Neuroscience 162 (2009) 549 –559
REVIEW
CEREBELLUM: HISTORY
M. GLICKSTEIN,a* P. STRATAb,c AND J. VOOGDd
a
Department of Cell and Developmental Biology, University College
London, Gower Street, London WC1E 6BT, UK
b
Department of Neuroscience and National Institute of Neuroscience–
Italy, University of Turin, Corso Raffaello 30, 10125 Turin, Italy
c
EBRI-Santa Lucia Foundation (IRCCS), via del Fosso di Fiorano 64,
00143 Rome, Italy
d
Department of Neuroscience, Erasmus Medical Center, 3000 CA
Rotterdam, The Netherlands
Abstract—This paper will outline the history of study of the
cerebellum from its beginnings to relatively recent times.
Although there is no unanimous agreement about what the
cerebellum does or how it does it, some principles of its
structure and function are well understood. The historical
approach can help to identify remaining questions and point
the way to future progress. We make no effort to separate
anatomical, physiological and clinical studies; rather, we
hope to emphasize their interrelation. The cerebellum has
always been seen as a distinct subdivision of the brain. Over
the years there was an increasingly accurate description of
its gross appearance and major subdivisions. By the begin-
ning of the 19th century, the classical descriptive anatomical
work was completed, and experimental study of the functions
of the cerebellum began. Lesions were made in the cerebel-
lum of experimental animals, and the behavioral deficits that
were caused by the lesion were studied and described. These
early animal studies powerfully influenced clinical interpre-
tation of the symptoms seen in patients with cerebellar dis-
ease. Several questions are implicit in the anatomical and clin-
ical studies of the nineteenth and early twentieth centuries,
some of which remain incompletely answered. Many of these
are addressed in other chapters in this volume. 1. Do different
parts of the cerebellum do different things? The uniformity of
the neuronal architecture of the cerebellar cortex suggests that
each small region must operate in a similar way, but it is also
clear that different regions control different functions. Is there a
systematic sensory and/or body representation? 2. What are
the functions of the cerebellar hemispheres? Massive in hu-
mans and very large in primates, their functions remain in dis-
pute. Because the size of the cerebellar hemispheres parallels
the development of the cerebral cortex, some have suggested
that the hemispheres in humans and the higher primates may
play a role in cognitive functions. 3. If one part of the cerebellum
is damaged, can another part take over? A related question is
whether normal motor function is possible in cases of complete
or near-complete agenesis of the cerebellum. 4. What are the
functions of the two distinctly different afferent systems to the
cerebellum; the climbing and mossy fibers? Crown Copyright
© 2009 Published by Elsevier Ltd on behalf of IBRO. All rights
reserved.
*Corresponding author. Tel: ⫹44-2067-388-0679.
E-mail addresses: m.glickstein@ucl.ac.uk (M. Glickstein), Piergiorgio.
strata@unito.it (P. Strata), janvoogd@bart.nl (J. Voogd).
0306-4522/09 $ - see front matter. Crown Copyright © 2009 Published by Elsevier Ltd on behalf of IBRO. All rights reserved.
doi:10.1016/j.neuroscience.2009.02.054
549
Key words: cerebellum, Purkinje, neuroscience history, Golgi
method, ramon y Cajal.
Contents
Experimental evidence of cerebellar function 552
Clinical interpretation of cerebellar lesions and disease 553
Functional localization and comparative anatomy 554
The related problems of recovery from cerebellar lesion and
cerebellar agenesis 555
Histological structure of the cerebellar cortex and the functions
of the two distinctly different afferent systems to the
cerebellum: the climbing and mossy fibers 555
Physiological and anatomical studies in the 20th century and
current questions 557
Some unsolved questions 558
Acknowledgments 558
References 558
The cerebellum has always been recognized as a distinct
subdivision of the brain. The cerebellum is not visible when
looking down at the surface of the human brain, since it is
overlaid by the occipital lobe of the cerebral cortex. In
many animals, the cortex does not extend over the cere-
bellum, so it can be seen from the top. This species differ-
ence lay at the root of Vesalius’ (1514 –1564) questioning
an earlier description by Galen. Vesalius (1543) wrote:
“Anatomists describe the site of the cerebellum as if it
filled the whole region of that prominence of the occiput,
that swelling which the mass of the people consider a
measure of the power of memory and talent. The very
highest part of the cerebellum extends only to the middle
[the middle here refers to the occiput; ed.] although some,
deluded by oxen and asses or by dreams, have written that
the cerebellum ascends from the posterior site of the fora-
men, to the lambdoid suture . . . .”
Vesalius was one of the first in a long tradition in
cerebellar research of questioning the worth of your pre-
decessor’s contribution. Vesalius suggested that Galen’s
descriptions were in error, probably because he based his
descriptions, not on the human cerebellum, but on that of
an ox.
The earliest anatomical descriptions were limited by
the usual procedure of dissecting brain structures from
above. Costanzo Varolio (1543–1575) introduced the prac-
tice of dissecting from below (Varolio, 1573). He described
and illustrated the pons, a massive structure in humans
which is intimately related to the cerebellum, and often
referred to as “the Pons of Varolius.”
The description of the external features was in place.
There followed an increasingly accurate portrayal of the
550 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559

Fig. 1. Figure of a split brain and cerebellum from Vieussens (1684). The floor of the fourth ventricle is visible, the initial portion of the spinal cord is
indicated by a T. The cerebellar halves are further dissected to show the corpus rhomboideum, indicated as an area with double hatching, located in
the white matter between N and P.

gross anatomical structure of the cerebellar cortex and dissected the brain from all approaches, and he gave a far
recognition of the cerebellar nuclei which are encased more accurate description of the cerebellar cortex and
within its white matter. nuclei than any of his predecessors.
Marcello Malpighi (1628 –1694) wrote (1665): “All the Malacarne named cerebellar subdivisions for their re-
fibres dispersed through the brain and cerebellum seem to semblance to some other structure, e.g. lingula (Italian;
have origin from the trunk of the spinal marrow contained linguetta, a cat’s tongue) uvula (Italian ugola). He wrote:
within the cranium like a noteworthy collection of fibres: for “. . . is seen a thick pyramidal eminence . . . . I have called
they ramify hither and thither from four reflected crura of the laminate pyramid . . . . to these lateral tangles . . . I give
the marrow until they end in the cortex in branching ex- the name of tonsils of the cerebellum and to the conical
tremities. This course is more apparent in the cerebellum eminence uvula, because they greatly resemble the parts
because the fibres are extended in the form of a tree and of the same names located in the human body.”
on the extreme branches, almost like leaves, the cortex is In Malacarne’s time cretinism was widespread in the
delicately placed. It is not attached to anything so it resem- Po valley due to a lack of iodine in this region of Italy. He
bles a free leaf.” described two cases of cretins that came to autopsy in
Shortly thereafter, the cerebellar nuclei were recog- which the cerebellum was smaller than normal, and with
nized by Raymond de Vieussens (1641–1716) as an ash fewer folia. Malacarne also suggested that the cerebellum
grey area of glandular substance (Fig. 1: the corpus rhom- might be involved in plastic changes. In his book on the
boideum), buried within the white matter of the cerebellum. anatomy of the cerebellum he described the presence of
Félix Vicq dÁzyr (1746 –1794) provided a more realistic variability in the number of folia from 500 to 780. In an idiot
drawing of the corpus rhomboideum, and renamed it as the they were 340. In a series of letters that he exchanged with
corps dentelé (Moreau de la Sarthe, 1805). The other
cerebellar nuclei were first labeled as the emboliform, glo-
bose and fastigial nuclei by Stilling (1864). Vicq dÁzyr drew
attention to the distinctness of the indentation on the su-
perior surface of the dentate and their relative absence on
its inferior surface (Fig. 2), corresponding to the two divi-
sions of the human dentate now known as the dorsal
microgyric division or palaeodentatum and its ventral, mac-
rogyric division, also known as the neodentatum, the latter
receiving its name from the large undulations of the cell
band in this part of the nucleus. A division of the monkey
dentate into a dorsal (motor) and ventral (nonmotor) por-
tions, with projections to (pre-)motor and pre-frontal corti-
cal areas and eye fields recently was proposed by Dum
and Strick (2002). See Fig. 3.
In the eighteenth century there was a succession of
increasingly accurate anatomical descriptions in which
Fig. 2. Diagram of a parasagittally dissected cerebellum from Vicq
subdivisions of the cerebellum began to be named. Vin- dÁzyr’s collected works (Moreau de la Sarthe, 1805). The dorsal part
cenzo Malacarne (1744 –1816) published the first work of the dentate nucleus, with small dentations is indicated with 21 and
entirely devoted to the cerebellum (Malacarne, 1776). He 22, the macrogyric ventral dentate with number 23.
 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559
Fig. 3. Title page of Malacarne’s “Il cervelletto,” the first book devoted
entirely to the cerebellum.
the Swiss anatomist Carlo Bonnet Malacarne (1791) dis-
cussed whether such variability is innate or was acquired
by experience. He proposed raising animal twins in poor
and enriched environments and determining the number of
folia. If the experiment was ever done, it has not been
reported.
By the end of the 18th century there was an accurate
picture of the gross anatomy of the cerebellum, and there
Fig. 4. Ventral aspect of the human cerebellum. From Reil (1807-1808). k, Quadrangular lobule; l, flocculus; m, posterior medullary velum; n, nodulus;
o, uvula; p, pyramis; q, transverse band (tuber vermis), connecting the gracile lobule of both sides; r, transverse commissure (folium vermis) of the
superior lobule; s, tonsil; t, biventral lobule.
551
were speculations about its functions. Although the ana-
tomical descriptions were accurate, early claims for cere-
bellar functions remained speculative; typically based on
very little evidence. Johann Christian Reil (1759 –1813)
and Karl Friedrich Burdach (1776 –1847) used alcohol fix-
ation to verify and extend Malacarne’s earlier description,
resulting in the classical nomenclature for the lobules of
the human cerebellum, still in use today (Reil, 1807-1808;
Burdach, 1819 –1826). Reil’s paper contains a meticulous
description of the cerebellar lobules. Beautiful engravings
illustrate this paper (Fig. 4).
At the time that these anatomical studies were being
reported, Volta’s discovery of a bimetallic source of elec-
tricity was the newest thing in physics. Reil suggested that
the alternating layers of grey and white matter constitute a
sort of voltaic pile, generating animal electricity. Other
interpretations were based on even less-well-founded ev-
idence.
The phrenologists Gall (1757–1828) and his followers
(Combe and Combe, 1838) viewed the cerebellum as the
organ of sexuality: philoprogenitiveness (love of making
babies). The claim was taken seriously, some authors
supporting the proposed role of the cerebellum in sexual
functions, others rejecting it. In this lectures on diseases of
the brain, reported in The Lancet, Andral cited M. Voisin
who had studied the heads of 372 convicts sentenced to
the gallerys at Toulon. On the basis of the shape of their
skulls he identified 20 of these as potential sex offenders.
Voisin wrote that “thirteen of these twenty had actually
been confined for rape or attempts upon female chastity.”
Voisin also cited Dr. Ferroresi of Torino who had obtained
the cure of a young girl afflicted with nymphomania and
two young men who suffered from a habit of masturbation
552 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559

clude that the production and the coordination of move-

Fig. 5. Bust of Rolando presently in the Department of Neuroscience
of the University of Turin, Italy.
by the simple application of ice to the back of the head,
below the occipital protuberance. Gall’s speculations about
cerebellar function attracted supporters, even though ex-
perimental evidence began to be available.
ments form two classes of essentially distinct phenomena
and that they reside in two classes of organs also essen-
tially distinct: with coordination in the cerebellum and pro-
duction in the spinal cord and medulla oblongata.”
Flourens dismissed as unfounded the claims for the
sexual functions of the cerebellum. He removed the cere-
bellum of a mature rooster. The animal was still deeply
interested in the hens, but motor dysfunction made it diffi-
cult for him to express his feelings towards them. The
animal’s deficit seemed to be entirely motor in character.
Flourens’ descriptions of the effects of cerebellar lesions
remained as the definitive work for much of the remainder
of the 19th century, and many of his conclusions remain
valid today.
Towards the end of the century Luigi Luciani (1840 –
1919) used improved operative technique and sterile pro-
cedures to re-address the question of cerebellar symp-
toms.
Luciani (1891) (Fig. 7) distinguished between the im-
mediate transient effects of lesions (unstabilized defi-
ciency), and their more permanent effects (stabilized defi-
ciency). He emphasized that the permanent effects of
cerebellar lesions could be understood in terms of elemen-
tary deficits in muscle control. The three cardinal symp-
toms according to Luciani were asthenia, or weakness,
atonia, lack of normal muscle tone, and astasia. Luciani
characterized astasia as an inability to maintain normal
fusion and continuity of movement, reflected in the inten-
tion tremor that is seen in cerebellar patients. He wrote:
“To this group of phenomena which include tremor, tituba-

EXPERIMENTAL EVIDENCE OF CEREBELLAR

FUNCTION
At the beginning of the nineteenth century, animal exper-
iments began to give a more accurate functional under-
standing. Luigi Rolando (1773–1831) identified the specif-
ically motor symptoms which follow cerebellar lesion
(1809) (Fig. 5). The lesions impaired motor and not sen-
sory or intellectual functions.
Rolando concluded that the cerebellum was the brain
region responsible for initiating movement, but his experi-
ments were rather crude. With the development of in-
creased surgical skill and later of aseptic technique more
accurate evaluation of the effects of cerebellar lesions
became possible. Two of the great experimenters of the
19th century were Pierre Flourens and Luigi Luciani.
Flourens (1794 –1867) (Fig. 6) made the fundamental
observation that animals are not paralyzed. Movements
are not completely lost after cerebellar ablation. He argued
that it is the coordination of movement, a property which
had not previously been considered by physiologists. Flou-
rens also noted there may be considerable recovery from
the initial symptoms caused by a cerebellar lesion.
Flourens (1824) wrote:
“. . . All movements persist following ablation of the
cerebellum: all that is missing is that they are not regular
and coordinated. From this I have been induced to con- Fig. 6. Pierre Flourens; portrait.
 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559 553

tion and rhythmical oscillating movements, we gave the

name astasia for the sake of brevity and owing to their
probable common origin.”

CLINICAL INTERPRETATION OF CEREBELLAR

LESIONS AND DISEASE
These 19th-century studies profoundly influenced clinical
interpretation. Two of the great twentieth-century neurolo-
gists based their analysis on experimental findings. Joseph
Babinski (1857–1932), following Flourens, emphasized
deficits in coordination especially of antagonistic muscles
used in rapid movement sequences (dysdiadochokinesis),
while Gordon Holmes, following Luciani, emphasized more
elementary losses in muscle control.
Babinski (1902) (Fig. 8) cited “Flourens’ memorable
experiments”:
“Cerebellar symptoms are asynergia, adiadokokinesia
and cerebellar catalepsy. I am adding to it excessive mo-
tions . . . . Adiadokinesia is the suspension or decrease of
the ability to carry out rapidly successive voluntary move-
ments. It is the very loss of this function which makes us
aware of its existence.”
Gordon Holmes (1876 –1965) (Fig. 9) based his inter-
pretation on the experiments and conclusions of Luciani.
He wrote (1917):
“Atonia or diminution of that slight constant tension Fig. 8. Joseph Babinski; portrait.
which is characteristic of healthy muscle, is such a con-
contractions . . . . Clinical observations consequently con-
stant and striking feature of all early injuries . . . that it is
firm Luciani’s conclusion that atonia, asthenia, and astasia,
obviously a primary and direct result of the lesion.”
the triad of symptoms to which he attributes all the func-
“Asthenia . . . lack of normal power in movements that
tional disturbances, resulting from cerebellar lesions.”
demand its exertion . . . . It is probably pronounced only
Both Luciani and Holmes agreed that cerebellar le-
when the cerebellar nuclei are involved.”
sions produced symptoms on the same side of the body as
“The tremor that occurs in maintaining an attitude and
voluntary muscle movement is due to this defect in the
regularity and stability of the muscular contractions. Lu-
ciani has described the condition as astasia, and has
attributed it to the imperfect fusion of the single twitch

Fig. 7. Luigi Luciani; portrait. Fig. 9. Gordon Holmes; portrait.

554 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559

Many of Bolk’s descriptive terms are still used (lobulus

Fig. 10. Lodewijk Bolk; portrait.
the lesion, but neither found evidence for a more detailed
localization. Holmes wrote: “But although my observations
lend no support to the theory of focal localization of func-
tion in the cerebellar cortex they cannot be accepted as
proof that such localization does not exist.”
André Thomas was particularly fascinated by the pa-
thology and neurology of the cerebellum, a topic over-
looked by J. M. Charcot. Thomas described olivoponto-
cerebellar atrophy with Dejerine and lamellar atrophy of
the cerebellar cortex.
simplex, ansiform lobule, crus I and II and paramedian
lobule). Bolk’s use of the terms “formatio vermicularis”
divided into the “crus circumludens” and the “uncus termi-
nalis” was discarded and replaced by the terms parafloc-
culus and flocculus.
Bolk’s and Larsell’s organization scheme formed the
basis for subsequent analysis of the dimensions of the
human and animal cerebellum by Sultan and Braitenberg
(1993) (Fig. 11). They measured the width of the cerebellar
folia, and the depths of the fissures between them. They
used the measurements to construct a flat map of an
unrolled cerebellar cortex in mammalian species ranging in
size from mouse to human.
Bolk, in the last pages of his great monograph tried to
make sense out of the comparative anatomical differ-
ences. Assuming that subdivisions would reflect differ-
ences in the capacity for complex movements, he looked
for differences among mammals in size of one or another
cerebellar region. In this he assumed a single somatotopic
plan, with vermian cerebellum involved in the control of
proximal muscles; hemispheres of distal muscles. Bolk
reasoned that the giraffe, whose neck movements are

FUNCTIONAL LOCALIZATION AND

COMPARATIVE ANATOMY
Comparative study revealed similarities and differences
among vertebrates in cerebellar structure. Vesalius and
others had emphasized the fact that the folding and fis-
sures of the cerebellum seem to be much less variable
than in the cerebral cortex. One of the great contributors to
comparative anatomy, often poorly recognized, was made
by the Dutch anatomist Lodewijk Bolk (1866 –1930).
Bolk (1906) (Fig. 10) compared the structure of cere-
bellum in 69 different mammals. He identified a common
plan in (virtually) all of them. Bolk divided the cerebellum
into four fundamental regions: a unitary anterior lobe com-
bined with lobulus simplex, a posterior vermis and paired
cerebellar hemispheres. All four of these divisions he be-
lieved to reflect independent “growth centres.” One of
Bolk’s most important contributions was his recognition
that in all mammals, despite the complexity of its folds,
there is a continuity of the folial chains of the vermis and
Fig. 11. “Unrolled” cerebellum in a series of mammals. Outlines of the
hemispheres. Lobules are delimited by sudden changes in
shapes of the cerebellar cortices, obtained by connecting the ends of
direction or local expansions of the folial chain. Bolk gave the most prominent folia. The scale is the same in the laterolateral and
us a common plan for interpreting cerebellar structure in all anteroposterior direction. Note division of posterior cerebellum in the
mammals. folial chains of vermis and hemispheres, and the relative width and
Bolk also contributed to cerebellar nomenclature. He length of these chains in different species. 1, Mouse; 2, bat; 3, flying
fox; 4, guinea pig; 5, rabbit; 6, pigeon; 7, hare; 8, chinchilla; 9, squirrel;
used a simple numbering/letter scheme for the vermis 10, dog; 11, cat; 12, macaque; 13, sheep; 14, human; 15, bovine.
which was later replaced and expanded by Larsell (1886 – Magnifications differ for the diagrams 1– 4, 5–9 and 10 –15. Adapted
1964) who recognized 10 vermian lobules (Larsell, 1970). from Sultan and Braitenberg (1993); J Hirnforschung, with permission.
 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559 555

highly subtle and complex, should show a greater devel-

opment of the appropriate cerebellar region. Bolk formu-
lated a hypothesis to account for the representation of the
body surface on the cerebellum in which he proposed that
there is a single somatotopic map. The suggestion of a
single somatotopic map was contradicted by the work of
Adrian (1943) and Snider and Stowell (1944) who demon-
strated a second map in the hemisphere of the anterior
lobe and the adjoining lobulus simplex, and in the parame-
dian lobule.
Larsell (1970) based his scheme of the folial pattern on
the continuity of the transverse fissures from the vermis
into the hemisphere (which is not always present) and on
their development as vermian fissures that extend into the
hemisphere (which is also not always the case). Larsell,
therefore, emphasized the composition of the cerebellum
of a series of 10 transverse units, where the hemispheral
lobules were considered as the lateral expansions of the
vermal ones, whereas Bolk emphasized the longitudinal
continuity in the folial chains of vermis and hemispheres,
and the relative independence of the lobules therein.

THE RELATED PROBLEMS OF RECOVERY

FROM CEREBELLAR LESION AND
CEREBELLAR AGENESIS
The initial symptoms which follow cerebellar lesions may
improve over time. In the extreme case of cerebellar agen-
esis, it has even been stated that the cerebellum may not
be necessary for normal movement. Although some au-
thors have concluded that cerebellar agenesis is not nec-
essarily associated with motor impairment, the conclusion
is probably wrong (Dow and Moruzzi, 1958). A young girl
who died from an unrelated disease was found at autopsy
to lack a cerebellum (Anton and Zingerle, 1914). Her
mother said that she had never learned to speak clearly
although she understood language. She could not stand
until the age of three, and then only with support. Another
case of agenesis (Tennstedt, 1965) was cited in one re-
view as having normal movement, but the family said he
only learned to walk when he was 7. The source of some
of the claims that agenesis may not be associated with
motor impairment was a case from Cambridge (Glickstein,
1994). A man came to autopsy whose brain seemed to
lack the cerebellum. On the death certificate the man was
listed as having been a “building worker.” The designation
“building worker” was interpreted as suggesting that the
man had motor skills. Although nothing was known, how-
ever, about the man when he was alive. The myth of
normal movement in cases of agenesis fed upon itself.
HISTOLOGICAL STRUCTURE OF THE
CEREBELLAR CORTEX AND THE FUNCTIONS
OF THE TWO DISTINCTLY DIFFERENT
AFFERENT SYSTEMS TO THE CEREBELLUM:
THE CLIMBING AND MOSSY FIBERS
Recognition of the cell and fiber types in the cerebellum
was closely related to the development of the neuron
Fig. 12. Jan Evangelista Purkinje; portrait.
doctrine; the idea that the brain and spinal cord are made
up of individual elements (later called neurons) and their
supporting elements. Neurons may touch one another, but
they do not fuse. The first characteristic cells in the human
brain to be accurately described were by Jan Evangelista
Purkinje in 1837 (Fig. 12), 2 years before Schwann (1839)
formulated and published the cell doctrine. In a paper read
in Prague, September 1837, Purkinje reported on his re-
cent observations with a newly acquired microscope using
achromatic objectives to observe thin, unstained sections
of the brain. After discarding the concept of elementary
nerve-fibers as hollow tubes, he discussed the ganglion-
like constitution of certain parts of the brain. Among other
structures he described the large neurons in the cerebellar
cortex which bear his name as “a great number of similar
corpuscles, which surround the yellow substance [the
granular layer, ed.]. Each corpuscle faces the yellow sub-
stance with a blunt, rounded pole and contains the central
nucleus within the space of its flask-like body; the other
pole is tail-like and directed outwards; its two prolongations
get lost in the grey substance near the outer periphery,
where this is covered by the meninges. This is visible over
the entire extent of the folial gyrations of the cerebellum
that thus displays the characteristics of a ganglion.” The
figure accompanying Purkinje’s paper (Fig. 13) shows a
distinct nucleolus within the nuclei of his corpuscles. The
nucleolus was not mentioned by Purkinje, but received its
name from his student Valentin (1836).
In Purkinje’s time there were few methods available for
fixing and staining neuronal tissue. Further progress re-
quired better methods of tissue preparation. There is a
paradox. The single most important tool for histological
study of the cerebellum and for ultimately validating the
556 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559
Fig. 13. Purkinje’s 1838 text and illustration of his description of the
monolayer of large corpuscles in the cerebellar cortex.
neuron doctrine itself, was based on the contributions of
Camillo Golgi (1883) (Fig. 14) and his “reazione nera.” But
Golgi never accepted the validity of the neuron doctrine.
Golgi made one fundamental mistake. He assumed
that the functionally important connections of brain are
axonal fusion in a plexus. The plexus he thought is made
up of axon collaterals and branches of incoming axons, the
“reticular” theory of neuronal organization. Soma and den-
drites, he believed, have a solely nutritive role.
Golgi’s views were challenged by Santiago Ramon y
Cajal (Fig. 15).
Fig. 14. Camillo Golgi; Portrait.
Fig. 15. Ramon y Cajal; portrait.
Ramon y Cajal emphasized that in no case did axons
or dendrites fuse; that interaction is made by contact from
an axon onto a dendrite or soma of the next cell in the
pathway. In addition to illustrating the cell types, Ramon y
Cajal also described the two fundamental afferents to the
cerebellar cortex: the mossy and climbing fiber. The rapid
spread of Ramon y Cajal’s ideas is remarkable. Six years
after his first publications using the Golgi method which
appeared in obscure Spanish journals, he was invited to
deliver the Croonian Lecture of the Royal Society in Lon-
don (Ramon y Cajal, 1894), where he reported on the fine
structure of the retina, the olfactory bulb, the cerebral
cortex and the cerebellum (Fig. 16).
The function of the two afferent systems of the cere-
bellum, the climbing and the mossy fibers, initially received
little attention. One of exception was the work of the Dutch
psychiatrist and anatomist Gerbrandus Jelgersma (1859 –
1942; Fig. 17). He discussed his ideas on the function of
the cerebellum and of the double innervation of the Pur-
kinje cells in a series of publications and monographs
(Jelgersma, 1886, 1928, 1932). He recognized the two
circuits connecting the cerebellum with the cerebral cortex,
through cerebellothalamic and corticopontine pathways
and with the periphery, through the crossed descending
limb of the superior cerebellar peduncle and the spinocer-
ebellar tracts. From his observations in cases of olivo-
ponto-cerebellar atrophy, in which the myelinated fibers of
the granular layer had disappeared, and of spinocerebellar
atrophy, where they had been preserved, he concluded
that ponto- and spino-cerebellar fibers terminate as mossy
and climbing fibers, respectively. The pontocerebellar
mossy fiber–parallel fibers pathway informs the Purkinje
cells on intended movements, the spino-cerebellar climb-
ing fibers on their execution. A mismatch between intention
 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559 557

to George Gray. The inferior olive was demonstrated to be

Fig. 16. The elements of the cerebellar cortex. From Ramon y Cajal’s
Croonian Lecture (Ramon y Cajal, 1894). Abbreviations: A, Purkinje
cell; B, basket cell. The more superficially located neuron is a stellate
cell; C, climbing fiber; D, Purkinje cell axon with collaterals; E, gran-
ule cell; F, parallel fiber; G, mossy fiber rosette; H, basket of the basket
cell axon.
and execution causes the Purkinje cells to send a correc-
tive stimulus to the cerebral cortex (high-level coordina-
tion) or to the spinal cord (low level coordination). Jelg-
ersma (1928) postulated the formation of new, temporary
connections in the learning process involved in cerebellar
coordination. Although Jelgersma’s anatomy was wrong,
his ideas on the function of the Purkinje cells and the role
of neuronal plasticity in cerebellar coordination predate
modern, similar concepts on long-term adaptation of
movement by the Purkinje cells of the cerebellum (Marr,
1969; Ito, 1982).
a major source of climbing fibers using silver impregnation
method to detect degenerating fibers (Szentágothai and
Rajkovitz, 1959; Desclin, 1974).
In the same years Masao Ito discovered the inhibitory
nature of the Purkinje cells, whose axons constitute the
only output from the cerebellar cortex. Ito’s discovery was
against the prevailing dogma that neurons with long axons
are excitatory and that inhibition is solely effected by inter-
neurons, local elements that switch excitation into inhibi-
tion (Eccles et al., 1967). Shortly thereafter Olov Oscars-
son (1979), using electrophysiology of spino-olivocerebel-
lar climbing fiber paths, Jan Voogd (1967) studying the
topography of white matter compartments in the cerebel-
lum, and David Armstrong (Armstrong et al., 1974) using
direct electrical stimulation of the inferior olive, demon-
strated that the cerebellar Purkinje cells are organized in a
series of longitudinal parasagittal bands. Each band re-
ceives its climbing fiber input from a well-defined cluster of
inferior olive cells, and projects to a specific locus in the
cerebellar nuclei. Olivary cells send collaterals to a defined
region of the cerebellar nuclei. The axon collaterals of the
climbing fibers, and the target of the Purkinje cell in the
cerebellar nuclei are in precise register.
In subsequent studies Voogd’s nomenclature for these
zones was generally adopted. A chemoarchitectonic zona-
tion of the cerebellar cortex was demonstrated by Scott
(1963) and Hawkes and Leclerc (1987). Hawkes’ zebrin
antibodies to a Purkinje cell–specific epitope was shown to
be present in a pattern of alternating bands of zebrin-
positive and -negative Purkinje cells. More recently the
Voogd’s olivocerellar climbing fiber zones B, C and D were
shown to be congruent with particular zebrin-positive or
-negative bands (Voogd et al., 2003). Voogd’s A zone,

PHYSIOLOGICAL AND ANATOMICAL STUDIES

IN THE 20TH CENTURY AND CURRENT
QUESTIONS
Ramon y Cajal’s great contribution could not be properly
utilized in his own day by physiologists. The development
of microelectrode recording and electron microscopy ush-
ered in a new age in cerebellar physiology in the 1960s.
Janos Szentágothai with morphological studies and John
Eccles by means of electrophysiological analysis provided
for the first time a complete picture of the functional archi-
tecture of the cerebellar cortex, identifying the excitatory
and the inhibitory nature of each cell type (Eccles et al.,
1967). Seminal contributions at ultrastructural level of the
synapses are due to Sanford Palay and at glomerular level Fig. 17. Gerbrandus Jelgersma, portrait.
558 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559
however, consists of a complicated array of zebrin-positive
and -negative strips, innervated by different parts of the
caudal medial accessory olive (Sugihara and Shinoda,
2004).
Studies on the developmental biology of the cerebellar
cortex, and the factors determining its zonation and the
patterns in the termination of its mossy and climbing fiber
afferents of the last decades were reviewed by Sotelo
(2004). That the zonal pattern is an intrinsic and funda-
mental property of the cerebellum was demonstrated in the
experiments of Zagrebelsky et al. (1996). They showed in
the mature cerebellum that regenerating climbing fibers
distribute according to this pattern.
These new discoveries served as a basis for future
development of cerebellar physiology. In a seminal theo-
retical paper, following a suggestion by Brindley, Marr
(1969) proposed that the cerebellar cortex acts as a learn-
ing device. He suggested that the pattern of activity of the
granule cells (1011 in human), with their parallel fibers,
could be stored by a long-lasting increase of the synaptic
efficacy if there was a simultaneous activation of the climb-
ing fiber. Marr provided a model of how the two excitatory
inputs which end on the Purkinje cell as mossy and climb-
ing fibers might interact. There were some undesirable
mathematical properties of the Marr model which led Albus
(1971) to propose a modification. Rather than an increase
in synaptic efficiency the climbing fiber would produce a
long lasting depression of the parallel fiber to Purkinje cell
synapses.
The issue of the involvement of such a depression in
motor learning has been initially supported by many studies
by using the eye blink conditioning (reviewed by Thompson
and Steinmetz, in this volume; Yeo et al., 1985). There is also
evidence that the olivo-cerebellar system functions as a bio-
logical clock whose fundamental purpose is to initiate pace,
and thereby coordinate movements (reviewed by Rodolfo
Llinás in this volume). It has also been demonstrated that the
inferior olive activity has a powerful inhibitory action on Pur-
kinje cells and has the role in shaping and maintaining the
architecture of the cerebellar cortex (reviewed by Cesa and
Strata in this volume).
For the past 40 years there have been many studies of
a possible role of the cerebellum in several forms of learn-
ing. Also the involvement in emotions and in cognitive
functions attracts a growing interest. These topics are still
controversial, and differing points of view are expressed by
several of the other authors in this volume.
SOME UNSOLVED QUESTIONS
1. How should we interpret the spatial extent of cerebellar
cortex? Are there many somatotopic maps? Current
textbooks, based on earlier studies of mossy fiber af-
ferents postulate two somatotopic maps; but the situ-
ation is far more complex.
2. How do mossy fiber and climbing fiber afferents inter-
act in their effects on the Purkinje cell?
3. A question almost forgotten: what is the meaning in the
great differences among mammals in development of
one or another region of cerebellum?
4. Finally what does the cerebellum do? How does it
calibrate reflex and saccadic movement? How does it
combine exteroreceptive, and proprioceptive informa-
tion in the control of movement? Does it initiate move-
ment, or is it essentially a sensory structure, predicting
the sensory consequences of a movement? Does it
also play a role in motor learning, language, cognition,
or emotion?
The cerebellum is the brain structure most critically
involved in reflex adjustments and the acquisition of fine
and sequential motor control. There are many other func-
tions that have been suggested for it, and several possible
mechanisms. We believe that the historical approach can
help focus on current questions. Problems have been
solved in the past, and they will continue to be solved in the
future.
Acknowledgments—We were guided to much of the early litera-
ture by the excellent volume by Clarke and O’Malley (1968). Some
of the quotes used in this article were taken from the translations
in that book.
REFERENCES
Adrian ED (1943) Afferent areas in the cerebellum connected with the
limbs. Brain 66:289 –315.
Albus JS (1971) A theory of cerebellar function. Math Biosci 10:25– 61.
Anton G, Zingerle H (1914) Genaue Beschreibung eines falles von
beiderseitigem Kleinhirnmangel. Arch Psychiatr 54:8 –75.
Armstrong DM, Harvey RJ, Schild RF (1974) Topographical localiza-
tion in the olivo-cerebellar projection: an electrophysiological study
in the cat. J Comp Neurol 154:287–302.
Babinski J (1902) Sur le rôle du cervelet dans les actes volitionnels
necessitant une succession rapide de mouvements (diadoco-
cinésie). Rev Neurol 10:1013–1015.
Bolk L (1906) Das Cerebellum der Säugetiere: eine verglichende
anatomische Untersuchung. Haarlem: Fischer.
Burdach KF (1819 –1829) Vom Baue und Leben des Gehirns. Leipzig:
Dyk.
Cesa R, Strata P (2009) Axonal competition in the synaptic wiring of
the cerebellar cortex during development and in the mature cere-
bellum. Neuroscience 162:624 – 632.
Clarke E, O’Malley CD (1968) The human brain and spinal cord.
Berkeley: University of California Press.
Combe G, Combe A (1838) On the functions of the cerebellum by Drs
Gall, Vimont and Broussais. Translated from the French by Combe
G. Edinburgh: MacLachlan and Stewart.
Desclin JC (1974) Histological evidence supporting the inferior olive as
the major source of cerebellar climbing fibers in the rat. Brain Res
77:365–384.
Dow RS, Moruzzi G (1958) The physiology and pathology of the
cerebellum. Minneapolis: University of Minnesota Press.
Dum RP, Strick PL (2002) Motor and nonmotor domains in the monkey
dentate. Ann N Y Acad Sci 978:289 –301.
Eccles J, Ito M, Szentágothai J (1967) The cerebellum as a neuronal
machine. New York: Springer Verlag.
Flourens P (1824) Rechérches expérimentales sur les propriétés et les
fonctions du système nerveux dans les animaux vertébrés. Paris:
Crevot.
Glickstein M (1994) Cerebellar agenesis. Brain 117:1209 –1212.
Golgi C (1883) recherches sur l’histologie des centres nerveux. Arch
Ital Biol 3:285–317 (1884) 4:92–123.
Hawkes R, Leclerc N (1987) Antigenic map of the cerebellar cortex:
the distribution of parasagittal bands as revealed by monoclonal
anti-Purkinje cell antibody map Q113. J Comp Neurol 256:29 – 41.
 M. Glickstein et al. / Neuroscience 162 (2009) 549 –559
Holmes G (1917) The symptoms of acute cerebellar injuries due to
gunshot injuries. Brain 4:461–535.
Ito M (1982) Cerebellar control of the vestibulo-ocular reflex—around
the flocculus hypothesis. Annu Rev Neurosci 5:275–296.
Jelgersma GJ (1886) De anatomische verhoudinegn van de den pons
varolui en de nucleus olicvaris tot de groote en de kleinehersenen
(the anatomical relations of the pons varpoli and the inferior olive to
the forebrain and the cerebellum). Psychiatry 4:156 –180.
Jelgersma JG (1928) Over schakelingen. Bijdrage tot de kennsi van
het cerebro-cverebellaire coordinatie systeem (On circuits. A con-
tribution to the knowledge of the cerebro-cerebellar coordination
system). Ned Tijdschr Geneesk 72:6272– 6280.
Jelgersma GJ (1932) Das Kleinhirn in anatomischer, physiologischer
und pathologischer. Hinsicht J Psychol Neurol 44:505– 612.
Larsell O (1970) The comparative anatomy and histology of the cere-
bellum from monotremes through primates. Minneapolis: Univer-
sity of Minnesota Press.
Luciani L (1891) Il cervelletto: nuovi studi di fisiologia normale e
pathological. Firenze : Le Monnier.
Malacarne MVG (1776) Nuova esposizione della vera struttura del
cervelletto umano. Torino: Briolo.
Malacarne MVG (1791) Sulla neuro-encefalotomia. Lettere ana-
tomico-fisiologiche di Vincenzo Malacarne e Carlo Bonnet, s.i.t.
Pavia.
Malpighi M (1665) De cerebro epistola. Bologna: Benati.
Marr D (1969) A theory of cerebellar function. J Physiol 202:437– 470.
Moreau de la Sarthe JI (1805) Oeuvres de Vicq-d’Azyr: recueillies et
publieìes avec des notes et un discours sur sa vie et ses ouvrages
par Jacq L Moreau (de la Sarthe) L [Band 6 and atlas]. Paris:
Duprat-Duverger.
Oscarsson O (1979) Functional units of the cerebellum; sagittal zones
and microzones. Trends Neurosci 2:143–145.
Purkinje J (1838) Die XV Versammlung deutscher Naturforscher und
Aerzte zu Prag im September 1837. Sitzungen der Fachabtheilun-
gen. V. Anatomie Ueber die scheinbar kanalikolöse Beschaffenheit
der elementaren Nervencylinder. pp 582–583. In: Isis von oken.
Ramon y Cajal S (1894) The Croonian Lecture. La fine structure des
centres nerveux. Proc R Soc Lond 55:444 – 467.
Reil JC (1807-1808) Fragmente über die Bildung des kleinen Gehirns
im Menschen. Arch Physiol 8:1–58.
Rolando L (1809) Saggio sopra le vera struttura del cervello dell’uomo
e degli animali e sopra le funzioni del sistema nervoso. Sassari:
Stamperia da S.S.R.M.
(Accepted 24 February 2009)
(Available online 9 March 2009)
559
Schwann T (1839) Mikroskopische Untersuchungen über die Übere-
instimmung in der Struktur und dem Wachstum der Thiere und
Pflanzen. Berlin: Reimer.
Scott TG (1963) A unique pattern of localization within the cerebellum.
Nature 200:793.
Snider RS, Stowell A (1944) Receiving areas of the tactile, auditory
and visual systems in the cerebellum. J Neurophysiol 7:331–357.
Sotelo C (2004) Cellular and genetic regulation of the development of
the cerebellar system. Prog Neurobiol 72:295–339.
Stilling B (1864) Untersuchungen über den Bau des kleinen Gehirns
des Menschen. Cassel: T. Fischer.
Sugihara I, Shinoda Y (2004) Molecular, topographic, and functional
organization of the cerebellar cortex: a study with combined aldo-
lase C and olivocerebellar labeling. J Neurosci 24:8771– 8785.
Sultan F, Braitenberg (1993) Shapes and sizes of different mammalian
cerebella. A study in quantitative comparative neuroanatomy. J
Hirnforsch 34:79 –92.
Szentágothai J, Rajkovits K (1959) Über den Ursprung der Kletter-
fasern des Kleinhirns. Z Anat Entwickl Gesch 121:130 –141.
Tennstedt H (1965) Kleinhirnaplasie beim erwachsenen. Zblatt Allg
Pathol 107:301–304.
Thompson RF, Steinmetz JE (2009) The role of the cerebellum in
classical conditioning of discrete behavioral responses. Neuro-
science 162:732–755.
Valentin GG (1836) Ueber die Verlauf und den letzten Enden der
Nerven. Nova Acta Acad Leopoldina 18:51–240, 531–543.
Varolio C (1573) De nervis opticis nonnulisque aliis prater communem
opinionem in humano capite observatis epistole. Padova: Metti.
Vesalius A (1543) De humani corporis fabrica libris septem. Basel:
Oporini.
Vieussens T (1684) Neurographia universalis. Lyons: Certe.
Voogd J (1967) Comparative aspects of the structure and fibre con-
nexions of the mammalian cerebellum. Prog Brain Res 25:94 –134.
Voogd J, Pardoe J, Ruigrok TJ, Apps R (2003) The distribution of
climbing and mossy fiber collateral branches from the copula pyra-
midis and the paramedian lobule: congruence of climbing fiber
cortical zones and the pattern of zebrin banding within the rat
cerebellum. J Neurosci 23:4645– 4656.
Yeo C, Hardiman M, Glickstein M (1985) Classical conditioning of the
nictitating membrane response of the rabbit. II. Lesions of the
cerebellar cortex. Exp Brain Res 60(1):99 –113.
Zagrebelsky M, Rossi F, Hawkes R, Strata P (1996) Topographically
organized climbing fibre sprouting in the adult rat cerebellum. Eur
J Neurosci 8:1051–1054.