Right compound eyes were formed in Xenopus embryos at stages 32–33 by the fusion of two nasal (NN), two ventral (VV) or two temporal (TT) halves. Shortly after metamorphosis the optic nerve from the compound eye was sectioned and the left intact eye removed. The retinotectal projections from the compound eye to the contralateral and ipsilateral tecta were studied by [3H]proline autoradiography and electrophysiological mapping between 6 weeks and 5 months after the postmetamorphic surgery. The results showed that NN and VV eyes projected to the entire extent of both tecta. In contrast, optic fibre projection from TT eyes, although more extensive than the normal temporal hemiretinal projection, failed to cover the caudomedial portion of the tecta. The visuotectal projections in all three combinations corresponded to typical reduplicated maps to be expected from such compound eyes, where each of the hemiretinae projected across the contralateral and ipsilateral tecta in an overlapping fashion. The rapid expansion of the hemiretinal projections of the compound eyes in the ipsilateral tectum following the removal of the resident optic fibre projection suggests that tectal markers may be carried and deployed by the incoming optic fibres themselves.

In Xenopus the entire retina projects topographically to the whole extent of the contralateral tectum (Gaze, 1958). Considerable work has been devoted to the understanding of how the orderly connexions between the axonal processes of the retinal ganglion cells and tectal neurons are formed during development and are reformed following optic nerve regeneration in adult animals. The most comprehensive hypothesis, advanced by Sperry (1951, 1963) suggests that the establishment of the topographic retinotectal connexions is based on selective cytochemical affinities between the axons of retinal ganglion cells and the corresponding tectal neurons; it is presumed that these positional specificities are generated independently from one another during embryogenesis.

Experimental evidence indeed supports the notion that individual ganglion cells differ from one another on the basis of their position in the retina and that these positional differences are expressed with reference to the tectal termination of their optic axons (Gaze, 1978). The polarization of the retina in relation to the future termination of the optic fibres has been shown to occur during early embryogenesis (Gaze, Feldman, Cook & Chung, 1979; Sharma & Hollyfield, 1980). The time at which similar tectal polarization occurs is not known. Experiments involving tectal rotation in metamorphic frogs have demonstrated the existence of tectal positional markers serving as targets for ingrowing optic axons (Levine & Jacobson, 1974). Since these studies were carried out in juvenile animals, where the retinotectal connexions are already formed, they failed to furnish information about the nature and the mode of generation of tectal positional markers.

Recent experiments in goldfish involving optic nerve regeneration from half retinae suggest that positional markers may be induced by the ingrowing optic fibres themselves, imprinting their position-related retinal label on the tectal cells (Schmidt, 1978; Schmidt, Cicerone & Easter, 1978). Consequently, an abnormal optic fibre input either during development or in regeneration could alter the tectal markers by generating a set appropriate to the nature of the incoming optic fibres.

The retinotectal projections from compound eyes (eyes made by the fusion of two nasal (NN), two temporal (TT) or two ventral (VV) halves) are abnormal in that each similar retinal half spreads across the whole extent of the tectum instead of being restricted to the corresponding part (Gaze, Jacobson & Székely, 1963; Straznicky, Gaze & Keating, 1974). Recent observations have demonstrated that the polarity of the retinal developmental programme is kept unaltered in fused eye fragments (Straznicky & Gaze, 1980; Gaze & Straznicky, 1980a). Further evidence for the stability of the retinal positional markers is derived from studying the patterns of optic fibre regeneration from compound eyes. The optic fibre regeneration from such eyes to the contralateral and ipsilateral tecta are strikingly different (Fig. 1). The compound-eye projection to the ipsilateral tectum (also innervated by the normal eye) is highly selective in that it is restricted to one half, corresponding to the type of the compound eye (Tay & Straznicky, 1978; Gaze & Straznicky, 1979, 19806). The projection from the same eye extends across the entire contralateral tectum as it did before the nerve was cut. The difference in the optic fibre distributions in the two tecta suggests that the compound eye possesses retinal positional markers characteristic of the two (similar) halves. It is also apparent from these studies that in the ipsilateral tectum the normal retinotopic ordering of fibre projections from both eyes is maintained in contrast to the contralateral tectum where it is not. It is thus conceivable that the topographic ordering of the incoming fibres on the tectum ipsilateral to the compound eye may be the result of their interaction with the resident fibres from the contralateral eye. If the basis of selective regeneration from a compound eye to the ipsilateral tectum is fibre–fibre recognition, one does not have to assume the existence of tectal markers independent from the optic fibre projection.

Fig. 1.

Drawings of the pathways and terminations of regenerating optic fibres from compound eyes to the contralateral and ipsilateral tecta following 3HP injection into the compound eye. Note that in contrast to the complete projections to the contralateral tectum only about half of the ipsilateral tectum is innervated by optic fibres from a TT (A), NN (B) and VV (C) eye.

Fig. 1.

Drawings of the pathways and terminations of regenerating optic fibres from compound eyes to the contralateral and ipsilateral tecta following 3HP injection into the compound eye. Note that in contrast to the complete projections to the contralateral tectum only about half of the ipsilateral tectum is innervated by optic fibres from a TT (A), NN (B) and VV (C) eye.

The experiments reported here attempt to test the validity of such a mechanism in animals with a right compound eye, where the left normal eye was removed 2 weeks after metamorphosis with the simultaneous sectioning of the optic nerve from the compound eye. The resultant spreading of the hemiretinal projections of the compound eye in the ipsilateral tectum supports the idea that positional markers may be carried and deployed in the tectum by the incoming optic fibres themselves.

Laboratory bred Xenopus laevis were used in this study.

Surgery

Eye operations

The right eyes of stages -32 to -33 embryos (Nieuwkoop & Faber, 1956) were operated on in full-strength Niu-Twitty solution to obtain double-nasal (KN), double-temporal (TT) or double-ventral (VV) eyes. Embryos were anaesthetized with 0·001 % solution of MS 222 (Tricaine Methanesulphonate, Sandoz) and the temporal half of the eye anlage was removed by suction and substituted by a left nasal half from another embryo to form an NN eye. Care was taken that neither the host’s nor the donor’s eye fragments included parts from the temporal half. Similarly TT and VV eyes were formed in other embryos. Twenty-four hours after the microsurgery, animals were checked and the unsuccessful cases as well as animals with grossly asymmetric compound eyes (60 % or more for one half) were excluded from further experiments. The NN, and TT and VV eye animals were kept separately and reared to metamorphosis and beyond.

Optic nerve transection and enucleation

Two weeks after metamorphosis animals were anaesthetized by immersion in a 0·1 %solution of MS 222. The optic nerve from the compound eye was approached through the roof of the mouth and cut 0·5 mm rostral to the chiasma. The proximal and distal stumps of the nerve were approximated to facilitate regeneration. Following the optic nerve transection, the left normal eye was enucleated. Autoradiographic and electrophysiological studies were performed between 42 and 150 days after optic nerve section and enucleation.

Electrophysiology

Animals of 60 days or longer postoperative survival were anaesthetized by intraperitoneal injection of 0·1 % aqueous solution of MS 222 and immobilized with 0·05 mg tubocurarine (10 mg/ml, Wellcome) administered intramuscularly. The tectum was exposed by opening the skull and the dorsal surface photographed at 50 × magnification. The animal was then set up with right eye centred on the optic axis of the perimeter for visuotectal recording. The mapping of visuotectal projections was similar to that previously described for adult Xenopus (Straznicky, Gaze & Keating, 1971). Single or multiunit action potentials were recorded from terminal arborizations of optic fibres at predetermined tectal positions, and the corresponding receptive fields in the visual field were mapped. Visuotectal maps were based on about 30–40 different tectal electrode positions in each aimai. After the visuotectal mapping the brains of animals were processed for autoradiography.

Histology

(i) Autoradiography

Twenty-four hours before recording and sacrifice 10 μCi L-5-[3H]proline (3HP) (Amersham, specific activity 21 Ci per m-mole) was injected into the compound eye. The head of the animal was fixed in Bouins’ solution, the dissected brain embedded in paraffin, serially sectioned at 10 μm in the transverse (coronal) plane and mounted on slides. The closely spaced, deparaffinized sections were coated with K2 emulsion (Ilford) exposed in light-tight boxes at 4 °C for 14 days, developed in Kodak D19 developer and counterstained with Harris’s haematoxylin.

(ii) Analysis of autoradiograms

In selected animals, every tenth serial section of the brain was drawn using a microscope camera-lucida attachment to record the size of the tecta and the distribution of 3HP-labelled axons in the contralateral and ipsilateral tecta. The extent of tectum covered with silver grains and the total tectal surface area were measured by a Hewlett & Packard digitizer (987A) and the two measurements were correlated. The autoradiographic grain density was established along the mediolateral extent of the optic-fibre-receiving layer of the tectum by measuring the reflected incident light over an oblong area of 20 × 60 μm (Straznicky, Gaze & Keating, 1979b). By moving the oblong medio-laterally in steps of 50 μm the grain density profile of the compound-eye projection was established. About 25 measurements were made on each of the three subsequent sections of the selected tectal area. Background activity was measured on the same section over non-visual brain tissue.

The autoradiographic visualization of the VV eye projection has been described previously (Straznicky, Gaze & Horder, 1979 a); however, no anatomical description on NN and TT eye projections has been published, therefore a short account of the autoradiographic optic fibre projections from NN and TT eyes is first given here to serve as reference to retinotectal projections to be described in this paper.

(A) Retinotectal projections from NN and TT eyes (optic nerve intact)

Three animals of each combination were injected with 3HP 5 months after metamorphosis. The autoradiographic silver grains filled the superficial layer evenly across the rostrocaudal and mediolateral extents of the tectum in NN eye animals (Fig. 2 A), as in the normal retinal projection. Unexpectedly, in none of the three TT eye animals did the silver grains cover the whole tectal surface; a wedge-shaped area in the caudomedial tectum was systematically devoid of silver grains (Fig. 2B). Graphical reconstructions and planimetric measurements (Fig. 2C) indicated a consistent 10-13 % deficit of TT-eye projection in the caudomedial tectum in animals without optic nerve section (Table 4).

Fig. 2.

Bright-field 3HP autoradiographs of contralateral retinotectal projections from a NN (A) and a TT (B) eye 5 months after metamorphosis and graphical reconstruction of the extent of the TT eye projection from serial sections (C). The plane of section is transverse; right is to the left, dorsal at the top. Note in (B) that the retinotectal projection from a TT eye does not extend to the midline. Stippled area in (C) indicates caudomedial tectum which is not innervated. Arrow points rostrally on the tectum. Scale 300 μm applies to (A), (B) and (C).

Fig. 2.

Bright-field 3HP autoradiographs of contralateral retinotectal projections from a NN (A) and a TT (B) eye 5 months after metamorphosis and graphical reconstruction of the extent of the TT eye projection from serial sections (C). The plane of section is transverse; right is to the left, dorsal at the top. Note in (B) that the retinotectal projection from a TT eye does not extend to the midline. Stippled area in (C) indicates caudomedial tectum which is not innervated. Arrow points rostrally on the tectum. Scale 300 μm applies to (A), (B) and (C).

(B) Retinotectal projections from compound eyes following optic-nerve section and left eye enucleation

NN eye projections

Ten animals were included in this group (Table 1). 3HP autoradiography demonstrated that both the contralateral and ipsilateral tecta were fully covered with silver grains, suggestive of a continuous retinotectal projection from the NN eye across the whole extent of both tecta. In five animals slightly lower grain density was measured in the rostral than in the caudal tectum (Fig. 3). Since 3HP autoradiography does not distinguish between fibres of passage and fibre termination it could not be determined whether or not the measured lower grain density corresponded to sparser innervation in the rostrolateral tectum. Silver grains were evenly distributed over the medial and lateral branches of the optic tract in the rostral pole of the tectum, indicating that regenerated optic fibres entered into the tectum via both these branches. Seven animals were recorded electrophysiologically 72–137 days after optic nerve section. The contralateral and ipsilateral visuotectal projections were typical NN maps in five animals (Fig. 4). The centre of the visual field projected to the rostral part, and the nasal and temporal poles to the caudal part of both tecta. In one animal an apparent projection deficit from the central visual field to the rostral part of the ipsilateral tectum was observed (Fig. 5) despite the presence of autoradiographic silver grains in this sector. In the seventh animal mapped in this group, only few points were recorded from the ipsilateral tectum, whilst the contralateral tectum had a typical NN map.

Table 1.

Retinotectal and visuotectalprojections to the contralateral and ipsilateral tecta in animals with right NN eye

Retinotectal and visuotectalprojections to the contralateral and ipsilateral tecta in animals with right NN eye
Retinotectal and visuotectalprojections to the contralateral and ipsilateral tecta in animals with right NN eye
Fig. 3.

Grain density counts over the optic-fibre-receiving layer of the rostral and caudal (stippled) parts of the right and left tecta in animal NN4: background activity is indicated by a broken line. The units of the ordinate represent number of grains over 1200 μm2area. Note lower grain counts rostrally in both tecta. Each segment of the histogram represents 140 μm. M = medial, L = lateral.

Fig. 3.

Grain density counts over the optic-fibre-receiving layer of the rostral and caudal (stippled) parts of the right and left tecta in animal NN4: background activity is indicated by a broken line. The units of the ordinate represent number of grains over 1200 μm2area. Note lower grain counts rostrally in both tecta. Each segment of the histogram represents 140 μm. M = medial, L = lateral.

Fig. 4.

Visual-field projections from the right NN eye to the contralateral and ipsilateral tecta in animal NN6. For this and all subsequent maps the conventions are the following. Right ipsilateral tectum is on the left and the left contralateral tectum is on the right. On the tectal diagram the small filled arrow points rostrally. The visual-field charts are centred on the optic axis of the eye extending out 100° radially. D, dorsal; V, ventral; T, temporal and N, nasal. The dorsal surface of the tecta shows rows of electrode positions, the corresponding rows of stimulus positions in the visual field are indicated with the same number. Open circles indicate electrode positions from where no visual responses were obtained. The large open arrows indicate the orientation of the nasotemporal axis of the visual field in relation to the rostrocaudal tectal axis. Note that each of the halves of the visual field projects to the entire contralateral and ipsilateral tectum.

Fig. 4.

Visual-field projections from the right NN eye to the contralateral and ipsilateral tecta in animal NN6. For this and all subsequent maps the conventions are the following. Right ipsilateral tectum is on the left and the left contralateral tectum is on the right. On the tectal diagram the small filled arrow points rostrally. The visual-field charts are centred on the optic axis of the eye extending out 100° radially. D, dorsal; V, ventral; T, temporal and N, nasal. The dorsal surface of the tecta shows rows of electrode positions, the corresponding rows of stimulus positions in the visual field are indicated with the same number. Open circles indicate electrode positions from where no visual responses were obtained. The large open arrows indicate the orientation of the nasotemporal axis of the visual field in relation to the rostrocaudal tectal axis. Note that each of the halves of the visual field projects to the entire contralateral and ipsilateral tectum.

Fig. 5.

Visual-field projections in animal NN3. Note that no visual responses were obtained from the rostral part of the ipsilateral tectum.

Fig. 5.

Visual-field projections in animal NN3. Note that no visual responses were obtained from the rostral part of the ipsilateral tectum.

VV eye projections

In all ten animals studied in this group (Table 2) the anatomical retinotectal projection spread across the whole rostrocaudal and mediolateral extents of the contralateral and ipsilateral tecta. In a few animals, however, the silver-grain density over the tectum decreased laterally, confirming our earlier report on VV eye projections during development and in regeneration (Straznicky et al. 1979,a). Silver grains were more or less evenly distributed over the medial and lateral branches of the optic tract in eight animals. In the remaining two animals optic fibres entered the tecta selectively via the medial branch. These observations indicate that during regeneration optic fibres of ventral retinal origin are not able to choose the appropriate medial branch for entry into the tectum without the presence of pre-existing fibre tracts. These findings provide indirect support for our previous suggestion that selective regeneration may be based on fibre-fibre recognition, recognizing the course of similar fibres in the tract and following them enroute to the tectum (Straznicky et al. 1979 a).

Table 2.

Retinotectal and visuotectal projections to the contralateral and ipsilateral tecta in animals with right VV eye

Retinotectal and visuotectal projections to the contralateral and ipsilateral tecta in animals with right VV eye
Retinotectal and visuotectal projections to the contralateral and ipsilateral tecta in animals with right VV eye

In all four animals recorded, both the contralateral and ipsilateral visuotectal projections were typical VV maps (Fig. 6).

Fig. 6.

Visual-field projections in animal VV8. The large open arrows indicate the orientation of the dorsoventral axis of the visual field in relation to the mediolateral tectal axis.

Fig. 6.

Visual-field projections in animal VV8. The large open arrows indicate the orientation of the dorsoventral axis of the visual field in relation to the mediolateral tectal axis.

TT eye projections

In contrast to the previous two groups 14 out of 15 animals with TT eyes showed a caudomedial deficiency in the autoradiographically demonstrable contralateral and ipsilateral retinotectal projections (Table 3).

Table 3.

Retinotectal and visuotectal projections to the contralateral and ipsilateral tecta in animals with right TT eye

Retinotectal and visuotectal projections to the contralateral and ipsilateral tecta in animals with right TT eye
Retinotectal and visuotectal projections to the contralateral and ipsilateral tecta in animals with right TT eye

The autoradiographic studies in these animals indicated that the caudomedial margins of both tecta were not innervated. The distribution of the silver grains in one representative animal, is shown in a pjioto-montage (Fig. 7). The rostral parts of both tecta are fully covered with silver grains, whereas, more caudally the silver grains are restricted to the lateral two thirds of the tecta. Although no strict boundary can be seen, a steep decrease of silver-grain density was established by grain density measurements (Fig. 8).

Fig. 7.

Bright-field photographs of autoradiographs of animal TT8. Rostral is on the bottom, caudal is at the top. Note the deficient innervation of the caudomedial tectum. Inset shows the extent of the TT-eye projection as seen from above and reconstructed from serial sections. Arrows point rostrally on the tectum. Scale 500 μm.

Fig. 7.

Bright-field photographs of autoradiographs of animal TT8. Rostral is on the bottom, caudal is at the top. Note the deficient innervation of the caudomedial tectum. Inset shows the extent of the TT-eye projection as seen from above and reconstructed from serial sections. Arrows point rostrally on the tectum. Scale 500 μm.

Fig. 8.

Silver-grain-density counts from the midtectal region of animal TT8 (A). Each segment on the histogram represents 140 μm. Broken line gives background activity over non-visual brain tissue. Arrows on the bright-field autoradiography of the corresponding tectal section marks the sites of grain density measurements (B). Scale 300 μm.

Fig. 8.

Silver-grain-density counts from the midtectal region of animal TT8 (A). Each segment on the histogram represents 140 μm. Broken line gives background activity over non-visual brain tissue. Arrows on the bright-field autoradiography of the corresponding tectal section marks the sites of grain density measurements (B). Scale 300 μm.

Reconstructions on the optic fibre projections from TT eyes revealed a consistent projection deficit varying between 7–32 % of the whole tectal area (Table 4). Although the number of animals is insufficient for more detailed analysis, the data show a tendency for the projection deficit to reduce with time after optic nerve section.

Table 4.

Retinotectal projection and tectal surface areas given in mm2in animals with right TT eye

Retinotectal projection and tectal surface areas given in mm2in animals with right TT eye
Retinotectal projection and tectal surface areas given in mm2in animals with right TT eye

In 11 out of 12 animals recorded, the contralateral and ipsilateral visuotectal maps correspond to typical TT eye projections, confirming that we had dealt with TT eye animals. In five animals with TT maps the caudomedial tecta (Fig. 9) failed to yield evoked potentials, substantiating the autoradiographic observations that this part of the tecta had not been innervated. In the other six animals the caudomedial deficiency of the TT eye projection was less extensive and consequently it was not demonstrable with a visuotectal recording. It is worth noting that animal TT3 had a normal visuotectal projection with accompanying complete autoradiographic retinotectal projections.

Fig. 9.

Visual-field projection in animal TT11. Broken line indicates the approximate extent of autoradiographic silver-grain deposition in the tecta.

Fig. 9.

Visual-field projection in animal TT11. Broken line indicates the approximate extent of autoradiographic silver-grain deposition in the tecta.

The aim of the present study was to determine whether regenerated optic fibres from the hemiretinae of a compound eye expand their assigned projection in the ipsilateral tectum, following the elimination of resident optic fibres of the normal eye. The compound-eye projection patterns were assessed anatomically and electrophysiologically and the two results were comparable. The main and consistent finding of this study was that each nasal, ventral and temporal half of the compound retina expanded its normal termination over the ipsilateral tectum.

Retinotectal-size-disparity experiments in adult fish and frogs indicate that the optic fibre projection is capable of reorganization in the form of compression of the entire retinal projection to a half tectum (Gaze & Sharma, 1970; Udin, 1977) and the expansion of the hemiretinal projection across the whole tectum (Straznicky, Tay & Lunam, 1978; Schmidt et al. 1978). Creating size disparity in the retinotectal relationship during early embryogenesis in animals with one compound eye have been reported to cause similar modification in visual projections (Gaze et al. 1963; Straznicky et al. 1974). The projection from each retinal half of a compound eye has been found to spread across the entire tectum when investigated in adult animals. Since the positional values of retinal ganglion cells in a perturbed eye remain stable (Straznicky & Gaze, 1980; Gaze & Straznicky, 1980a, b) the altered retinotectal topographic relationship must therefore involve changes at the tectal level.

In order to account for the results of developmental and regeneration studies on retinotectal connexions, Willshaw & von der Malsburg (1979) have proposed an elegant model for a mechanism of tectal marker induction. According to this model the tectum is initially devoid of markers. As optic fibres arrive into the tectum they interact with each other and with the tectal cells resulting in the generation of tectal positional markers. The model presumes the decay of the tectal markers after the removal of the resident optic fibre projection.

The expanded representation of the hemiretinae of the compound eyes in the ipsilateral tectum in the present study stands in contrast to our previous observations where these projections were restricted, in the presence of optic fibres from the normal eye, to the corresponding portion of the ipsilateral tectum (Tay & Straznicky, 1978; Gaze & Straznicky, 1979, 1980b). These observations indicate that the resident optic fibre projection from the normal eye maintains the whole range of tectal markers to which the hemiretinal projections from compound eyes align themselves. Since the ipsilateral tectum had been innervated by the intact eye up to two weeks after metamorphosis it had to carry the normal complement of markers at the time of the arrival of optic fibres from the compound eye. The first check point 6 weeks after operation however, showed an expanded hemiretinal projection across the entire ipsilateral tectum from NN and VV eyes and partial expansion from TT eyes. The very early regeneration pattern was not studied in this series of experiments.

Comparable experiments in fish following partial retinal ablations have shown that the initial regeneration from a residual hemiretina was restricted to the appropriate tectal half, followed later by a subsequent expansion of this projection across the entire tectum (Schmidt, 1978). In contrast, when the optic fibres from a hemiretina regenerated into a tectum which had been denervated for many months, the projection expanded initially across the entire tectum. These results indicate that tectal markers, serving as targets for regenerated optic fibres disappear during long-term denervation.

Signs of rostral, lateral and caudal expansion of the projections are noticeable in some animals in the present study as judged by the lower grain densities in the rostrolateral and lateral portions of the ipsilateral tectum in animals with NN or VV eyes and by the consistent caudomedial projection deficit in TT eye animals. From these observations we may infer that the early regeneration may have also been selective, as with the goldfish results. It is quite apparent however, that the decay of tectal positional markers in Xenopus is more rapid than in goldfish, allowing the spreading of the hemiretinal projection to occur across the tectum within a shorter time. The earliest visuotectal mappings were carried out 56–84 days after optic-nerve section. All the recordings showed the restoration of characteristic NN, TT and VV projections in both tecta. Thus not only had the extent of the hemiretinal projections in the ipsilateral tectum changed, but a new retinotopic ordering was also generated corresponding to the nature of the compound eye. The remarkably rapid change (within 6–8 weeks) in the set of tectal markers induced by the hemiretinal projection makes it unlikely that positional markers, independent from the previous innervation, existed.

The interpretation of the present results on the deficient TT eye projections with intact as well as with regenerated optic nerve is difficult. None of the previous studies on TT eye projections in adult frogs reported a discernible projection deficit by visuotectal mapping. We think that the phenomenon is transient, present during larval and young postmetamorphic life. Temporal fibres prefer rostrolateral tectum both during development and in regeneration (Straznicky et al. 1979: Gaze & Straznicky, 1980b). Consequently the expansion of the temporal hemiretinal projection is towards the caudomedial tectum. This tectal sector is furthest away from the incoming fibres and hence the last to be innervated.

In view of the goldfish results (Schmidt, 1978) and of the present observations, it is likely that detailed tectal markers are induced by the ingrowing optic fibres, yet rostrocaudal and mediolateral polarity cues are also needed for the orientation of the retinotectal map. Previous experiments with tecta which were completely deprived of optic fibre input throughout development have shown that regenerating optic fibres form an orderly but rotated projection on a rotated tectum (Straznicky, 1978). The presence of at least minimum tectal polarity cues prior to initial innervation is indicated also by observations on the development of the compound-eye projection (Straznicky et al. 1979 b). The results show that the initial outgrowth of optic fibres from TT, NN and VV eyes is to the corresponding half of the growing tectum. It appears conceivable that tectal polarity cues determine the orientation and fibre-fibre interactions the extent and the orderliness of the retinotectal projection. The possible underlying mechanisms, instrumental in retinotectal map formation have been discussed in detail elsewhere (Straznicky, Gaze & Keating, 1981).

Ms Teresa Clark’s skilled assistance in preparing the autoradiography, Ms Jenny Hiscock’s assistance in the planimetric measurements and Miss Laima Visockis’ secretarial assistance are gratefully acknowledged. The research was supported by a grant from the Australian Research Grants Committee and from the Flinders University Research Budget.

Gaze
,
R. M.
(
1958
).
The representation of the retina on the optic lobe of the frog
.
Q.J. exp. Physiol. Cogn. Med. Sci
43
,
209
214
.
Gaze
,
R. M.
(
1978
).
The problem of specificity in the formation of nerve connections
.
In Specificity of Embryological Interactions
(ed.
D. R.
Garrod
, pp.
53
93
.
London
:
Chapman and Hall
.
Gaze
,
R. M.
&
Sharma
,
S. C.
(
1970
).
Axial differences in the reinnervation of goldfish optic tectum by regenerating optic nerve fibres
.
Expl Brain Res
.
10
,
171
181
.
Gaze
,
R. M.
&
Straznicky
,
C.
(
1979
).
Selective regeneration ofopticfibers fromacompound eye to the ipsilateral tectum in Xenopus
.
J. Physiol
.
293
,
57
58
.
Gaze
,
R. M.
&
Straznicky
,
C.
(
1980a
).
Stable programming for map orientation in disarranged embryonic eyes in Xenopus
.
J. Embryol. exp. Morph
.
55
,
143
165
.
Gaze
,
R. M.
and
Straznicky
,
C.
(
1980b
).
Regeneration of optic nerve fibres from a compound eye to both tecta in Xenopus’. evidence relating to the state of specification of the eye and the tectum
.
J. Embryol. exp. Morph
.
60
,
125
140
.
Gaze
,
R. M.
,
Jacobson
,
M.
and
Székely
,
G.
(
1963
).
The retinotectal projection in Xenopus with compound eyes
.
J. Physiol
.
165
,
484
489
.
Gaze
,
R. M.
,
Feldman
,
J. D.
,
Cook
,
J.
&
Chung
,
S. H.
(
1979
).
The orientation of the visuotectal map in Xenopus: developmental aspects
.
J. Embryol. exp. Morph
.
53
,
39
66
.
Levine
,
R. L.
&
Jacobson
,
M.
(
1974
).
Development of optic nerve fiber projection is determined by positional markers in the frog’s tectum
.
Expl Neurol
.
43
,
527
538
.
Nieuwkoop
,
P. D.
&
Faber
,
J.
(
1956
).
Normal Table of Xenopus laevis
.
Amsterdam
:
(Daudin) North Holland Publ. Co
.
Schmidt
,
J. T.
(
1978
).
Retinal fibers alter tectal positional markers during the expansion of the half retinal projection in goldfish
.
J. comp. Neurol
.
177
,
279
300
.
Schmidt
,
J. T.
,
Cicerone
,
C. M.
&
Easter
,
S. S.
(
1978
).
Expansion of the half retinal projection to the tectum in goldfish. An electrophysiological and anatomical study
.
J. comp. Neurol
.
177
,
257
278
.
Sharma
,
S. C.
&
Hollyfield
,
J. G.
(
1980
).
Specification of retinotectal connexions during development of the toad Xenopus laevis
.
J. Embryol. exp. Morph
.
55
,
77
92
.
Sperry
,
R. W.
(
1951
).
Mechanism of neural maturation
.
In Handbook of Experimental Psychology
(ed.
S. S.
Stevens
), pp.
236
280
.
New York
:
Wiley
.
Sperry
,
R. W.
(
1963
).
Chemoaffinity in the orderly growth of nerve fiber patterns and connections
.
Proc. natn. Acad. Sci., U.S.A
.
50
,
703
710
.
Straznicky
,
K.
(
1978
).
The acquisition of tectal positional specification in Xenopus. Neurosci
.
Letts
.
9
,
177
184
.
Straznicky
,
C.
&
Gaze
,
R. M.
(
1980
).
Stable programming for map orientation in fused eye fragments in Xenopus
.
J. Embryol. exp. Morph
.
55
,
123
142
.
Straznicky
,
K.
,
Gaze
,
R. M.
&
Keating
,
M. J.
(
1971
).
The retinotectal projections after uncrossing the optic chiasma in Xenopus with one compound eye
.
J. Embryol. exp. Morph
.
26
,
523
542
.
Straznicky
,
K.
,
Gaze
,
R. M.
&
Keating
,
M. J.
(
1974
).
The retinotectal projection from a double-ventral compound eye in Xenopus laevis
.
J. Embryol. exp. Morph
.
31
,
123
137
.
Straznicky
,
K.
,
Tay
,
D.
&
Lunam
,
C.
(
1978
).
Changes in retinotectal projection in adult Xenopus following partial retinal ablation
.
Neurosci. Letts
.
8
,
105
111
.
Straznicky
,
C.
,
Gaze
,
R. M.
&
Horder
,
T. J.
(
1979a
).
Selection of appropriate medial branch of the optic tract by fibers of ventral retinal origin during development and in regeneration. An autoradiographic study in Xenopus
.
J. Embryol. exp. Morph
.
50
,
253
267
.
Straznicky
,
C.
,
Gaze
,
R. M.
&
Keating
,
M. J.
(
1979b
).
Selective optic fibre projection from compound eyes to the tectum during development in Xenopus
.
Neurosci. Abstr
.
5
, p. 181.
Straznicky
,
C.
,
Gaze
,
R. M.
&
Keating
,
M. J.
(
1981
).
The development of the retinotectal projections from compound eyes in Xenopus
.
J. Embryol. exp. Morph
.
62
(in the press).
Tay
,
D.
&
Straznicky
,
K.
(
1978
).
The pattern of early optic nerve regeneration from compound eyes to the tectum in Xenopus
.
Proc. Austr. Physiol. Pharmacol. Soc
.
9
, p.
65
.
Udin
,
S. B.
(
1977
).
Rearrangements of the retinotectal projections in Rana Pipiens after unilateral caudal half-tectum ablation
.
J. comp. Neurol
.
173
,
561
582
.
Willshaw
,
D. J.
&
Von Der Malsburg
,
C.
(
1979
).
A marker induction mechanism for the establishment of ordered neural mappings: its application to the retinotectal problem
.
Phil. Trans. Roy. Soc. Ser. B
287
,
203
243
.