Variants of Congenital Ocular Motor
Apraxia: Associations With Hydrocephalus,
Pontocerebellar Tumor, and a Deficit
of Vertical Saccades
I r e n e A n t e b y , M D , -~B e n j a m i n Lee, M D , b M i c h a e l N o e t z e l , M D , -~and L a w r e n c e T y c h s e n , M D -~

Background:Congenital ocular motor apraxia (COMA)is characterized by the inability to generate volitional horizontal
saccadic eye movements in the absence of other focal neurologic abnormalities. Subjects:We report on two children
(ages 5 months and 3years)whose COMA did not adhere to these classic criteria.The children were followed up clinically
with serial ocular motor examinations and neuroimaging over a period of 3 years. Results: In the first child horizontal
COMAwas associated initiallywith neonatal communicating hydrocephalus. Two and one half years after the first signs
of COMA, a fourth ventricle medulloblastoma appeared. The second child, who recovered from a periventricular
hemorrhage caused by perinatal asphyxia, manifested vertical COMA and compensatory vertical head thrusts.
Conclusions:COMA may be associated with hydrocephalus, pontocerebellar tumor, and periventricular hemorrhage.
These rare variants of COMA emphasizethatthe eye movement deficits may arise from several locations, cerebral as well
as pontocerebellar, in the neuronal pathways generating saccades.(J AAPOS 1997;1:201-8)

ongenital ocular motor apraxia (COMA) is
characterized by the inability to generate voluntary
saccadic eye movements. Reflexly induced saccades
and quick phases of vestibular nystagmus are relatively
preserved. 14 In classic cases of COMA the saccadic deficit
is horizontal, whereas vertical saccades are normal. The
disorder is typically recognized at the age of 5 to 6 months
when the infant acquires head and neck control, manifested by head thrusts to substitute for the inability to
make saccades.1 The etiology and pathogenesis of COMA
is currently unknown. Brain stem tumors, 5-7 cerebellar
abnormalities,8, 9 and hypoplasia or agenesis of the corpus
callosumS, 10, ll have been reported rarely in association
with COMA.
We describe two children with COMA, one of whom
had an atypical clinical course and the other atypical

c

From the Depaf~tment of Ophthalmology and Visual Sciences,~ the Mallinckrodt Institute of Radiology, b and the Department of Neurology, c St. Louis Children's Hospital,
Washington University School of Medicine, St. Louis, Missouri.
Supported by National Institutes of Health program project grant No. O1 NS 17763-I 2
(L. T.), National lnstitutes of Health grant No. R O1 EYI O214-O1A 3 (L. T.), a McDonnell
Center for Higher Brain Function Research Grant (I.A.), Knights Templar Eye Foundation Research Grant (I.d.), and in part by National Institutes of Health grant No.
EY02687 and an unrestricted grant to the Department of Ophthalmology and Visual
Sciencesfrom Research to prevent Blindness, Inc.
Reprint requests: L. Tychsen, MD, St. Louis Children's Hospital, Washington University School of Medicine, One Children's Place, Room 2 South 89, St. Louis, MO 63110.
Copyright © 1997 by the American Assodation for Pediatric Ophthalmology and Strabismzar.
1091-8531/97 $5.00 + 0 75/1/84884
Journal

of AAPOS

clinical signs. In the first case horizontal COMA was
associated with neonatal communicating hydrocephalus
followed years later by the appearance of a fourth
ventricle neoplasm. In the second case a child who had
a cerebral periventricular hemorrhage from birth
asphyxia manifested a rare form of persistent vertical
COMA.
R E P O R T O F CASES
Case 1
A 5-month-old male infant was referred to our unit for
suspected poor vision. His mother noted that he was
unable to follow moving targets but did smile when she
placed herself directly in his line of sight. The infant, a
twin, was delivered at a gestational age of 37 weeks by
cesarean section, which was prompted by preeclampsia.
The child's birth weight was 3990 gm. Routine neonatal
examination documented a head circumference above the
95th percentile. A head computed tomographic (CT) scan
was performed (Figure 1), which revealed communicating
hydrocephalus. The ventricular system was slightly dilated
but had otherwise normal configuration. There was no
evidence of a mass lesion. Because of progressively
increasing head circumference (46.5 cm), a
ventriculoperitoneal shunt was placed by the pediatric
neurosurgery service when the patient was 3 months old.
Repeat CT scans showed a functioning shunt and normal
head growth.
December

1997

201

202

Journal ofAAPOS
Volume 1 Number 4 December 1997

Anteby et al.

FIG. 1. CT scan of child No. 1 at age of 2.5 months showing mildly
dilated frontal (short arrow) and occipital horns (longer arrow) of
lateral ventricles. Child had unusually large head circumference. CT
supported diagnosis of communicating hydrocephalus.

FIG. 2. Sagittal Tl-weighted MRI after infusion of gadolinium showing
large, intensely enhancing tumor within fourth ventricle compressing
pons and ventral part of cerebellum (bottom arrow). Cerebral
aqueduct (top arrow) is dilated slightly.

Examination in our unit 2 months after the shunt
procedure revealed that the infant fixated well with each
eye but could not initiate voluntary saccadic eye movements to fixate on a novel target presented at an eccentric
location along the horizontal plane. Vertical saccades
appeared clinically to be of normal latency, amplitude, and
velocity. The fast phases of small-field optokinetic
nystagmus (OKN), induced by a hand-held drum, were
infrequent in the horizontal plane but normal vertically.
When the vestibuloocular reflex was evoked by wholebody rotation, fast phases were deficient so that the eyes
tended to remain deviated at the extreme edges of the
orbits (e.g., to the right edge of each orbit as the child's
head rotated to the left). Versions evoked by head rotation
were normal. The infant was orthotropic and made
normal convergence and divergence eye movements. The
visual field was fully intact, as evidenced by the fact that the
infant made head thrusts to acquire targets in all visual
field quadrants. Pupillomotor and funduscopic examinations were normal. A diagnosis of COMA was made.
A follow-up examination at age 2 years documented
improvement in the initiation of horizontal voluntary
saccades. Horizontal small-field optokinetic nystagmus
was normal with no nasotemporal asymmetry under
conditions of monocular viewing. 12 Visual field testing
with use of the evoked-saccade method, 13 pupillomotor
examination, and funduscopic examinations again had
normal results. General motor development was considered by the parents to be slightly delayed compared with
that of the normal twin brother. Language skills were
appropriate. Follow-up examinations in the pediatric

neurosurgery unit had revealed no evidence of shunt
malfunction. Follow-up magnetic resonance imaging
(MRI), performed at age 2 years, showed no intracranial
abnormality.
At the age of 3 years (i.e., 2.5 years after the initial
diagnosis of COMA), the patient was admitted to the
pediatric neurology service for sudden onset of ataxic gait
causing frequent falls. MRI of the brain showed a 4.3 × 3.8
cm posterior fossa mass occupying and enlarging the
fourth ventricle (Figure 2). The mass displaced the pons
anteriorly and compressed the ventral cerebellum. A
smaller second mass lesion was present in the cerebellar
tonsils. A subtotal resection of the tumor was performed
through an infratentorial posterior craniotomy incision.
Histopathologic examination documented a medulloblastoma. The child received chemotherapy with cisplatin and
etoposide followed by radiation therapy to treat cerebellar
and lumbar spine metastasis. Follow-up examination 8
months after the craniotomy showed no additional eye
movement deficits as a consequence of the tumor
resection. The ataxia had resolved, but the ocular motor
apraxia remained unchanged.
Case 2
A 3-year-old boy was referred to our unit for evaluation of
abnormal head movements present since infancy. He was
born in Tiibingen, Germany, by a cesarean section
prompted by preeclampsia at 36 weeks' gestation. The
pregnancy was complicated by fetal growth retardation,
diagnosed by intrauterine ultrasonography in the third
trimester. The birth weight was 2060 gm. In the neonatal

Journal of AAPOS
Volume I Number 4 December 1997

Anteby et al.

FIG. 3. Typical vertical head thrust observed in child No. 2. 1, After introduction of target in superior
visual field, child's strategy is to thrust head (and orbits) upward, activating vertical slow-phase
vestibuloocular reflex, which allows visual axes to be vertically realigned on target. 2, Examiner begins
to move target downward abruptly. 3, Child initiates downward head thrust, activating upward slow
phase of vestibuloocular reflex, deviating eyes upward in orbits. 4, Child hyperflexes his head so that
head deliberately overshoots target. 5, Child raises his head slightly, now using downward slow-phase
of vertical vestibuloocular reflex to achieve alignment of his visual axes on target.

203

204

Journal ofAAPOS
Volume 1 Number 4 December 1997

Anteby et aL

TABLE 1. Reports of horizontal COMA in presence of intracranial tumors

Authors

Patient
sex, age

Interval
between COMA
onset and
diagnosis of
tumor

Zaret et al. (1980)7

Male, 8 months

4 months

Lyle (1961)6

Male, 16 months

Summers et al. (1987)~
Case No. 1

Type of lesion

Improvement
of COMA
after tumor
excision

Cerebellar vermis,
cerebral peduncle, thalamus

Small cell neoplasm

Yes

11 months

Ports, lateral wa]l fourth
ventricle, cerebellar hemisphere

Medulloblastoma

Yes

Female,10 months

7 months

Cerebellar hemisphere
(venous sinus)and cerebellar vermis

Lipoma

No

Male, 5 months

2.5 yr

Fourth ventricle, compressing
pons and cerebellum

Medul]oblastoma

No

Location of
lesion

period he was noted to be initially hypotonic, followed by
hypertonicity. A small left hemispheric subependymal
hemorrhage (attributed to perinatal asphyxia) was documented on an MR/scan obtained at age 1 month. Followup neurologic examination performed at age 2.5 years
showed mild diplegic spastic cerebral palsy affecting gross
motor functions and language skills.
On examination in our unit the mother reported that
from early infancy the patient had displayed consistently unusual vertical head movements. He was unable
to generate voluntary vertical saccades. T o refixate a
target presented in the superior or inferior visual field,
the child used vertical head thrusting (Figure 3) and the

vertical vestibuloocular reflex. His strategy was to blink
and flex his head forward or backward rapidly so that his
head overshot the target, evoking an upward (or
downward) slow-phase vestibuloocular reflex movement. T h e slow-phase deviation allowed him to achieve
fixation, after which he returned his orbital axis to align
with the object. Fast phases of small-field and full-field
vertical O K N were rare. In the horizontal plane the
patient fixated and followed normally with prompt,
normometric saccades and fast phases of OKN. Cover
testing documented that he was orthotropic. Pupillomotor and funduscopic examinations were normal.
Visual acuity was normal (2 0/36 in the right eye and 2 0/

TABLE 2. Clinical features and birth history of children with vertical COMA

Authors

Patient
age, sex

Birth, weight,
gestational age

Perinatal
asphyxia

Neuroimaging
lesions (age)

Neurologic
development

Hughes et ah (1985)~8

Male, 4.5 yr

1135 gm, 30 wk

Yes

CT: normal (1 wk and 4.5 yr)

De]ayed motor development,
deficit language skills

Ro et al. (198 9)~8

Female, lyr

3200 gm, 41.5 wk

Yes

CT and MRI: normal (2 wk)

Hypotony in infancy, delayed
motor development

Ebner et al. (1990)20

Male, 2.5 yr

3150 gm, not reported

Yes

CT: low-density areas on
both sides of third ventricle
(2.5 yr) MRI: bilateral lesions
at mesencephalic-diencephalic
junction (4.5 yr)

Delayed motor development

Buncic and Crawford
(1985)2~

Female,
7 months

"Normal," term

Yes

CT and MRI: bilateral, medial
thalamic lesions (7 months)

Severe memory deficit

Case No. 2

Male, 3 yr

2060 gm, 36 wk

Yes

MRI: hemispheric subependymal
Mild spastic cerebral palsy
hemorrhage (1 month), normal (3 yr) deficient gross motor function
and language skills

Journal ofAAPOS
Volume 1 Number 4 December 1997

Anteby et al.

20Y

Cortex generated
volitional saccade

paqueduct
',
MR>

HORIZONTAL
PATHWAYS
postulated
level of lesion
case # 1

C
I [¢,2"~.~,~]

LR

tal
Semicircular Canal
"Head thrust" (VOR)

FIG. 4. Horizontal saccadic eye movement pathways in coronal view. Postulated lesion in case 1 is between
level of oculomotor and abducens nuclei, involving either decussating corticopontine fibers or saccaderelated neurons of pontine raphe. Generation of horizontal saccades involves activation of both posterior
parietal cortex (PPC) and frontal eye field (FEF)of one cerebral hemisphere. Signals descend from both
these, through the internal capsule and basal ganglia (bold downward arrows), to ipsilateral midbrain
superior colliculus (SC), which sends projection to neurons of pararnedian pontine reticular formation
(PPRF) on opposite side of pons. Frontal eye field also sends parallel projection to paramedian pontine
reticular formation that does not involve synapse in superior colliculus. Activation of left paramedian
pontine reticular formation excites motoneurons of ipsilateral abducens nucleus (6) and interneurons,
which projectthrough medial longitudinal fasciculus (MLF)to contralateral medial rectus motoneurons in
the oculomotor nucleus (3). When corticopontine pathways are impaired, as in horizontal COMA, infant
learns to head thrust. Head thrusting activates neurons within semicircular canal and hence vestibuloocular
reflex (VOR),generating slow-phase rnovements that compensate partially for saccadic deficit. MR, Medial
rectus; LR, lateral rectus. Dotted line from cerebral aqueduct indicates midline. 8, Vestibular nucleus.

33 in the left eye), as documented with use of the
spatial-sweep visual-evoked potential method.
Vertical COMA was diagnosed. A repeat MRI of the
brain, obtained at age 3 years, revealed no sign of
structural abnormality. No white matter or cortical lesion
was detectable in the region of the previous neonatal
subependymal hemorrhage.
DISCUSSION
Cogan 1 introduced COMA into the clinical lexicon in
1952 at the Jackson Memorial Lecture. The key features,
described in four patients, included the absence of willed
eye movements in the horizontal plane only and telltale,
jerky overshooting of the head on fixation. None of

Cogan's patients in this report had an associated neurologic abnormality. He presumed the syndrome was
congenital; its pathogenesis was obscure.
C O M A and hydrocephalus
Hydrocephalus, especially of the noncommunicating type,
may cause a variety of neuro-ophthalmologic manifestations in children. The motility abnormalities associated
with increased intracranial pressure include the setting sun
sign; the dorsal midbrain syndrome; sixth, fourth, or third
nerve palsies; comitant esotropia; skew deviation; and
internuclear ophthalmoplegia. 14 Our case No. 1 had
communicating hydrocephalus resulting from extraventricular obstruction of cerebrospinal fluid flow in associa-

206

Journal ofAAPOS
Volume 1 Number 4 December 1997

Anteby et al.
Cortex gene
volitional sac
postulated
- level of lesion
case # 2

CERTICAL
ATHWAYS

SR
IR;
10

~ R

al
)

FIG. 5. Vertical saccadic eye movement pathways in coronal view. Postulated lesion in case 2 is in
subcortical white matter neurons or contiguous basal ganglia. Generation of vertical saccades
involves activation of posterior parietal cortex (PPC) and frontal eye field (FEF) in both cerebral
hemispheres. For sake of clarity, descending projections are shown fully for only right frontal eye
field and posterior parietal cortex. Both send descending projections through internal capsule and
basal ganglia to midbrain and pons. Right and left rostral interstitial nucleus of medial longitudinal
fasciculus (riMLF) receives projections from frontal eye field, from superior colliculus (SC) and
paramedian pontine reticular formation (PPRF),and from contralateral rostral interstitial nucleus of
medial longitudinal fasciculus through interconnecting fibers of posterior commissure (PC).Rostral
interstitial medial longitudinal fasciculus provides signals to vertical eye muscle motoneurons in
oculomotor 3and trochlear 4nuclei. Vertical head thrusting excites neurons within the posterior (Post)
and anterior (Ant)semicircular canals, and hence the vestibuloocular reflex (VOlt),generating vertical
slow-phase vestibuloocular reflex movements that compensate partially for saccadic deficit. SR,
Superior rectus; IR, inferior rectus; I0, inferior oblique; SO,superior oblique. Oottedlinefrom cerebral
aqueduct indicates midline. 8, Vestibular nucleus.

don with COMA. The COMA in this infant did not
resolve after the shunting procedure and subsequent
control of intracranial pressure. PeBenito and CraccC s
described a case of C O M A associated with
noncommunicating hydrocephalus. The child they reported had congenital aqueductal stenosis, and a
ventriculoperitoneal shunt was placed at age 1 month
because of enlarged ventricles and multiple arachnoid
cysts. The child was referred to them to rule out blindness,
a common misdiagnosis in infants with COMA who are
younger than age 6 months and who have poor head
control)6, 17They established the diagnosis of CONLA at
age 8 months, after the onset of typical head thrusts.
Follow-up beyond infancy in their case also documented
motor and speech delay.
COMA and intracranial t u m o r
Among the six dozens of cases of COMA reported in the
literature, four (listed in Table 1) have been associated with
intracranial tumors (for a prevalence of-5 %)?-7 The tumors
in each of these four cases were located in the midbrain, pons,

or cerebellum, implying a brainstem locus for the COMA. It
is important to note, however, that the brainstem is the most
common site of pediatric brain tumors. In our case No. 2
several years lapsed between the onset of COMA and the
diagnosis of tumor. In the previous reports (see Table 1,
column 3) the diagnosis of brain tumor was established 4 to
11 months after the appearance of COMA. The lag period in
the cases reported by Lyle 6 in 1961 and Zaret et al.7in 1980
could be attributed in part to lack of sensitive neuroimaging
techniques. In our case state-of-the-art CT and MRI scans,
performed serially from age 1 month, failed to reveal, for a
span of 2.5 years, any posterior fossa or brainstem lesion.
Vertical C O M A
It is important to differentiate vertical COMA from
acquired ocular motor apraxias. In the acquired type a
horizontal saccadic deficit is accompanied by a vertical
saccadic deficit.3 In our case horizontal saccades were
normal. The only manifestation of an ocular motility
disorder was the inability to initiate vertical saccades and
OKN fast phases.

Journal ofAAPOS
Volume 1 Number 4 December1997
Only four cases of isolated vertical COMA have been
reported. 18-21Our case No. 2 had perinatal asphyxia and
a hemorrhage in the region of the internal capsule noted
on MRI scanning at age 1 month. Although perinatal
insults are uncommonly associated with horizontal
COMA, asphyxia at birth (Table 2, column 4) has been
a feature of each of the previously reported cases of
vertical COMA. I8-21 Neonatal asphyxia is known to
cause damage to the germinal matrix in the region of the
internal capsule and basal ganglia, structures through
which pass corticopontine fibers, which are important
in saccadic control.
Pathophysiologic features of C O M A
The pathophysiologic features of COMA remain
obscure. Our two cases, along with those listed in
Tables 1 and 2, lead to the conclusion that horizontal
and vertical COMA are due to lesions at different levels
of the saccadic pathway. As shown in Figure 4, the
generation of horizontal saccades in primates appears to
involve activation of both the posterior parietal cortex and
the frontal eye field of the cerebral hemispheres.22-25
Signals are conveyed from the posterior parietal cortex
and the frontal eye field in one cerebral hemisphere,
through the internal capsule and basal ganglia, to the
ipsilateral superior colliculus, which sends a projection
to neurons of the paramedian pontine reticular formation on the opposite side of the pons. W h e n the
corticopontine pathway is impaired, as in horizontal
COMA, the infant learns to head thrust, activating
neurons within the semicircular canal, and hence the
vestibuloocular reflex, to partially compensate for the
saccadic deficit?, 4 In each of the reported cases of
horizontal COMA with tumor (Table 1), the tumor has
involved the dorsal pons or cerebellum (depicted as the
"level of lesion" in Figure 4).
The generation of vertical saccades (Figure 5) appears
to involve activation of the frontal eye field and posterior
parietal cortex in both cerebral hemispheres.=, 23 The
frontal eye fields and posterior parietal cortexes send
descending projections through the internal capsule and
basal ganglia to the brainstem. The right and left rostral
interstitial medial longitudinal fasciculus (riMLF) receive
projections from the frontal eye field, the superior
colliculus, and the paramedian pontine reticular formation
and are interconnected through fibers of the posterior
commissure. The riMLF provides the signals to vertical
eye muscle motoneurons in the nuclei of cranial nerves III
and IV. Each of the reported cases of vertical COMA has
been associated with perinatal hypoxic-ischemic encephalopathy (Table 2). Perinatal hypoxic encephalopathy is
known to damage especially subcortical neurons in the
region of the internal capsule and basal ganglia, 26,27which
is depicted as the "level of lesion" in Figure 5.
Our cases and those previously reported in the
literature support the notion that dysgenesis of saccade-

Anteby et al.

207

related neurons, at brainstem or cerebral levels, gives rise
to horizontal, or rarely, vertical COMA. The improbable
concurrence of horizontal COMA and brainstem tumor,
and the latency between onset of horizontal COMA and
subsequent appearance of tumor, implies that in a
minority of cases an occult neoplastic process, present
from early infancy, contributes to the neuronal dysfunction responsible for COMA.
References
1. Cogan DG. A type of congenital ocular motor apraxia presenting
jerky head movements. Trans Am Acad Ophthalmol Otolaryngol
1952;56:853-62.
2. Cogan DG, Chu FC, Reingold D, Tychsen L. A long-term followup of congenital ocular motor apraxia. Neuro-Ophthalmol
1980;1:145-7.
3. Leigh JR, Zee DS. The neurology of eye movements. 2nd edition.
Salem (MA): FA Davis; 1991.
4. Zee DS, Yee RD, Singer HS. Congenital ocular motor apraxia.
Brain 1977;100:581-99.
5. Summers GC, MacDonald JT, Wirtschafter JD. Ocular motor
apraxia associated with intracranial lipoma. J Pediatr Ophthalmol
Strabismus 1987;24.'267-9.
6. Lyle DJ. Discussion of ocular motor apraxia with a case
presentation. Trans Am Ophthalmol Soc 1961;59:274-85.
7. Zaret CR, Behrens MM, Eggers HM. Congenital ocular motor
apraxia and brainstem tumor. Arch Ophthalmol 1980;98:328-30.
8. Fielder All, Gresty MA, Dodd KL, Meller DH, Lefene MI.
Congenital ocular motor apraxia. Trans Ophthalmol Soc U K
1986;105:589-98.
9. Eda I, Takashim S, Kitahara T, Ohno D, Takeshita K. Computed
tomography in congenital ocular motor apraxia. Neuroradiology
1984;26:359-62.
10. Orrison W, Rohertson J. Congenital ocular motor apraxia: a
possible disconnection syndrome. Arch Neurol 1979;36:29-31.
11. Borchert MS, Sadun AA, Sommers JD, Wright KW. Congenital
ocular motor apraxia in twins: findings with magnetic resonance
imaging. J Clin Neuro-Ophthalmol 1987;7:104-7.
12. Tychsen L. Binocular vision. In: Hart WM, editor. Adler's
physiology of the eye: clinical applications. 9th edition. St. Louis:
Mosby-Year Book; 1992. p. 773-853.
13. Tychsen L. Pediatric ocaflar motility disorders of neuro-ophthalmic
significance. In: Shults WT, editor. Ophthalmology clinics of North
America. Volume 4. Philadelphia: WB Saunders; 1991. p. 615-43.
14. Brodsky MC, Baker RS, Hamed LM. Pediatric neuroophthalmology. New York: Springer-Verlag; 1996.
15. PeBenito R, Cracco JB. Congenital ocular motor apraxia. Clin
Pediatr 1988;27:27-31.
16. Rosenberg ML, Wilson E. Congenital ocular motor apraxia
without head thrusts. J Clin Neuro-Ophthalmol 1987;7:26-8.
17. Gittinger JW, Sokol S. The visual-evoked potential in the
diagnosis of congenital ocular motor apraxia. Am J Ophthalmol
1982;93:700-3.
18. Hughes JL, O'Connor PS, Larsen PD, Mumma JV. Congenital
vertical ocular motor apraxia. J Clin Neuro-Ophthalmol

1985;5:153-7.
19. Ro A, Gummeson B, Orton RB, Cadera W. Vertical congenital
ocular motor apraxia. CanJ Ophthalmol 1989;24:283-5.
20. Ebner R, Lopez L, Ochoa S, Crovetto L. Vertical ocular motor
apraxia. Neurology 1990;40:712-3.
21. Buncic JR, Crawford JS. Congenital vertical apraxia [abstract].
Proceedings of the annual meeting of the American Association for
Pediatric Ophthalmology and Strabismus; 1997 Mar 3-7; San Juan,
Puerto Rico.

208

Journal ofAAPOS
Volume I Number 4 December1997

Anteby et al.

22. Sparks DL. The neural translation of sensory signals into
commands for the control of saccadic eye movements: the role of
the primate superior colliculus. Physiol Rev 1985;66:118-71.
23. Tusa RJ. Saccadic eye movements: supranuclear control. Bull Soc
Beige Ophtalmo11989;237:67-111.
24. Sparks DL. Signal transformations required for the generation of
saccadic eye movements. Annu Rev Neurosci 1990;13:309-36.
25. Arai IG Keller EL, Edelman JA. Two-dimensional neural network

O

model of the primate saccadic system. Neural Networks
1994;7:1115-35.
26. VolpeJJ. Brain injury in the premature infant--current concepts. In
Proceedings of the Current Concepts of Cerebral Palsy Following
Preterm Birth Conference; 1993 Nov 18-20; St. Louis.
27. Volpe JJ. Hypoxic-ischemic encephalopathy: neuropathology and
pathogenesis. In: Volpe JJ, editor. Neurology of the newborn.
Volume 22. Philadelphia: WB Saunders; 1987. p. 209-35.

N THE MOVE?
~en

us your newa

r e s s a t l ec a s t s i x w e e k s a h e a d

D o n ' t m i s s a s i n g l e i s s u e o f t h e j o u r n a l ! To e n s u r e p r o m p t s e r v i c e w h e n y o u c h a n g e y o u r a d d r e s s ,
please photocopy and complete the form below.

Please send your change of address notification at least six weeks before your move to ensure continued service.
We regret we cannot guarantee replacement of issues missed due to late notification.
J O U R N A L TITLE:
Fill in the title of the journal here.
OLD ADDRESS:
Affix the address label from a recent issue of the journal here.

NEW ADDRESS:
Clearly print your new address here.
Name
Address
City / State / ZIP

COPY A N D MAIL T H I S F O R M TO:
J o u r n a l S u b s c r i p t i o n Services
M o s b y - Y e a r Book, Inc.
11830 Westline I n d u s t r i a l Dr.
St. Louis, M O 63146-3318

OR FAX TO:
314-432-1158

Mosby

OR P H O N E :
1-800-453-4351
OutsidetheU.S.,call
314-453-4351