J Neurophysiol 96: 2827–2829, 2006;

doi:10.1152/classicessays.00042.2006.

ESSAYS ON APS CLASSIC PAPERS

Roger Sperry: pioneer of neuronal specificity

Bernice Grafstein
Department of Physiology and Biophysics, Weill Medical College of Cornell University, New York, New York

This essay looks at the historical significance of two APS classic papers that are
freely available online:
Sperry RW. Optic nerve regeneration with return of vision in anurans. J Neu-
rophysiol 7: 57– 69, 1944 (http://jn.physiology.org/cgi/reprint/7/1/57).
Sperry RW. Restoration of vision after crossing of optic nerves and after
contralateral transplantation of eye. J Neurophysiol 8: 15–28, 1945 (http://jn.
physiology.org/cgi/reprint/8/1/15).

WHEN ROGER SPERRY (Fig. 1) received the Nobel Prize in 1981 the responses. On the other hand, control animals that had
for his work on functional specialization of the cerebral hemi- undergone regeneration of the optic nerve without rotation of
spheres, some of his admirers felt that the prize had been the eye always showed correct responses under the same
awarded for the wrong body of work. Or even that he deserved conditions. Thus the incorrect responses from the rotated eyes
a second prize! How the organization of the nervous system could not be attributed to any inherent defect in the regenera-
manifested itself in behavior was always at the forefront of his tion process itself, and they could not be corrected by any
interests, but the split-brain studies for which he received the functional process such as learning, even if they were of a
Nobel Prize represented only a fraction of the avenues by maladaptive nature.
which he chose to explore this subject. Another approach that Moreover, by testing visual responses after having made
figured importantly in his career was to examine the behavioral lesions in various regions of the optic tectum, Sperry showed
effects of perturbing the nervous system, especially in lower that in the regenerated animals these regions still retained their
animals, where both the brain structure and the behavioral appropriate correspondence with the original retinal sectors,
repertoire were apparently simpler to interpret. These experi- even when the eye had been rotated. Histological examination
ments led him to hypotheses about basic mechanisms regulat- of the regenerated optic nerves showed that axon outgrowth in
ing the wiring of the nervous system, hypotheses that even the region of the lesion was quite disorderly, but it was not
today influence our thinking about neuronal organization, par- possible to determine whether the axons might at some point
ticularly in development. have sought out an appropriate pathway, or whether the only
Some of Sperry’s classic experiments appear in two papers axons that survived were the ones that had by chance reached
in the Journal of Neurophysiology in 1944 and 1945. The their appropriate termination sites. It was also not possible to
object of the experiments in the first of these papers (Sperry assess the precision of the retinal projections.
1944) was to confirm in adult and immature anuran amphibians
what he himself had previously observed in the more primitive
newt (Sperry 1943a,b): when severed, optic axons reconnected
to the optic tectum (the principal visual center in these animals)
and visual function eventually returned, indicating that reflex
relations in the brain had been correctly restored. To test how
the restored vision had come about, regeneration was com-
bined with a perturbation of the visual field produced by
rotation of the eye (Sperry 1943b).
Sperry tested vision in his experimental animals using flies
impaled on a thin wire as a food lure for the adults and rotary
optokinetic stimulation in the case of the tadpoles. He found
that after a few weeks had been allowed for regeneration of the
optic nerve, the animals showed persistently reversed re-
sponses. Neither development of the tadpoles nor repeated
exposure of the adults to the stimuli produced any correction of

Address for reprint requests and other correspondence: B. Grafstein, Dept.

of Physiology and Biophysics, Weill Medical College of Cornell Univ., 1300
York Ave., New York, NY 10021 (E-mail: bgraf@med.cornell.edu). FIG. 1. Roger Sperry in 1973. (Photo courtesy of Dr. Robert W. Doty.)

http://www.the-aps.org/publications/classics 0022-3077/06 $8.00 Copyright © 2006 The American Physiological Society 2827
Downloaded from journals.physiology.org/journal/jn (138.099.029.009) on August 24, 2020.
2828 ESSAYS ON APS CLASSIC PAPERS
Sperry inferred from these experiments that “the ingrowing
optic fibers must possess specific properties of some sort by
which they are differentially distinguished in the [brain] ac-
cording to their respective retinal origins” (Sperry 1944, p. 67).
Moreover, “a basic embryonic specification [may arise]
through central self-differentiation of the nuclear [i.e., neural?]
mass itself” (Sperry 1944, p. 68) to account for the likelihood
that the central neurons to which the optic axons connected
were likewise differentially specified, in order for appropriate
connections to be made.
In his second paper in the Journal of Neurophysiology
(Sperry 1945), Sperry set the regenerating optic axons a more
difficult task by excising the optic chiasma and rerouting the
optic nerves (which are usually completely crossed in the
amphibians that he used) to connect each eye to the ipsilateral
lobe of the optic tectum. He tested these animals by optokinetic
stimulation and directed food presentation, as before, but also
made detailed examination of their spontaneous movements
and escape reactions. Again he observed inverted responses to
stimulation, made more difficult by the fact that these animals
showed prominent spontaneous circling. Some of these ani-
mals were studied over a period of several months, with no
qualitative change in their reactions.
Another approach he used was to transplant one eye to the
opposite side, thus reversing its dorsal-ventral orientation with-
out changing its nasal-temporal orientation, or vice versa. This
operation in previous workers’ hands had not led to restoration
of vision, but Sperry nevertheless found good visual recovery
in urodeles and in metamorphosing frog tadpoles at the stage of
forelimb emergence. In these animals, the visuomotor re-
sponses were altered from normal in the direction that would
have been predicted from their eye orientations. No correction
of these abnormal responses was seen even 2–3 mo after
recovery. Histological studies showed a disorderly progression
of the optic axons along the nerve, with almost all of them
crossing to the contralateral tectum and only a scattering
selecting the route to the side to which they would originally
have connected. Thus it was evident that the regenerating
axons were not passively following their old trajectories, and
hence their final orderly reconnection to the optic tectum must
have been determined by an intrinsic property reflecting the
specific location of their cell bodies in the retina. Sperry
inferred that the specific local properties would have been
produced originally in the embryo “through a field differenti-
ation of the retina” (Sperry 1945, p. 15). This process would be
analogous to the establishment of polarity in the developing
limb bud, with two differentiation gradients, corresponding to
the anteroposterior and dorsoventral axes, determined sepa-
rately.
Subsequently, data that would be necessary to support this
hypothesis were obtained in Sperry’s laboratory by demon-
strating the detailed connection pattern of the regenerating
axons (Attardi and Sperry 1963). A series of histological
studies in the visual system of fish showed that after localized
lesions in either the tectum or retina, or after surgical rerouting
of the optic tracts into foreign regions of the tectum, the
regenerating axons were seen inevitably to have grown back to
their appropriate end-zones, bypassing, if necessary, more
readily accessible but inappropriate synaptic regions. Studies
in other systems, such as innervation of skin (Miner 1956) and
muscles (Sperry and Arora 1965), likewise showed preferential
J Neurophysiol • VOL
Downloaded from journals.physiology.org/journal/jn (138.099.029.009) on August 24, 2020.
connections to the correct targets despite various surgical
perturbations of the neuron-target pathways. Sperry was led to
define this specificity as arising from “each axon linking only
with certain neurons to which it becomes selectively attached
by specific chemical affinities” (Sperry 1963, p. 704).
Despite this accumulation of evidence, however, Sperry’s
hypothesis was not readily accepted. One factor undoubtedly
was that virtually all his experiments were carried out in lower
vertebrates and hence failed to immediately attract the attention
of the community of neuroscientists oriented to “real,” i.e.,
mammalian, organisms. Hence progress in this area fell to a
relatively small number of workers, who perceived the impor-
tance (and convenience) of lower animals for establishing basic
biological principles that might prevail throughout the animal
kingdom.
Historical factors also worked against the acceptance of hy-
potheses that apparently did not adhere to the principle of Oc-
cam’s razor, namely that “one should not increase, beyond what
is necessary, the number of entities required to explain anything”
(http://pespmc1.vub.ac.be/OCCAMRAZ.html). Sperry’s original
experiments were undertaken at a time when the idea that
physical and chemical forces governed the form of biological
structures, particularly as espoused by D’Arcy Thompson
(1917), was still an important illuminating principle. Paul
Weiss’s now-classic textbook on development (Weiss 1939)
and J. Z. Young’s essay on axonal morphology (1945) strongly
emphasized this view in relation to the dynamics of growth in
the nervous system. Weiss exerted a powerful influence with
his demonstration that the outgrowth of axons in tissue culture
was determined more by the physical properties of the sub-
strate than any positive attraction by target tissue (Weiss 1934)
and that regenerating axons, given their choice of pathways,
showed no preference for the direction in which their target lay
(Weiss and Taylor 1944).1 If Occam’s razor was to be applied,
the view that axon growth was essentially determined by
mechanical factors left little room for Sperry’s postulates of
specific chemical affinities and attractive forces.
On the other hand, numerous observations that restoration of
motor function might occur even with apparently random
reinnervation of muscles (reviewed by Weiss 1936) had given
rise to explanations that required that motor neurons could
undergo functional modification subsequent to their contact
with the muscles (reviewed by Grafstein 2001a). These views
posed a particularly difficult problem for Sperry himself. He
had been a graduate student of Weiss’s and had shown that
incorrect innervation of muscles, at least in mammals, led to
incorrect movements that were not subject to any functional
modification or learning (Sperry 1941). In interpreting his
studies on the visual system he initially tried to leave room for
Weiss’s concepts of functional modulation (Sperry 1944) but
finally chose to regard them as unimportant in the face of his
own ideas (Sperry 1963).2
At the present time, of course, the resistance to Sperry’s
ideas has vanished (see review by Meyer 1998). Indeed, these
1
It is ironic that this demonstration of the absence of attraction factors in
regenerating nerve was eventually found to have been invalid due to an
ingenious but incorrect choice of substrate: nerve growth was tested in an
arterial sleeve that may have nullified the tropic influence of the degenerating
nerve stump (Politis et al. 1982).
2
But not without retaining a lifelong antipathy to Weiss and his ideas
(reviewed by Grafstein 2001a, b).
96 • DECEMBER 2006 • www.jn.org
 ESSAYS ON APS CLASSIC PAPERS
ideas may be seen as the embodiment of many principles that
we now regard as fundamental to nervous system development
and regeneration. Diffusible signals for attraction and repulsion
of growing axons and factors involved in selective synapse
formation have been identified and are being explored in great
detail (e.g., see Purves et al. 2004). A postulate that worried
Sperry no longer seems unreasonable—that there might not be
enough distinct chemical labels available to account for so
many different neurons and so many different neuronal inter-
connections in the nervous system. Now we have no qualms
about whether there is enough information available—there are
plenty of genes, plenty of time-dependent regulators of gene
expression, and plenty of posttranslational mechanisms for
modification of gene products.
REFERENCES
Attardi DG and Sperry RW. Preferential selection of central pathways by
regenerating optic fibres. Exp Neurol 7: 46 – 64, 1963.
Grafstein B. Half a century of regeneration research. In: Regeneration in the
Central Nervous System, edited by NA Ingoglia and M Murray. New York:
Marcel Dekker p. 1–18, 2001a.
Grafstein B. Bernice Grafstein. Hist Neurosci Autobiog 3: 246 –282, 2001b.
Meyer RL. Roger Sperry and his chemoaffinity hypothesis. Neuropsychologia
36: 957–980, 1998.
Miner N. Integumental specification of sensory fibers in the development of
cutaneous local sign. J Comp Neurol 105: 161–170, 1956.
Politis MJ, Ederle K, and Spencer PS. Tropism in nerve regeneration in
vivo. Attraction of regenerating axons by diffusible factors derived from
J Neurophysiol • VOL 96 • DECEMBER 2006 •
Downloaded from journals.physiology.org/journal/jn (138.099.029.009) on August 24, 2020.
2829
cells in distal nerve stumps of transected peripheral nerves. Brain Res 253:
1–12, 1982.
Purves D, Augustine GJ, Fitzpatrick D, Hall WC, Lamantia A-S, Mc-
Namara JO, and Williams SM. Neuroscience (3rd ed.). Sunderland, MA:
Sinauer Associates 2004, p. 537–539.
Sperry R. The effect of crossing nerves to antagonistic muscles in the hind
limb of the rat. J Comp Neurol 75: 1–19, 1941.
Sperry R. Visuomotor coordination in the newt (Triturus viridescens) after
regeneration of the optic nerve. J Comp Neurol 79: 33–55, 1943a.
Sperry R. Effect of 180 degree rotation of the retinal field on visuomotor
coordination. J Exp Zool 92: 263–279, 1943b.
Sperry RW. Optic nerve regeneration with return of vision in anurans.
J Neurophysiol 7: 57– 69, 1944.
Sperry RW. Restoration of vision after crossing of optic nerves and after
contralateral transplantation of eye. J Neurophysiol 8: 15–28, 1945.
Sperry RW. Chemoaffinity in the orderly growth of nerve fiber patterns and
connections. Proc Natl Acad Sci USA 50: 703–710, 1963.
Sperry RW and Arora HL. Selectivity in regeneration of the oculomotor
nerve in the cichlid fish, Astronotus ocellatus. J Embryol Exp Morphol 14:
307–317, 1965.
Thompson DW. On Growth and Form. Cambridge, UK: Cambridge Univer-
sity Press, 1917.
Weiss P. In vitro experiments on the factors determining the course of the
outgrowing nerve fiber. J Exp Zool 68: 393–344, 1934.
Weiss P. Selectivity controlling the central-peripheral relations in the nervous
system. Biol Rev Camb Philos Soc 11: 494 –531, 1936.
Weiss P. Principles of Development. A Text in Experimental Embryology. New
York: Holt, 1939.
Weiss P and Taylor AC. Further experimental evidence against “neurotro-
pism” in nerve regeneration. J Exp Zool 95: 233–257, 1944.
Young JZ. The history of the shape of a nerve-fibre. In: Essays on Growth and
Form presented to D’Arcy Wentworth Thompson, edited by Le Gros Clark
WE and Medawar PB. Oxford: Clarendon Press, p. 41–94, 1945.
www.jn.org