/ . Embryo!. exp. Morph. Vol. 35, I, pp. 41-57, 1976
41
Printed in Great Britain
Reorganization of retinotectal
projection of compound eyes after various
tectal lesions in Xenopus
By K. STRAZNICKY 1
Department of Anatomy, School of Medicine, The University of Zambia,
and Division of Developmental Biology, National Institute for
Medical Research, London
SUMMARY
The retinotectal connexions of double nasal (NN), double temporal (TT) and double
ventral (VV) eyes in juvenile Xenopus were mapped after post-metamorphic removal of the
rostral, caudal, medial or lateral tectal halves with the subsequent cutting of the optic nerve
of the operated eye.
A whole visual field projection occurred from NN eyes on to residual caudal tectum, from
TT eyes on to residual rostral tectum and from VV eyes on to residual medial tectum. When
optic fibres from NN, TT or VV eyes grew into inappropriate rostral, caudal or lateral tectal
halves respectively, there was a projection deficit in the nasal and temporal or in the dorsal
and ventral poles of the visual field. The persisting scotomas in the visual field indicated that
only the optic fibres of the central retinal fundus had succeeded in connecting with an
inappropriate tectal half whereas the peripheral opticfibreshad not. The incongruous results
of optic nerve regeneration obtained in the various recombinations may be taken to indicate
that the assumption of early embryonic pattern regulation is inadequate to explain the
findings in this experimental situation.
INTRODUCTION
Experiments in adult goldfish have revealed a reorganization of the retinotectal projection following partial tectal ablations. After removal of the caudal
half (Gaze & Sharma, 1970), the rostral half (Sharma, 1972a), the caudomedial
sector (Yoon, 1971), or a rostro-caudal strip of the optic tectum (Sharma,
1912b), the entire visual field projection redistributed itself in a correct retinotopic order over the residual tectum. Using a gelatine film barrier across the
medio-lateral extent of the tectum, Yoon (1972) demonstrated that the visual
projection could compress onto the rostral half of the tectum; later, after
absorption of the gelatine film barrier in the same animal, the field compression
disappeared and the normal retinotectal projection was restored across the
whole tectum. These results indicate that the retinotectal projection in these
1
Author's address: Department of Human Morphology, School of Medicine, The Flinders
University of South Australia, Bedford Park, South Australia, 5042, Australia.
42
K. STRAZNICKY
animals is modifiable and shows certain somewhat elastic properties, and also
that the experimental reorganization of the projection is reversible.
The results of tectal ablation experiments in Anura stand in direct contrast
to those obtained in goldfish. After larval removal of the rostral or the caudal
tectal halves (Straznicky, Gaze & Keating, 1971a) or after post-metamorphic
partial tectal ablations (Straznicky, 1973) in Xenopus, a consistent defect in the
corresponding visual field was found. Similar observations have been made in
the tree frog (Meyer & Sperry, 1973) following tectal ablations or the implantation of a gelatine film separating the rostral and caudal tectal halves. The
resulting visual field scotomas persisted unaltered for a period of up to one year.
In these cases it can be assumed from the incomplete visual field projections
that only the appropriate retinal fibres connected selectively with the residual
tectal halves. Thus late larval or post-metamorphic tectal lesions revealed no
signs of any plastic readjustment of the retinotectal projection in Anura.
Meyer & Sperry (1973) have interpreted these findings as further supporting
evidence in favour of a position-specific determination of the retinotectal
connexions, presumably on the basis of the selective cytochemical affinities
between retinal and tectal neurons suggested by Sperry (1951, 1965).
Compound eyes, surgically constructed from two nasal (NN), temporal (TT)
or from two ventral (VV) halves (Gaze, Jacobson & Szekely, 1963, 1965;
Straznicky, Gaze & Keating, 1974) in Xenopus appear to form a projection,
during neurogenesis, from each half retina across the entire tectal surface,
which normally receives optic fibres from the whole eye. A comparable result
has been found in Xenopus after post-metamorphic section of the optic nerve,
where regenerating optic fibres from a double nasal or double temporal eye
spread over the whole tectum (Straznicky, Gaze & Keating, 1971 b). These
results indicate that interference with the retina, as long as it is performed at
an early embryonic stage, permits plasticity in the developmental patterning of
the retinotectal connexions in Xenopus. It has been suggested by the same
authors that, in relation to compound eyes, the reorganization of the retinotectal projection, in terms of spreading, might occur at one of two different
levels of the retinotectal system. It is possible that (i) the 'rules' of interconnexion between optic fibres and tectal cells are such as to permit modification
of the projection pattern after certain operations on retina or tectum. Thus a
change in the pattern of retinotectal connexions in this situation need not
indicate any alteration in cell specificities. This alternative has received support
from recent experiments in which the visual projections were mapped during
development in Xenopus (Gaze, Keating & Chung, 1974). The observations on
the developing retinal projection in Xenopus suggested that the retinal input
to a particular tectal cell progressively alters during larval life and that the
development of the adult retinotectal projection is completed and stabilized
only by the beginning of the metamorphic climax. Alternatively, since the
compound eyes were formed at an early embryonic stage, it is possible (ii) that
Retinotectal projection of compound eyes
43
each half of the operated eye may undergo an embryonic pattern regulation
after which both halves would have the whole range of nasotemporal and dorsoventral specificity characteristics. Thus each half of a compound eye would
represent a whole eye in terms of specificity.
In order to be able to determine which of these alternatives is operative in the
establishment of the compound eye projection the following experimental
approach was devised. Compound nasal (NN), temporal (TT) or ventral (VV)
eyes were formed in Xenopus embryos and in addition, after metamorphosis,
the rostral, medial, lateral or caudal half of the optic tectum was removed.
To give an equal chance to optic fibres from all over the retina to reconnect with
the residual tectum the optic nerve of the operated eye was simultaneously cut.
The present report shows evidence for the complete reorganization of the
compound eye projection over an appropriate tectal half. In this context
'appropriate' means appropriate to the nature of the compound eye; i.e.
caudal half tectum to NN, rostral half to TT and medial half to VV eyes
respectively.
The systematic presence of a full-field compound eye projection to an
appropriate tectal half casts doubt on the idea of embryonic pattern regulation
as an adequate explanation for the peculiarities of the compound eye projection.
METHODS
Xenopus laevis embryos of stage 32 (Nieuwkoop & Faber, 1956) were
anaesthetized with an 0-1 % MS222 (Tricaine methanesulphonate, Sandoz)
and the right optic cup cut in half along the vertical midline. The nasal half
was then removed and replaced by the temporal half of the eye cup from the
left eye of another embryo, thus forming a TT eye. In other embryos in a
similar way, NN eyes were made. In the third group of embryos the right eye
cup was cut in half along the horizontal midline and the dorsal half replaced
by a ventral half from the left eye of another embryo to obtain a VV eye.
Three to four weeks after metamorphosis, 20 NN, 20 TT and 12 VV eye
animals underwent a second operation. Animals were anaesthetized with
MS222 (0-3 %), the left tectum was exposed and either its rostral, caudal,
medial or lateral half was removed. The corresponding right optic nerve was
simultaneously cut. The piece of cartilage and flap of skin were approximated
and sealed with tissue adhesive (isobutyl 2-cyanoacrilate monomer, ETHICON).
A few animals with TT, NN or VV eyes were kept without tectal operation for
control purposes. After the two successive operations the following recombinations were obtained:
(a)
(b)
(c)
(d)
NN eye
NN eye
TT eye
TT eye
with
with
with
with
residual
residual
residual
residual
rostral
caudal
rostral
caudal
half
half
half
half
tectum
tectum
tectum
tectum
(NNR),
(NNC),
(TTR),
(TTC),
44
K. STRAZNICKY
(e) VV eye with residual medial half tectum (VVM),
(/) W eye with residual lateral half tectum (VVL).
The six groups of animals were reared to a size of 5-7 cm body length (approximately 3-5 months after the 2nd operation) and then used for electrophysiological mapping of the retinotectal projection of compound eyes to the residual
tectal halves. The electrophysiological and histological methods used have been
described in detail in previous papers (Straznicky et al. 19716; Straznicky,
1973). Under ether anaesthesia the animals were decerebrated and given
O-O3-O-O5 mg tubercurarine intramuscularly and both right normal and left
operated tectum were exposed and covered with liquid paraffin. A picture was
then taken from the dorsal surface of the tecta and a x 50 magnified print
made with a 10 x 10 mm rectangular grid superimposed on it. The animal was
set up with the right, operated eye centred on the fixation point of an 'Aimark'
perimeter. The stimulus, a black disc of diameter of 5°-10°, evoked multi-unit
action potentials in the superficial layers of the tectum, and these were recorded
with a lacquer-insulated tungsten electrode with a naked tip diameter about
1-2 jura. The electrode position was changed systematically at the intersections
of the grid in 100-^m steps across the tectum on the basis of the print and at
each location the receptive field position was determined through the operated
eye. The receptive fields from which constant localized responses could be
evoked were about 10°-40° in diameter. Electrode positions from which no
responses were elicited or where the responses happened to be too faint or
inconstant to be localized were marked on the chart with an ' 0 ' . The heads of
the operated animals were sectioned serially either in the sagittal plane (with
rostral and caudal tectal lesions) or in the transverse plane (with medial and
lateral tectal lesions) and stained with Holmes' silver method. The maximum
length of the operated tectum either across the rostro-caudal or the mediolateral extent was measured and compared to the length of the corresponding
extent of the normal tectum on the other side of the brain in most of the experimental animals. With the help of the photographs of the dorsal surface of the
tectum, the actual electrode positions in the histological sections could be
determined with an accuracy of 100 fim. A camera lucida reconstruction was
finally made from the frontal or sagittal sections through the tectum indicating
the size and the exact location of the lesion as well as the approximate site of
the rows of electrode positions on the dorsal surface of the residual tectum.
RESULTS
The area of tectal ablation was well recognizable at the time of the mapping.
The tectal wound was covered with the pigmented pia layer or, in some cases,
the healing process was so complete that hardly any scar was noticeable,
although the operated left tectum was approximately half the size of the
normal one, either across the rostro-caudal or across the medio-lateral extent.
Retinotectal projection of compound eyes
45
Five animals died between the 2nd operation and the time of electrophysiological
recording. From the remaining 47 animals, 32 successful recordings were
obtained. The mapping results of 6 animals had to be excluded because in each
more than two thirds of the operated left tectum was present. All six cases gave
almost complete compound eye projections. The present report comprises the
results of electrophysiological mapping and histological observations obtained
in 26 animals.
(1) Normal visual field projection
The visual field projection through the right eye was recorded in three
normal juvenile Xenopus. The maps obtained were the same as described by
Gaze (1959) in that the nasal half of the visual field projected to the rostral
half of the tectum, the temporal half of the visual field to the caudal half, the
dorsal part of the visual field to the medial half and the ventral part of the
visual field to the lateral half of the tectum (Fig. 1 A).
(2) Visual projections of compound eyes without tectal lesions
Three animals each with an NN eye, three with a TT eye and two with a
VV eye were mapped. The compound eye projections in these cases were similar
to the visual field maps obtained by Gaze et ah (1963) from NN and TT eyes
and by Straznicky et ah (1974) from VV eyes. Each of the hemiretinae projected
to the whole tectum; thus each electrode position had two corresponding field
positions, arranged more or less symmetrically about the vertical or about the
horizontal midlines. With NN eyes the nasal and temporal poles of the visual
field projected to the caudal pole of the tectum, whereas the central field was
represented on the rostral pole of the tectum (Fig. 1B). With TT eyes the nasal
and temporal poles of the visual field projected to the rostral pole and the
central sector of the visual field to the caudal pole of the tectum (Fig. 2A).
With VV eyes the dorsal and ventral poles projected to the medial edge of the
tectum, while visual field points along the horizontal midline were represented
on the lateral edge of the tectum (Fig. 2B).
(3) Visual projections of compound eyes with various tectal lesions
Three to five months elapsed between the time of the second operation and
the mapping of the visual projections. This period, according to earlier observations (Straznicky, 1973), is sufficient for the completion of the optic nerve
regeneration to the residual tectum. In fact, in all but one of the recorded
animals the retinotectal projection was ordered, suggesting that the regeneration
process had gone to completion within the time span of the experiments. The
character of the visual projections will be described according to the type of
operation performed.
46
K. STRAZNICKY
id) Animals with right double nasal eye and with left residual rostral tectal half
(NNR)
Six operated animals comprised this group, in which optic fibres from a
double nasal eye innervated the residual rostral tectal half. The compound eye
projections showed a regular deficit in the extreme nasal and temporal fields.
A typical map is presented in Fig. 3. Wide visual field scotomas appear both
L. tectum
L. lectum
0
28
27 26
25
24
23 22
NN eye
visual field
Fig. 1. In this and each of the following figures the numbers on the tectum represent
electrode positions. The corresponding stimulus positions ate indicated on the chart
(lower part of the diagram) of the visual field by the same number. A, The projection
of the visual field of the right eye to the left tectum in a normal animal. B, The projection of a right compound nasal eye (NN) to the normal left tectum. The temporal
and nasal poles of the visual field project to the caudal end of the tectum, whereas
the central retina is represented on the rostral end of the tectum.
in the temporal and nasal poles (shaded areas), which otherwise could have
projected to the removed caudal tectal half. The rest of the visual field projected
in a retinotopic manner to the available rostral half tectum. In one animal poor
responses were obtained, and only from the central region of the visual field.
The responses in the field in this case were otherwise retinotopically organized.
Retinotectal projection of compound eyes
47
(b) Animals with right double nasal eye and with left residual caudal tectal half
(NNC)
Of the five animals of this group four gave a full field visual projection where
both the nasal and the temporal (through a double nasal retina) halves of the
visual field projected to the whole extent of the residual caudal half tectum
(Fig. 4). The majority of the electrode positions corresponded each to two
L. tectum
L. tectum
26 25 24 23^N
0 22 21 20 19 18
0 17 16 15 14 13
12 11 10 9 8
0
X
7
3
6
i
5 4
y
R. VV eye
visual field
Fig. 2. The projection of right compound temporal (TT) and ventral (VV) eyes to
the normal left tectum. In TT eyes (A) the nasal and the temporal poles of the visual
field project to the rostral, the central field to the caudal end of the tectum. In VV
eyes (B) the dorsal and the ventral poles of the visual field project to the medial edge,
the centre of the field to the lateral edge of the tectum.
stimulus positions, symmetrically arranged about the vertical midline. The
multi-unit receptive fields of responses coming from the edge of the residual
tectum were wide, with diameters up to 30°-40° as compared to the 10°-15°
width of the receptive fields of good, localized responses. In one animal (Fig. 5)
the field projection to the caudal half tectum was complete with the exception
of a narrow central sector of visual field. The histology of the operated eye of
48
K. STRAZNICKY
this animal looked normal. No noticeable scar was present in the retina which
might have caused the central visual scotoma.
(c) Animals with right double temporal eye and with left residual rostral tectal
half(TTR)
Five animals were successfully recorded in this group. In one animal no
response could be evoked from the temporal visual field, perhaps because of a
L. tectum
0
14
8
17
0 ^
16 15
13 12 11 10
7
6
5
9
4
R. NN eye
visual field
NNR
Fig. 3. The projection of the right visual field of the NN eye to the left rostral half
tectum of an NNR animal, mapped 120 days after tectal ablation. The hatched area
of the visual field indicates the extent of the scotoma. A sagittal section along the
thick arrow in the left tectum is illustrated on the right of the figure and the shaded
area shows the size of tectal ablation. Thin arrows of the sagittal section mark rows
of tectal electrode positions. In this and each of the following figures ' 0 ' on the
tectum represents electrode position from where no localizable response was elicited.
failure of connexions between the transplanted hemiretina and the optic
tectum. The original temporal retinal half (the nasal half of the visual field),
however, projected in a retinotopic order to the residual tectum. Responses
were faint and rather wide along the vertical meridian of the field and these
projected to near the cut edge of the rostral tectum. In the other four animals
Retinotectal projection of compound eyes
49
there was a good restoration of the visual projections; thus the entire field was
represented on the residual rostral tectum (Fig. 6). The TT eye projection to a
whole (Fig. 2 A) tectum or to a residual rostral tectum appears to be very
similar, indicating a successful compression of visual projection to an appropriate half tectum.
L. tectum
20 19%
18 17 16s\
0 15 14 13 12 0
* - — •
0
0 11
6 5
10
4
9
3
8 7
2 .1.
R \
R. NN eye
visual field
NNC
Fig. 4. Full field visual projection occurred from the NN eye on to the appropriate
caudal tectal half in an NNC animal. Recording was performed 95 days after the
tectal ablation. A sagittal section along the thick arrow in the left tectum is illustrated on the right of the figure and the shaded area shows the size of the tectal ablation. Thin arrows of the sagittal section mark rows of tectal electrode positions.
(d) Animals with right double temporal eye and with left residual caudal tectal
half(TTC)
Four animals were recorded in this group. In all animals visual responses
were absent from the peripheral third of the temporal and nasal halves of the
visual field. The representative case is seen in Fig. 7. The field scotomas correspond closely in location and dimension with the tectal lesion. Electrode
positions in the first three rows of the residual tectal half gave reduplicated
visual responses disposed approximately symmetrically about the vertical
4
EM B
35
50
K. STRAZNICKY
meridian. Only single visual responses were obtained from the more caudal
tectal electrode positions and these came from the centre of the visual field.
(e) Animals with right double ventral eye and with left residual medial tectal
half(VVM)
Full field visual projections were obtained in all the four animals recorded.
The double ventral projections were basically similar to those found from VV
eyes in animals without tectal lesions (Fig. 2B), although responses from the
L. tectum
R. NN eye
visual field
NNC
Fig. 5. The projection of the right NN eye to the left caudal half tectum of an NNC
animal 120 days after the tectal ablation. Note the narrow gap in the centre of the
visual field (hatched area) from which no visual responses could be evoked.
lateral edge of the residual tectum were wider than normal. The visual field
positions (Fig. 8) were symmetrically distributed about the horizontal meridian.
The rows of visual field positions from the caudal part of the residual tectum
were bent slightly toward the temporal pole of the field, thus exhibiting a
'cartwheeling phenomenon' as described earlier in the projections from some
Retinotectal projection of compound eyes
51
VV eyes (Straznicky et a I. 1974). No signs of scotoma can be seen as a result of
the lateral tectal lesion.
(/) Animals with right double ventral eye and with left residual lateral tectal
half(VVL)
Two animals were recorded in this group. One animal gave very poor and
random responses from the central sector of the visual field. Some of the
L. teclum
L. tectum
15
19
14
18
13
0
0
10
6
9
5
17 16
12 11
8
4
7
3
0
2
R. TT eye
visual field
TTC
Fig. 6
Fig. 7
Fig. 6. The projection of the right TT eye to the left rostral half tectum of a TTR
animal .150 days after the tectal ablation. The whole visual projection is compressed
on to the residual rostral tectal half.
Fig. 7. The projection of the right TT eye to the left caudal half tectum of a TTC
animal 135 days after the tectal ablation. Responses were absent from the nasal
and temporal poles of the visual field (hatched area). Compare this figure to Fig. 3.
electrode positions yielded one, others two responses. The unorganized central
projection seemed to be similar to the incomplete early optic nerve regeneration
reported in frog by Gaze & Jacobson (1963). In the other animal the double
4-2
52
K. STRAZNICKY
ventral eye projection revealed large scotomas in the dorsal and ventral poles of
the visual field (Fig. 9). The responses from the central third of the visual field
were retinotopically organized both along the nasotemporal and dorsoventral
directions.
L. tectum
,C
R. VV eye
visual field
VVM
Fig. 8. The projection of the right VV eye onto the appropriate medial tectal half in
a VVM animal. The animal was mapped 110 days after the tectal ablation. A frontal
section along the thick arrow in the left tectum is illustrated on the right of the figure
and the shaded area shows the size of tectal ablation. Thin arrows on the frontal
section mark tectal recording rows.
DISCUSSION
The present experiments were designed to investigate the possible occurrence
of 'embryonic pattern regulation' in compound eyes, by analysis of the reorganization of compound eye projections after partial tectal removal. The
visual field maps obtained fall into two categories:
(1) In animals of groups TTR, NNC and VVM the consistent finding after
optic nerve regeneration was the relative completeness of the compound eye
projections. The visual field maps are very similar to those described earlier by
Gaze et al. (1963) and Straznicky et al. (1974) except that the projections in
Retinotectal projection of compound eyes
53
the present experiments have been compressed onto a residual tectal half. In
these animals the compound temporal, nasal and ventral eyes innervated the
appropriate rostral, caudal and medial halves of the tectum respectively, which
halves in normal animals would receive optic fibres from the temporal, nasal or
ventral halves of the retina. Tectal ablation experiments in fish (Gaze & Sharma,
L. tectum
R. VV eye
visual field
VVL
Fig. 9. The projection of the right VV eye to the left (inappropriate) lateral tectal half
in a VVL animal. The animal was mapped 120 days after the tectal ablation. A
frontal section along the thick arrow on the left tectum is illustrated on the right of
the figure and the shaded area shows the extent of the tectal ablation. Thin arrows
on the frontal section mark rows of tectal recording positions. The thick arrow
with the short shaft indicates the extreme lateral extent of the optic tectum.
1970; Yoon, 1971) have shown evidence of compression of the whole visual
field projection onto residual tectal halves. It has been suggested by Yoon
(1971) that the reorganization of the retinotectal projections in fish may be
based on the modification of the positional specific properties of individual
tectal cells, which leads to the recreation of a complete range of cellular specificities in the remaining tectal fragment. Although visual field compression
54
K. STRAZNICKY
occurred in TTR, NNC and VVM animals the situation here is rather more
complex than in the case of goldfish partial tectal ablation. ]n the present
combinations the optic fibres of compound eyes formed complete projections
in the sense that both similar retinal halves were represented in the appropriate
tectal halves.
The full-field visual projections, however, need not necessarily involve
modification of the position-specific properties of either the tectal or retinal
cells. They could merely reflect a completed optic fibre regeneration from the
compound eyes to the appropriate tectal halves. This is, in fact, what should
happen in the absence of both retinal and tectal regulation.
(2) In the other three groups (TTC, NNR and VVL), where the compound
eyes connected with the 'wrong' tectal halves, the visual field projections were
incomplete in that the peripheral nasal and temporal or the peripheral dorsal
and ventral fi.elds were systematically missing. It can be inferred from the
consistent visual field scotomas that only the central retinal fibres succeeded in
projecting to the inappropriate tectal halves while the peripheral retinal fibres
did not. The optic nerve of a juvenile Xenopus consists of about 60000 myelinated and unmyelinated nerve fibres (Wilson, 1971). From the size of the
scotomas one can estimate that about one third or half of the optic fibres failed
to reach the residual tecta. The most probable target for these strayed fibres
seems to be the appropriate tectal loci of the right intact tectum, and the
pretectal neuropil area which normally receives an ordered projection from the
contralateral retina (Lazar, 1971).
The present results have two direct implications. The first relates to the
possible existence of embryonic pattern regulation in a compound eye. It has
been suggested (Gaze, 1970) that retinal cells in the centre of a double nasal or
double temporal eye could perhaps take up the specific character of the missing
pole; thus each half of a NN or TT compound eye would have a full range of
specificities from the nasal to the temporal poles. Similarly each half of a
double ventral eye would comprise the full range of specificity properties from
the ventral to the dorsal poles, the latter being in the geometrical centre
of the compound retina. In terms of specificity properties, therefore, the compound eyes from this view point could be regarded (Straznicky et al. 1971 b)
as 'two small normal eyes' assembled in an unconventional manner. Recent
experimental results (Hunt & Jacobson, 1974) in various Xenopus compound
eye situations have indeed indicated alterations in the visual field projection that
can be taken as signs for the possible remodelling of the specificity properties
of fused eye fragments.
If embryonic pattern regulation were to occur in the retina after the formation
of the compound eye, the projections to be expected from the various recombinations would be as shown in Fig. 10. In NNC, TTR and VVM combinations
the secondarily generated temporal, nasal and dorsal poles (in the centre of the
retina) should remain without an appropriate tectal half, consequently a central
Retinotectal projection of compound eyes
55
scotoma in the visual field should be present. Only one out often cases indicated
the presence of a narrow central scotoma; the rest exhibited 'conventional'
full-field compound eye projections. In the case of retinal regulation one would
therefore have to assume in the present experiments that, after partial tectal
removal, either the ganglion cells of the retinal centre (the secondarily generated
temporal, nasal and dorsal poles) become respecified again or alternatively that
the compound eye projection exhibits elastic properties of connexion. In NNR,
Nature of
compound eye
/
\
/
\
/
v
\
Type of
tectal ]v]
ablation
Predicted
visual T
field
NNR
NNC
TTR
TTC
VVM
VVL
Fig. 10. The nature of compound eyes and the types of tectal ablation in the six
experimental groups are summarized. Note that in three groups the actual findings
matched the predicted visual field projections (with assumed retinal regulation), in
the other three groups they did not.
TTC and VVL combinations only the centre of the retina projected to the
residual tecta since the appropriate tectal half was missing. In these cases the
expected projections and the actual findings match each other and this could
suggest that retinal regulation had occurred following the original operation
to form the compound eye. It is, however, very unlikely that in half of the
combinations pattern regulation occurred whereas in the other half it did not.
For the reasons given above it is fair to conclude that embryonic pattern
regulation of compound eyes has not been substantiated by the present experiments.
The second implication of our results relates to the presence of a central
visual field projection in NNR, TTC and VVL animals. It has been shown
56
K. STRAZNICKY
(Gaze et al. 1963; Straznicky et al. \91\b) that each half of a compound eye
connects across the whole tectum. After post-metamorphic partial tectal
removal the compound eye can connect, in order, across only a half tectum,
but this will only happen if the nature of the compound eye (NN, TT, VV) is
appropriate to the residual half tectum (caudal, rostral, medial). The different
results from the groups NNR, TTC, VVL and NNC, TTR, VVM underline
the importance of the retinal and tectal poles in achieving a whole projection on
to a residual tectum. It is conceivable that, in matching up the retinal projection
on to the tectum, the tectal poles act as 'road signs' to which the corresponding
optic fibres are recruited, in a retinotopic order. When the poles of the residual
tectum and the compound eye are grossly mis-matched, as in NNR, TTC
and VVL animals, the reorganization of the projection is incomplete and
restricted to the central part of the retina. Thus to this limited extent the retinotectal projection shows plasticity, but the nature of the plasticity is obscure and
is beyond the scope of the present paper. The most general statement we can
make about the presence of a central retinal projection on to an inappropriate
tectal half is that differences in behaviour between fibres from central and
peripheral retina may be involved. This problem, however, remains to be
elucidated by further investigations.
The author wishes to thank Mr N. Newton (Lusaka) for his skilled assistance with making
the tectal histology.
REFERENCES
GAZE, R. M. (1959). Regeneration of the optic nerve in Xenopus laevis. Q. Jl exp. Physiol
44, 290-308.
GAZE, R. M. (1970). The Formation of Nerve Connections. London: Academic Press.
GAZE, R. M. & JACOBSON, M. (1963). A study of the retinotectal projection during
regeneration of the optic nerve in the frog. Proc. Roy. Soc. B 157, 420-448.
GAZE, R. M. & SHARMA, S. (1970). Axial differences in the reinnervation of the goldfish
optic tectum by regenerating optic nerve fibres. Expl Brain Res. 10, 171-181.
GAZE, R. M., JACOBSON, M. & SZEKELY, C. (1963). The retinotectal projection in Xenopus
with compound eyes. /. Physiol., Lond. 165, 484-499.
GAZE, R. M., JACOBSON, M. & SZEKELY, C. (1965). On the formation of connections by
compound eyes in Xenopus. J. Physiol., Lond. 176, 409-417.
GAZE, R. M., KEATING, M. J. & CHUNG, S. H. (1974). The evolution of the retinotectal map
during development in Xenopus. Proc. R. Soc. Lond. B 185, 301-330.
HUNT, R. K. & JACOBSON, M. (1974). Development of neuronal locus specificity in Xenopus
retinal ganglion cells after surgical eye transection or after fusion of whole eyes. Devi Biol.
40, 1-15.
LAZAR, G. (1971). The projection of the retinal quadrants on the optic centres in the frog.
Ada morph. hung. 19, 325-334.
MEYER, R. L. & SPERRY, R. W. (1973). Tests for neuroplasticity in the anuran retinotectal
system. Expl Neurol. 40, 525-539.
NIEUWKOOP, P. D. & FABER, J. (1956). Normal Table o/Xenopus laevis (Daudin). Amsterdam:
North Holland Publish.
SHARMA, S. (1972C). Reformation of retinotectal projections after various tectal ablations in
adult goldfish. Expl Neurol. 34, 171-182.
Retinotectal projection of compound eyes
SHARMA, S.
57
(19726). Redistribution of visual projection in altered optic tecta of adult goldfish.
Proc. natn. Acad. Sci., U.S.A. 69, 2637-2639.
SPERRY, R. W. (1951). Mechanism of neuronal maturation. In Handbook of Experimental
Psychology (ed. S. S. Stevans), pp. 236-280. New York: Wiley.
SPERRY, R. W. (1965). Embryogenesis of behavioural nerve nets. In Organogenesis (ed. R. L.
De Haan & H. Ursprung), pp. 161-186. New York: Holt, Rinehard & Winston.
STRAZNICKY, K. (1973). The formation of the optic projection after partial tectal removal in
Xenopus. J. Embryol. exp. Morph. 29, 397-409.
STRAZNICKY, K., GAZE, R. M. & KEATING, M. J. (197.1a). The establishment of retinotectal
projections after embryonic removal of rostral or caudal half of the optic tectum in
Xenopus laevis toad. Proc. int. Un. Physiol. Sci. 9, 540.
STRAZNICKY, K., GAZE, R. M. & KEATING, M. J. (19716). The retinotectal projection
after uncrossing the optic chiasma in Xenopus with one compound eye. /. 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.
WILSON, M. (1971). Optic nerve fibre counts and retinal ganglion cell counts during development of Xenonpus laevis (Daudin). Q. Jl exp. Physiol. 56, 83-91.
YOON, M. (1971). Reorganization of retinotectal projection following surgical operation on
the optic tectum in goldfish. Expl Neurol. 33, 395-411.
YOON, M. (1972). Reversibility of the reorganization of retinotectal projection in goldfish.
Expl Neurol. 35, 565-577.
{Received 8 May 1975, revised 28 September 1975)