis projects rather sparsely to the ipsilateral tectum. In contrast, the posterior portion of the posterolateral nucleus projects to both the superficial and deep laminae of the rostral two-thirds of the ipsi- and contralateral tectum. The significance of these and reciprocal tectothalamic projections for behavioral expression is considered.     

Organization of the visual system in larval lampreys: an HRP study.

The organization of the visual system of larval lampreys was studied by anterograde and retrograde transport of HRP injected into the eye. The retinofugal system has two different patterns of organization during the larval period. In small larvae (less than 60-70 mm in length) only a single contralateral tract, the axial optic tract, is differentiated. This tract projects to regions in the diencephalon, pretectum, and mesencephalic tegmentum. In larvae longer than 70-80 mm, there is an additional contralateral tract, the lateral optic tract, which extends to the whole tectal surface. In addition, ipsilateral retinal fibers are found in both small and large larvae. Initially, the ipsilateral projection is restricted to the thalamus-pretectum, but it reaches the optic tectum in late larvae. Changes in the organization of the optic tracts coincide with the formation of the late-developing retina and consequently, the origin of the optic tracts can be related to specific retinal regions. The retinopetal system is well developed in all larvae. Most retinopetal neurons are labeled contralaterally and are located in the M2-M5 nucleus of the mesencephalic tegmentum, in the caudolateral mesencephalic reticular area and adjacent ventrolateral portions of the optic tectum. Dendrites of these cells are apparent, especially those directed dorsally, which in large larvae extend to the optic tectum overlapping with the retino-tectal projection. These results indicate that in lampreys, visual projections organize mainly during the blind larval period before the metamorphosis, their development being largely independent of visual function.     

Observations on the projections and intrinsic organization of the pigeon optic tectum: an autoradiographic study based on anterograde and retrograde, axonal and dendritic flow.

Three aspects of the labelling pattern seen after the injection of 13 different radioactive amino acids into the pigeon optic tectum have been described: The efferent projections of the optic tectum; the specific labelling of two pathways; and the dendritic organisation of tectal layer III neurons based on the retrograde and anterograde movement of label within these dendrites. Discrete injections of tritiated amino acid that involved all or only the superficial tectal layers suggested that layer III gave rise to the massive non-topographically organised and bilateral projections (fibers crossing within the decussato supraoptica ventrlis) upon the nuclei rotundus, subpraetectalis and interstitio-praetecto-subpraetectalis and to the ipsilaterally directed pathways terminating within the nuclei praetectalis, triangularis, subrotundus, dorsolateralis anterior thalami, posteroventralis and ventrolateralis thalami. Layer III neurons may also be the source of efferents to the posterior dorsolateral thalamus (the layer III pathway), the pontine grey and, bilaterally to the reticular formation and of the layer IV or tectal commisural pathway terminating within the contralateral tectal cortex. In contrast projections originating from layer II were generally topographically organised and terminated either within certain of the isthmic nuclei (n. isthmi pars parvocellularis, n. isthmo-opticus and n. semilunaris) or ran within layer I (layer I pathways) to end in the pretectum (griseum tectale) and ventral thalamus (n. ventrolateralis thalami, n. geniculatus, pars ventralis). A small projection from layer II upon the ipsilateral nucleus rotundus may also be present. Triated serine and tyrosine were found to be particularly effective in labeling perikarya as well as axons and terminals. The layer I pathway could be selectively labelled after tectal injections of 3H-GABA while the cell bodies of Ipc neurons were labelled in a retrograde fashion after tectal injections of 3H-glycine, serine or alanine. Intrinsic tectal labelling was found by correlation with Golgi material to reflect both anterograde and retrograde transport of label within dendrites of layer III cells. Anterograde movement of label indicated that the terminal portions of layer III cell dendrites ended in an orderly radial arrangement within sublayers IIb and IId, while the retrograde movement of label resulted in the labelin of layer III perikarya outside the injection field.     

Retinal projections in larval,transforming and adult sea lamprey,Petromyzon marinus

Unilateral enucleations were performed on larval, transforming and adult sea lampreys. Following 5 to 11 days survival, the animals were sacrificed and the brains were processed using a modified Fink-Heimer technique. In larvae, contralateral optic projections were found to the posterior one-third of the dorsal thalamus, the pretectum, and the optic tectum. No ipsilateral projections were present in the larvae. In enucleated transforming and adult lampreys, degenerating axons were observed in the optic chiasm and bilaterally in the optic tracts. Retinal efferents projected bilaterally to a lateral neuropil region (“tractus opticus”) in the posterior one-half of the dorsal thalamus. Contralaterally, a conspicuous dorsomedial cell group (lateral geniculate nucleus) also received a projection. Contralateral projections to the superficial layers of the pretectum and optic tectum were observed. Ipsilateral retinal projections to the pretectum and optic tectum in transforming and adult lampreys were restricted to a small zone at the ventrolateral margins of the pretectum and tectum. The changes in distribution of retinofugal projections during transformation appear to be occurring at the same time that the eye differentiates into its adult form.